I Introduction
Driven by the skyrocketing growth of mobile devices and the wide applications of the Internet of things (IoT), future wireless communication systems have triggered the explosive demand and urgent need for ultrahigh capacity, ultralow latency, and massive connectivity over the scarce wireless resources [1, 2, 3, 4]. Specifically, an everincreasing volume of mobile data traffic is expected to appear in the coming years, which may overwhelm the limited spectrum resources significantly and increase the power consumption dramatically [2, 3, 4]. Due to the unprecedented surge of mobile data traffic, researchers have been motivated to develop new transmission technologies for maximizing the achievable throughput and minimizing the deployment cost. Among various technologies, spatial modulation (SM) has been envisioned as one prospective digital modulation technology to achieve high spectral efficiency and energy efficiency yet enjoy a simple design principle. Although early attempts, which can be dated back to the beginning of the 21st century, have been made to explore the preliminary SM [5, 6], they were sporadic and did not receive much attention in the early days. After 2008, research work on SM has begun to grow explosively and the systematic introduction of the SM concept has been reported in some tutorial/overview literature [7, 8, 9, 10], which attract a lot of attention from researchers. By activating one transmit antenna to convey index information, SM reaps the spatial gain with a single radio frequency (RF) chain and enjoys the following additional benefits [7, 6]:

lower detection complexity with a smaller number of receive antennas and lower complexity of RF circuits;

free of interchannel interference;

no need for interantenna synchronization;

compatibility with massive multipleinput multipleoutput (MIMO) [13].
In SM, extra information bits are typically conveyed through the index of one active transmit antenna, in addition to the information bits conveyed by the conventional constellation symbol. Based on the antennaswitching mechanism, the active antenna index changes according to the spatial information bits. In particular, only one RF chain is required at the SM transmitter to activate one out of multiple transmit antennas for a constellation symbol transmission, significantly saving the energy consumption in downlink communications, and dramatically reducing the hardware cost at the user terminal in uplink communications. Moreover, due to freedom of the interchannel interference, SM can be a better candidate technology than Vertical Bell Laboratories Layered Space Time (VBLAST) for highmobility wireless communication systems, where the channel correlation is weakened while the interchannel interference effect is aggravated [14]. Space shift keying (SSK) can be viewed as a simplified variant of SM [15, 16, 17, 18, 19, 20, 21], which embeds the overall information into the index of one active antenna only without involving any constellation symbol. Other variants of SM technology typically consist of generalized (G)SM [22, 23, 24, 25], quadrature SM (QSM) [26], differential (D)SM [27, 28, 29], receive (R)SM [30, 31, 32, 33], and generalized (G)SSK [34]. Generally speaking, SM refers to a new modulation family of communication systems that conveys additional information typically through the activation states of transmit antennas. By choosing different activation patterns at the transmitter, various SM members provide a flexible design to meet different specific requirements and tradeoffs among the spectral efficiency, energy efficiency, deployment cost, and system performance. For information recovery, the receivers of SM have to execute two main tasks [35]:

detecting the indices/states of active antennas;

demodulating the constellation symbols embedded on the active antennas/states (if applicable).
However, it is not trivial to effectively detect both spatial and constellation information effectively while maintaining low complexity for SM members under different channel conditions [25, 36]. On the other hand, linkadaptive SM, which relies on the feedback from the receiver to alter its transmission pattern (e.g., modulation order, transmit power and antenna selection), was extensively investigated to achieve better system performance and channel utilization [37, 38, 39, 40]. Please note that the detection performance of plain SM highly relies on the distinctness of the channel signatures/fingerprints associated with different transmit antennas. As a result, plain SM enjoys the technical challenge to operate in rich scattering propagation and stationary [41] environments. With further leverage of advantage preprocessing techniques, such as orthogonal pulse shaping [42] and trellis coded modulation (TCM) [43, 44], the problem of lack of channel distinctness can be well addressed in SM systems. Furthermore, as plain SM utilizes the space domain to convey index information via one active antenna, no transmitdiversity gain is provided by plain SM to combat channel fading effects. To overcome the lack of transmitdiversity inherent in plain SM, transmitdiversity enhanced SM using spacetime block coding (STBC) is also a promising direction to improve the error performance [45, 46, 47, 48, 49].
Due to its many promising advantages, SM serves as an attractive energyefficient modulation technology with flexibility of working together with others in emerging communication systems. Thanks to the sparsity inherent in SM signals, compressedsensing (CS) theory is a powerful tool for lowcomplexity signal reconstruction even when the number of available measurements is much smaller than the signal dimension (i.e., the number of receive antennas is much less than that of transmit antennas in underdetermined systems), particularly in the case of largescale MIMO [50, 51, 52, 53]. On the other hand, due to the significantly reduced wavelength in the millimeterwave (mmWave) frequency band, mmWave systems can be equipped with a large number of antennas at the transceiver in a highly compact manner for the implementation of largescale MIMO [54]. As a result, the SM family emerges as a promising lowcost and highefficiency candidate for largescale MIMO with a large number of transmit antennas while a much smaller number of RF chains are required for antenna activation. More recently, noticing the great potentials of nonorthogonal multiple access (NOMA) in supporting massive connectivity and low transmission latency, NOMA aided SM emerges as an attractive and novel technology for multiuser communications, which achieves high spectral efficiency and energy efficiency while maintaining lowcomplexity transceiver design [55, 56, 57]. As a result, NOMA aided SM technology strikes an appealing tradeoff among spectral efficiency, energy efficiency, deployment cost, and interference mitigation [58]. Simultaneous wireless information and power transfer (SWIPT) is another emerging important technology, which aims at delivering wireless information and energy concurrently. SM finds its special fit to SWIPTenabled wireless systems since it has the potential to leverage the inactive antennas for energy harvesting without incurring any loss of the spectral efficiency.
Moreover, due to the broadcast nature of wireless communications, the security of SM transmission is an essential problem in practice [59, 60, 61]. Interestingly, both physical layer security (PLS) and SM highly rely on the randomness and discrimination properties of the wireless interface, which can be exploited to achieve confidential information exchange among legitimate nodes while impairing the potential eavesdropper at the same time. Specifically, SM employs a fast antennaswitching mechanism to achieve the random selection of transmit antennas according to the spatial information bits and the legitimate channel state information (CSI), which can incur fast timevarying environments to confuse the eavesdropper and degrade its decoding performance. In other words, SM shows a great potential not only in its increased energy efficiency, but also in the feasibility of secure transmission. Furthermore, since visible light communications (VLC) enjoys high security and supports the fast switching in lightemitting diodes (LEDs), SM technology can also be readily integrated into VLC, which provides the flexibility to control the luminance of the LED array and the communication throughput [62].
Although originated in the space domain, the concept of SM is not exclusive to the space domain, but can be generalized and applied to other signal domains such as frequency/time/code/angle domain or even across multiple domains. As a consequence, considerable interest and attention have been paid in recent years to fully develop various forms of the SM concept in diverse wireless communication applications [35, 63, 64]. The use of the ON/OFF keying mechanism to embed index information has been mostly applied to entities in single domains such as spatial domain (e.g., antenna, RF mirror, and LED), frequency domain (e.g., subcarrier), time domain (e.g., time slot), code domain (e.g., spreading code and modulation type), and angle domain (e.g., angle of arrival (AoA) and polarization state). To further enhance the system performance and enjoy a more flexible design, multidimension entities are also developed, which include more than one dimension for performing the ON/OFF keying mechanism.
The objective of this survey is to present a comprehensive overview of the latest results and progresses in SM research. In particular, Section II reveals the basic principles and variants of SM. In addition, Section III presents several performance enhanced techniques for SM systems, including linkadaptive SM, TCM aided SM and diversity enhanced SM. The combinations of SM with other promising techniques and applications of SM to various emerging and practical communication systems are discussed in Sections IV and V. Finally, some important extensions of SM to other domains and entities are studied in Section VI, and concluding remarks with future directions are drawn in Section VII.
Ii Basic Principle and Variants of SM
In this section, we first introduce the basic principle of SM and then discuss some variants of SM, including (G)SM and (D)SM. We consider the implementations of SM in a MIMO system with transmit and receive antennas.
Iia SingleRF SM
At the beginning of this century, SM has emerged as a novel MIMO technology that works with a single RF chain and exploits the active antenna index to convey additional information based on the antennaswitching mechanism [6, 7, 65]. As a result, the information of SM is not only explicitly transmitted as one phaseshift keying (PSK)/quadrature amplitude modulation (QAM) symbol, but also implicitly transmitted by selecting the index of one active antenna for each channel use. Given the number of transmit antennas and the modulation order of the signal constellation , the spectral efficiency of SM is
(1) 
where [bpcu] stands for bits per channel use. Specifically, the first part of bits determines the index of the active antenna and the second part of bits is used to modulate the constellation symbol . Consequently, as the constellation symbol is carried by the
th antenna, the transmit vector of SM can be expressed as
(2) 
where all elements of except the th one are zeros. As shown in Fig. 1, using the antennaswitching mechanism at the transmitter, the index of active antenna changes randomly for each channel use according to the incoming information bits. Specially, when , the SM system degenerates into the SSK system, in which the overall information bits are mapped into the index of active antenna in the space domain and the corresponding spectral efficiency is
(3) 
At the receiver side, both the index of active antenna and the corresponding constellation symbols must be detected for information recovery. Various detection algorithms have been developed for SM systems, which can be divided into two main categories: joint detection and decoupled/separate detection. For example, the maximumlikelihood (ML) detector is the most typical representative of joint detection, which searches jointly over all possible transmit antennas and constellation symbols to achieve optimal detection performance [66]. However, the ML detector has relatively high complexity which linearly scales with the number of transmit antennas and the size of the constellation. On the other hand, various suboptimal lowercomplexity detectors were also developed, which typically involve a twostep decoupled process, i.e., first determine the index of the active transmit antenna and then demodulate the constellation symbol carried on the active transmit antenna [67, 68, 69, 70, 71, 72]. As a result, the size of the search space is reduced from to . Although the twostep decoupled/separate detectors have smaller search complexity, they usually suffer from performance degradation compared with the optimal detector. Therefore, it is an essential and challenging problem to design lowcomplexity detection algorithms to achieve optimal or nearoptimal performance for the SM receiver. It is noted that by exploiting the sparsity of the SM signal, the implementations of lowcomplexity detection algorithms based on CS theory are developed to achieve nearoptimal performance [50, 73].
On the other hand, the practical shaping filter is employed to evaluate the impact of antenna switching for bandlimited SM and a multipleRF antenna switching mechanism is developed to overcome the shortage of singleRF SM scheme in [74]. Moreover, the authors of [75] evaluate the impact of pulse shaping on the bandwidthversusenergyefficiency tradeoff for bandlimited SM and emphasize the advantage of singleRF SM in largescale antenna array implementation. Moreover, the design of SM detection algorithms taking into account of channel correlation [76, 77]
, imperfect channel estimation
[78, 79, 80, 81], performance analysis [82, 83, 84, 85], cooperative protocol [86, 87, 88], and coded systems [69, 89, 90, 91] are also extensively investigated in the literature.It is also worth pointing out that the concept of SM provides a multipleantenna fullduplex node additional freedom to decide which antennas for signal transmission and which antennas for reception simultaneously, increasing the spectral efficiency while remaining the RF chains advantages. The authors of [92] for the first time proposed this idea, which was later extended to fullduplex relaying, such as in [93]. Note that in the above introduction the number of transmit antennas has been assumed to be a power of . When this assumption cannot be satisfied, effective schemes such as fractional bit encoding [91]
, bit padding
[94], and constellation order varying [95] can be applied to improve the spectral efficiency of the SM system. The investment for the benefit is the increased susceptibility to error propagation.IiB Generalized (G)SM
Although the singleRF SM enjoys high energy efficiency with one active antenna, one major disadvantage is that spectral efficiency suffers from a slow logarithmic growth with an increasing number of transmit antennas. Hence, the constraint of a single active antenna is relaxed in (G)SM to provide more spatial information by allowing more than one antennas to be simultaneously activated to transmit the same PSK/QAM symbol [22, 23]. Since the same PSK/QAM symbol is transmitted from all active antennas, this (G)SM scheme requires only one RF chain and is also free of interchannel interference and there is no need for interantenna synchronization. For each channel use, out of () transmit antennas are selected to carry the same constellation symbol while the remaining antennas are inactivated, resulting in the following spectral efficiency^{1}^{1}1 stands for the binomial coefficient and is the floor function.
(4) 
Although the spatial information significantly increases in (4) compared to (1), the constellation information is limited by one PSK/QAM symbol. To further improve the spectral efficiency, the concept of (G)SM can be further extended by allowing different active antennas to carry different PSK/QAM symbols [24, 96], resulting in the spectral efficiency
(5) 
In particular, (G)SM enables a flexible design to achieve a compromise between the spectral efficiency, the deployment cost (the number of RF chains), and the error performance by choosing the required number of active antennas [97]. It is worth pointing out that the singleRF SM and full activation MIMO become two special cases of (G)SM with and , respectively.
The transmitter diagram of the (G)SM system is depicted in Fig. 2, where the incoming information bits are separated into two parts for different purposes. Specifically, unlike the classical MIMO scheme that maps all incoming bits to constellation points for all transmit antennas, a block of incoming bits is divided into the space domain and constellation domain, respectively. The first part with bits is applied to the index selector for activating antennas (i.e., determining the indices of those active antennas)
(6) 
where is the index of the th active antenna with and stands for the active antenna combination (AAC), whose bit mapping procedure can be implemented using the lookup table when the number of AACs is small or combination strategy when the number of AACs is large [98, 99, 100].
It is worth noting that among possible combinations, only AACs are permitted while the other AACs are illegal due to the integer constraint of mapping bits. Without loss of generality, the set of legal AACs is denoted as . The second part with bits is then applied to the symbol modulator for generating constellation symbols transmitted by the active antennas, which generates the transmitting block as
(7) 
where is drawn from an order constellation and transmitted by the th active antenna. Denote , where and .
After transmission through the MIMO channel, the signal model at the receiver side is expressed as
(8) 
where strands for the transmit power,
denotes an additive white Gaussian noise (AWGN) vector with a zero mean and normalized variance,
is a constellation symbol vector carried by the active antennas, and consists of the columns of corresponding to active transmit antennas. The receiver of (G)SM has to detect the indices of the active antennas and demodulate the corresponding constellation symbols simultaneously for information recovery.By performing a joint detection over the active antenna indices and constellation symbols, the ML detector for (G)SM can be expressed as
(9) 
We can observe from (9) that the size of the search space is , resulting in a considerably high search complexity, especially for largescale MIMO or high order constellations. Therefore, many researchers have focused on the design of lowcomplexity detection algorithms for the (G)SM system, e.g., the CS based detections [50, 51, 52, 53], the ordered block minimum meansquarederror (OBMMSE) based detections [101, 102], the message passing based detection [100], the Gaussian approximation based detections [103, 104], and sphere decoding (SD) based detections [105, 106, 107]. However, the abovementioned detection algorithms either suffer from suboptimal performance with considerable loss or encounter a fluctuating complexity that depends on the channel conditions, making them unsuitable for practical implementations. Moreover, most existing detectors output harddecisions, which suffer from some performance loss when channel codes are deployed in the coded (G)SM system. In [108], a soft OBMMSE based detector is proposed to output softdecisions for the coded (G)SM system, which sorts the possible AAC based on the specific metric and then proceeds to estimate the constellation symbols associated with the MMSE solution for each AAC. To achieve nearoptimal performance but with low complexity for coded (G)SM systems, soft detection algorithms based on deterministic sequential Monte Carlo (DSMC) and best SD are proposed in [25, 36], respectively. Fig. 3 shows the performance results of some representative detectors in terms of bit error rate (BER) for the uncoded (G)SM system. It can be observed from Fig. 3 that by showing the ML detector as the benchmark, the detectors proposed by [25] and [36] can achieve nearoptimal performance while other detectors still suffer from some performance loss. To examine the performance of soft decoders, Fig. 4 shows the comparison results in terms of BER for the coded (G)SM system, where the turbo receiver and rate1/2 constraint length3 convolutional code are employed. It can be observed that the DSMC based detector in [25] achieves nearly the same performance as the MAP detector (acts as the performance benchmark for the turbo receiver) under different iteration numbers, which outperforms the soft OBMMSE detector in [108] significantly.
IiC Differential (D)SM
Since the SM transmitter embeds the spatial information in the active antenna implicitly, the SM receiver has to detect such spatial information by distinguishing different channel fading states associated with different transmit antennas, which requires the CSI for coherent detection. However, the CSI requirement at the receiver increases the deployment cost due to the pilot overhead and channel estimation complexity. Alternatively, differential encoding of the SM symbols, which dispenses with any CSI at the transceiver while inheriting the advantages of SM, emerges as an attractive solution with low deployment cost. A differentially encoded spacetime shift keying (STSK) modulation scheme is presented in [109] as the primitive idea of (D)SM, which employs the Cayley unitary transform and conveys information via the activation state of the spacetime dispersion matrix. To overcome limitation on the realvalued constellation of [109], the authors of [68] further develop a differentially STSK scheme with QAM to achieve higher bandwidthefficiency.
On the other hand, a permutationbased (D)SM scheme is developed in [27], which conveys information via the antenna activation order and is applicable to an arbitrary number of transmit antennas. Specifically, as the original SM of antennas has activation states, the (D)SM will transmit an spacetime block, which is one permutation of block activation states. As a result, we have different permutations for the spacetime block and the transmitter of (D)SM selects one out of permutations according to the previous spacetime block and the incoming spatial information bits, which is illustrated in Fig. 5. For each (D)SM block, the first part with bits is applied to the permutation selector to determine a permutation , while the second part with bits is then applied to the symbol modulator for generating constellation symbols in the form of the diagonal matrix , resulting in the information block
(10) 
where denotes the diagonal operation. Finally, the (D)SM block matrix is calculated by
(11) 
where the initial block matrix of (11
) can be assumed to be the identity matrix, i.e.,
. As it can be seen, the spacetime block is a fullrank matrix and has only one nonzero entry in each row or column. By leveraging the timecoherent property for noncoherent detection, the (D)SM can successfully avoid the requirement of pilot insertion at the transmitter and channel estimation at the receiver to achieve low deployment cost. Fig. 6 shows the BER comparison results between noncoherent detection (D)SM and coherent detection SM with , targeting at the spectral efficiency of bps/Hz. It can be observed that compared with coherent detection SM, the performance loss in terms of BER for the noncoherent detection (D)SM is less than dB. Fig. 7 shows the BER performance results of (D)SM under different parameter settings, targeting at the spectral efficiency of bps/Hz. By showing the singleantenna differential PSK (DPSK) as a benchmark, it can be observed that the performance gain of (D)SM increases as the number of receive antennas grows.A reduced complexity detector is investigated for (D)SM in [29] to achieve nearoptimal performance, which is derived from the ML principle and adopts a twostep operation. The introduction of amplitude phase shift keying (APSK) modulation to (D)SM is investigated in [110] and [111] to enhance the transmission efficiency and performance. In [112]
, the performance of (D)SM with two transmit antennas over Rayleigh fading channels is analyzed in terms of the average bit error probability, which shows a less than 3 dB performance loss compared with the coherent SM. To fill the performance gap between the DSM and the coherent SM, the stateoftheart in rectangular DSM concept is extensively investigated in
[113, 114, 115, 116, 117].On the other hand, DSM also has various applications. For example, the concept of (D)SM is then implemented in dualhop networks [118], which reaps the transmitdiversity and reduces the receive complexity by using precoding. To achieve the improved BER performance of (D)SM, the concept of Gray coding is applied, where the TrotterJohnson ranking and unranking theory is adopted to perform index permutations [119]. To capture the full transmitdiversity of (D)SM, schemes that use cyclic signal constellation and algebraic field extensions are developed in [120] and [121], respectively. Inspired by the (G)SM concept, the authors of [122] generalize the differential STSK scheme by allowing a subset of spacetime dispersion matrices to carry more than one realvalued constellations. Furthermore, the differential transmission scheme for (G)SM with multiple active transmit antennas is described in [123], which strikes a more flexible tradeoff between the diversity gain and transmission rate.
IiD Receive (R)SM
As a reciprocal scheme of SM technology, RSM, a.k.a. precoding aided SM (PSM), has also received a lot of research interest due to its simplified receive structure. Specifically, by applying the concept of SM at the receiverside, RSM utilizes the indices of receive antennas to convey spatial information in addition to conventional constellation information of PSK/QAM symbol. With the aid of transmitter precoding, RSM benefits from both high beamforming gain and low complexity design at the receiver side, which is highly desired for the downlink MIMO transmission.
With the perfect CSI available at the transmitter, the initial RSM is explored in [30], where both zeroforcing (ZF) and minimum meansquared error (MMSE) precoding schemes are considered. Two precoding schemes for RSM under imperfect CSI at the transmitterside are then developed in [31]. Moreover, the concept of RSM is also generalized by activating more than one antennas at the receiverside [32, 33], in which error performance and the lowcomplexity detectors are investigated. In [124], a nonlinear RSM scheme is developed with the leverage of vector perturbation, which conveys implicit information via the activation pattern at the receiverside. The authors of [125] develop a new RSM scheme to achieve both transmit and receive diversity, where several associated detection algorithms are introduced to meet different requirements in terms of complexity and reliability. To reduce the significant channel estimation overheads required at the transmitter, a twostage RSM based on partial CSI is proposed in [126] for correlated channels. Moreover, to overcome the performance degradation due to the channel correlation, an orthogonality structure is designed for generalized RSM [127]. Based on the ML criterion, both coherent and incoherent detection schemes are studied in [128], which are then simplified to the singletap detection problem.
More recently, the performance of RSM in various application scenarios are also studied and discussed. An upper bound on the error performance of the MMSEbased RSM system is derived in [129]. In [130, 131, 132], the power allocation problem for RSM is formulated, where new approximate solutions are proposed to achieve a higher performance gain. The error performance, diversity order, and coding gain of RSM under the shadowing MIMO channel are analyzed in [133] and [134]. In [135], the integration of multistream RSM under a broadcast channel is studied for the multiuser scenario, where the error performance, diversity order, and coding gain are derived. The authors of [136] develop an amalgamated scheme of QSM and RSM and derive a closedform upper bound on its error performance. Moreover, the designs of RSM for secure communication are also considered in [137, 138, 139]. In [140], the integration of RSM into the simultaneous wireless information and power transfer system is studied, where the rateenergy tradeoff problem is revealed with performance optimization.
Iii Performance Enhancement for SM
As Section II, the system performance of SM and its variants highly depends on the distinctness of the channel signatures/fingerprints associated with different transmit antennas. To avoid significant performance degradation, those SM members require rich scattering in the propagation environment, which, however, may not be realistic in practical systems. Fortunately, with preprocessing techniques, such as linkadaptive, precoding/TCM, and spacetimecoded transmission, the error performance of the SM family can be improved significantly in adverse environments.
Iiia LinkAdaptive SM
Under the slow fading channel model, linkadaptive SM can be applied to achieve better BER performance and higher channel utilization by adapting the parameter settings (e.g., modulation order, transmit power and number of transmit antennas) or precoding matrices according to the channel condition. Specifically, after channel estimation, the receiver determines the optimal transmission parameters in terms of modulation order, power allocation profile and antenna selection mode, or the optimal precoders, which is then sent back to the transmitter using a limited feedback link [37, 38, 39, 40].
In [141], adaptive SM mechanisms, which adapt the modulation type and/or transmit power through the limited reliable feedback channel, are introduced to cognitive radio systems to enhance both the spectral and energy efficiency of the secondary system. An adaptive SM scheme based on the Huffman coding technique is proposed in [142], which activates transmit antennas with distinct probabilities via Huffman mapping with a variablelength code. In addition, [143] exploits spatial correlation as the partial CSI at the transmitter to realize the adaptive SM principle, which is shown to achieve performance improvement. An adaptive unified linear precoding scheme based on the maximum criterion of the minimum Euclidean distance for (G)SM systems is proposed in [144], which maintains the RF chains and is shown to improve the system error performance relative to the (G)SM systems without precoding. The adaptive SM principle is also introduced in mmWave communication systems, where the shortest Euclidean distance of the SM symbols is maximized in an iterative manner to improve error performance [145].
IiiB Precoding/TCM Aided SM
It is worth pointing out again that the detection performance of the spatial information highly depends on the distinctness of channel impulse responses associated with different transmit antennas. As a result, the detection performance suffers from spatiallycorrelated fading channels as the channel impulse responses become very similar from different transmit antennas.
The precoding and TCM are two effective techniques to overcome the negative impact of spatial correlation on SM. In [146], a diagonal precoding matrix is applied to the SM transmit vector before transmission, which is designed to minimize the asymptotic average BER while preserving the average power budget and without any explicit knowledge of the channel coefficients at the transmitter. On the other hand, in [43] and [44] the encoding scheme based on TCM is investigated for SM systems. Specifically, the spatial information bits are convolutionally encoded and interleaved according to the TCM principle in [43], which differentiates the transmit antennas into multiple subsets to maximize the spatial distance between each transmit antenna within the same subset. However, although the scheme proposed in [43] benefits from some performance improvements in correlated fading channels, such improvements vanish in uncorrelated fading channels, compared with the conventional SM without TCM. To break the bottleneck of [43], another TCM based SM scheme is proposed in [44], which reaps the advantages of trellis coding and achieves performance improvement over the conventional SM in both uncorrelated and correlated fading channels.
IiiC TransmitDiversity Enhanced SM
It has been reported in [7] that no transmitdiversity gain can be achieved by SM transmission even though the transmitter is equipped with multiple antennas. Hence, many researchers tried to develop transmitdiversity techniques with encoding mechanisms to overcome the lack of transmitdiversity inherent in SM.
In [45], the Alamouti code is exploited as the STBC to integrate with SM, which reaps the secondorder transmitdiversity and enjoys the system simplicity by activating a pair of transmit antennas. Moreover, owing to the orthogonality of the Alamouti STBC on a spatialconstellation diagram, an optimal detector is further developed for the STBCSM system with linear complexity. By resorting to the nonvanishing determinant property, the orthogonal STBC is embedded into the spatial constellation to achieve the transmitdiversity order of two [46]. However, although the orthogonal STBC based SM scheme in [46] achieves higher spectral efficiency without the rotation phase optimization involved in [45], it is limited by its antenna configuration, which requires that the number of transmit antennas should be even and not less than four. To overcome the spectral efficiency degradation in the STBCSM system of [45], an enhanced STBCSM making use of the cyclic structure and signal constellation rotation is developed in [47], which improves the spectral efficiency without scarifying the transmitdiversity. By optimizing the power and phase allocation, a SM scheme based on modified multistrata STBC is also proposed to further improve the spectral efficiency while maintaining the same transmitdiversity order [48]. In [147], temporal permutations are introduced as a new dimension to improve the performance of spatially modulated STBCs and ensure a full transmit diversity order.
To enjoy the high transmitdiversity order while avoiding channel estimation, the STBC is also integrated into the (D)SM, which reaps the diversity order without the requirement of CSI at the receiver [148]. Inspired by the QSM proposed in [26], the inphase and quadrature parts of the constellation symbols are linearly multiplexed by a pair sets of dispersion matrices in [149] to obtain the diversityachieving QSM transmission mechanism, which achieves the secondorder transmitdiversity and improves the spectral efficiency by using additional spatial information embedded in a dispersion matrix associated with the quadrature part.
Iv (G)SM Integration with Other Promising Technologies
Iva CompressedSensing (CS) Theory for SM
For (G)SM systems, conventional detection methods such as ML and linear detections can be applied. However, the ML detector has prohibitively high complexity that exponentially grows with the number of active antennas whereas linear detectors, e.g., MMSE and ZF suffer from significant performance loss. Recently, by exploiting the sparse features in (G)SM symbols, several researchers focused on the design of lowcomplexity CSbased detection schemes to achieve satisfactory performance. Therefore, CS theory becomes a powerful tool that makes the signal reconstruction possible even when the number of available measurements is much smaller than the signal dimension (i.e., the number of receive antennas is much less than that of transmit antennas in underdetermined systems), particularly in the case of largescale MIMO [150]. Since a very large number of transmit antennas are used in largescale MIMO while only a few of them are active during each transmission, the (G)SM symbols can be considered sparse since most elements of the transmitted block are zeros.
By formulating the antenna detection of SM as an norm convex optimization problem, [50] applies the normalized CS to detect the active antenna indices, which saves the computational complexity significantly in the largescale MIMO implementation. Against the limitation that basis pursuit (BP) algorithm is not suitable for detecting sparse signals corrupted by noise, [51] makes some adaptations to counter the effect of noise and formulates the detection of (G)SM signals as a convex quadratic programming problem. A sparse Bayesian learning based detection algorithm is then proposed in [104] to reconstruct the (G)SM symbol effectively with low complexity, which shows outstanding performance over other CSbased schemes in the highly underdetermined scenarios (i.e., the receiver only has a small number of antennas). By formulating the (G)SM detection as an norm optimization problem under the constraint of fixed sparsity, Bayesian CS is applied in [53] to solve the detection problem, which shows a better performance than the BP based detector. Several efficient CSbased detectors under the orthogonal matching pursuit framework are proposed in [52] for largescale (G)SM systems, which achieve satisfactory performance with considerably reduced complexity.
The application of CS theory to the multiuser detection problems of SM systems also received great attention from many researchers. In [73], the detection problem for SM systems over the multipleaccess channel is investigated, in which a lowcost detector based on CS theory is proposed by exploiting the structured sparsity of the problem. Exploiting the distributed and group sparsity in largescale SM uplink systems, [151] develops a structured CSbased multiuser detection algorithm with satisfactory performance and low deployment cost. The joint detection problem of user activity and (G)SM signal is investigated over the terrestrial return channel [152], in which the blocksparse CSbased method is developed for solving the joint detection problem by exploiting the structured sparsity inherent in multiuser (G)SM systems. In [153], after partitioning transmit antennas into multiple groups to perform independent SM, corresponding detectors based on thresholdaided CS and message passing are proposed to achieve nearoptimal performance with low complexity. In [154], an effective Bayesian CS detector is proposed with low deployment cost for STBC based QSM systems.
IvB Nonorthogonal Multiple Access (NOMA) Aided SM
NOMA, which exhibits great potential in supporting both massive connectivity and low transmission latency, has been envisioned as one of the most promising technologies to accommodate more active users in the future network [155]. In NOMA networks, multiple users share the same channel resource and are multiplexed in the power domain via superposition coding at the transmitter [156]. To mitigate interuser interference, multiuser detection (MUD) techniques such as successive interference cancellation (SIC) are typically employed at the receiver. On the other hand, most research works of SM focus on the pointtopoint transceiver design and current multiuser SM systems usually suffer from interuser interference [157]. Recalling that NOMA can effectively mitigate interuser interference, NOMA aided SM emerges as a competitive technology in multiuser communications to achieve high spectral efficiency and energy efficiency yet maintain low deployment cost and reduced interuser interference. In particular, NOMASM strikes an appealing tradeoff between spectral efficiency, energy efficiency, and interference mitigation. Towards this direction, researchers are motivated to promote the deployment of NOMA aided SM and aim to explore both the potential benefits of SM with lowcomplexity design and NOMA with high bandwidth efficiency.
In [152] and [158], (G)SM is introduced to the NOMA system to improve both spectral efficiency and energy efficiency in the terrestrial return channel and uplink channel, respectively, in which the joint detection problem of user activity and (G)SM signals is explored by exploiting the blocksparse model. NOMAbased SM is proposed in [55] for downlink multiuser communications, which mitigates interuser interference effectively via SIC and achieves high spectral efficiency at the cost of some performance degradation. In [56], NOMAaided precoded SM is proposed for downlink multiuser communications under overloaded transmissions, in which the demodulation of precoded SM is integrated into SIC to obtain effective detection algorithms and analytical results in terms of spectral efficiency, implementation cost, multiuser interference, and mutual information are also derived to evaluate the system performance. In [159] and [160], NOMAaided SM schemes with finite alphabet inputs are proposed for uplink and downlink communications, respectively, in which the lower bound on mutual information is derived to characterize the system’s achievable rate.
It is noted that NOMA generally exploits interuser channel strength disparities to achieve an effective spectral utilization. Nevertheless, NOMA performance may degrade considerably when different users have a similar channel gain. Noticing this issue, [161] proposes a switching mechanism between NOMA and SM, which maintains better spectral efficiency under minor interuser channel strength disparities. Against the switching mechanism, the combination scheme of NOMA and SM is developed for uplink transmissions in [162], in which a hybrid detection scheme is proposed to estimate the antenna index and then the constellation symbols. In [163] and [164], the combination scheme of NOMA and SM is proposed for vehicletovehicle communications, in which the index of one active antenna embeds one specific user’s information and the superposition coded symbol carried by the active antenna conveys the information for other users.
It is worth pointing out that in the works mentioned above, the implementation of superposition coding and SIC may incur a high computational cost as the number of users grows large. Moreover, interference cancellation can be imperfect in practice, which causes error propagation and further degrades the system performance. To circumvent this issue, the study of [58] develops a primitive SM assisted NOMA technique with antenna partitioning to support multiuser access, which avoids the utilization of SIC at the receiver and refrains from multiuser interference. Furthermore, the SMaided NOMA scheme is investigated under a threenode cooperative network in [57], in which the information carried by the index of one active antenna and the constellation symbol is separately delivered to two users to avoid the utilization of SIC as well as improve the system performance. To support massive connectivity, SM aided codedivision multipleaccess is proposed as a NOMASM scheme for uplink communications in [165], whose performance and lowcomplexity signal detection problem are also investigated.
IvC Security provisioning in SM
Due to the broadcast characteristics of the wireless interface, like other wireless transmission technologies, SM transmissions also face the security and information leakage challenges. Specifically, the confidential messages delivered to legitimate users may be intercepted by malicious eavesdroppers. As a result, ensuring secure transmission of SM becomes a critical topic in the application of wireless networks. By complementing cryptographic techniques at the upper networking layers [166]
, PLS is an efficient alternative to improve the secrecy performance by exploiting the features of the wireless interface (e.g., randomness, variability, discrimination, reciprocity, etc.)
[61]. It is worth pointing out that from the physical layer perspective, both PLS and SM highly rely on the randomness and discrimination properties of the wireless interface. Thus, an interesting problem arises when exploiting the channel properties to achieve confidential information exchange among legitimate nodes using SM techniques such that eavesdroppers fail to recover the information. Therefore, advanced signal processing techniques may be applied to amplify the discrimination between the legitimate channel and wiretap channel, which benefits the legitimate node and impairs the eavesdropper at the same time. Moreover, SM activates one antenna randomly for each channel use via a fast antennaswitching mechanism, which presents a fast timevarying environment to the received signal of the eavesdropper. This informationdriven antennaswitching mechanism makes it more difficult for the eavesdropper to intercept the modulation information carried on the randomly activated antenna, especially when the mapping rule is unknown to the eavesdropper. Therefore, SM enjoys a great potential not only in the increased spectral efficiency and energy efficiency, but also in enhancing secure transmission.In [59] and [60], some preliminary results on the secrecy capacity for both SSK and SM are presented, and the effects of the constellation size, received signaltonoise ratio (SNR), and number of antennas at the transceiver on the secrecy behavior are examined. Assuming that the CSI of the eavesdropper is available in the multipleinputsingleoutput (MISO) system, [61] studies the secrecy mutual information of SM under the finite input alphabet constraint and develops a precoding based SM method to enhance the security performance. Under the assumption of an available global CSI at the transmitter, the optimization problem of the precodingaided SM system is investigated in [139] to improve the secure performance, which maximizes the ratio of the received signal power at the legitimate user to that at the eavesdropper. However, acquiring the eavesdropper’s CSI may not be possible at the transmitter for passive eavesdroppers. Without knowing the eavesdropper’s CSI, artificial interference is introduced to the SM system for ensuring secure transmission in [138, 167, 137, 168], where [137] extends the precodingaided SM technique to a multiuser downlink case against the multiple antenna eavesdropper.
In [169], two random antenna selection methods are developed for precoding based SM to improve the secrecy rate, which severely degrades the reception quality of the passive eavesdroppers without degrading the received signal at destination. In [170], two transmit antenna selection methods based on the maximization of signaltoleakageandnoise ratio and secrecy rate are developed for artificial interferenceaided SM systems to enhance the transmission security. A fullduplex legitimate receiver is introduced to the secure SM system in [171], which not only receives the secret information from the source, but also emits jamming signals to cause timevarying interference at the passive eavesdropper. Under the framework of QSM, an antieavesdropping scheme is proposed by transmitting mixed signals of the information symbol and the carefully designed artificial noise via randomly activated antennas [172]. In [173], by exploiting the randomness of the constellation mapping principle to encrypt the secret information, a novel scheme is presented for the secret key exchange and authentication in the SM system.
In [174], based on the channel reciprocity of timedivision duplexing communications and knowledge of legitimate CSI, the transmitter dynamically redefines the mapping rule of the spatial information to the active antenna indices for security purpose, and then chooses the active antenna according to the redefined mapping rule. Since a redefined mapping rule is unknown by the eavesdropper due to lack of the legitimate CSI, the security of the spatial information can be guaranteed whereas the constellation information is not secure. Attentive to the security issue of the constellation information, [175] dynamically rotates the active antenna indices and the constellation symbols at the transmitter according to the instantaneous legitimate CSI, which is totally unknown to the eavesdropper. As a result, both the spatial information and constellation information are secure from the eavesdropper. Moreover, by leveraging the random state of the legitimate channel, the mappingvaried SM scheme is proposed in [176], which dynamically rearranges the mapping patterns of the antenna index bits and constellation bits according to the instantaneous legitimate channel quality. Without knowing the legitimate channel state, the eavesdroppers are confused with variable information mapping patterns and the PLS can be further enhanced.
The design of the SM technique for secure cooperative communication systems is also investigated in [177] and [178]. In [177], a securely distributed SM scheme is deployed in multiple nondedicated relays to defend against eavesdropping attacks and coordinate the transmission of sourcedestination information via the indices of the selected relays. In particular, the selected relays are equipped with specifically designed precoder to facilitate the decoding of spatial information at the legitimate receiver while confusing the eavesdropper. In [178], a precoded SMbased cooperative network using relay selection and jamming is investigated with secrecy considerations, where Alice applies precoded SM and Bob emits a jamming signal simultaneously for security protection in the first phase while multiple relays coordinate the information transmission and jam the eavesdroppers in the second phase. Moreover, the power allocation optimization problem among the cooperative relays and the performance in terms of secrecy capacity and outage probability are studied.
V Applications of SM to Emerging Communication Systems
Va SM in mmWave Communications
Driven by the high cost and exhaustion of radio resources in low frequency bands (below 6 GHz), the exploitation of mmWave frequency band (around 60 GHz) alleviates the data traffic burden and provides super high data services in a costefficient manner [179, 180]. However, working in highfrequency bands, mmWave communications suffer from high propagation loss. To overcome this drawback, beamforming technology can be applied with a large number of antennas to compensate for high propagation attenuation. Fortunately, attributed to the significantly reduced wavelength in the mmWave frequency band, mmWave MIMO systems can be equipped with a large number of antennas at the transceiver in a highly compact manner to achieve high beamforming gains [54]. For the conventional mmWave MIMO system, the high compaction of antennas imposes the prohibitively high cost at the transceiver due to the use of a large number of RF circuits, which hinders the practical implementation. To explore the practical implementation of RFchainlimited mmWave systems, the (G)SM technique is a promising lowcost and highefficiency design option by activating a subset of transmit antennas connected to a small number of RF chains. As a result, (G)SM becomes an attractive technology in mmWave communications for reaping benefits of both beamforming gain and low cost with reduced RF chains.
Fig. 8 shows the transmitter diagram of the (G)SM aided mmWave communication systems. A total number of transmit antennas are partitioned into analog precoders, each of which comprises antenna elements for analog precoding, i.e., . For the implementation of (G)SM aided mmWave scheme, the number of RF chains is less than the number of analog precoders . Unlike the conventional mmWave MIMO scheme that conveys information in the constellation domain, the incoming data of (G)SM aided mmWave scheme are separated into two parts in the constellation domain and space domain, respectively. Specifically, the space domain data are fed to the index selector to determine the indices of out of analog precoders for activation. Since the remaining analog precoders are set to be idle, the outputs of RF chains are connected with activated analog precoders via a fast switcher and the total combination number of analog precoders is . On the other hand, the constellation data are fed to the symbol modulator to generate constellation symbols, and then independently processed with RF chains. For each analog precoder, antenna elements are deployed as phase shifters for the analog procession. A diagonal matrix of size is applied as an analog precoder, whose th diagonal entry is given by
(12) 
where is the rotation angle of the th transmit antenna. Similar to (7), we denote the (G)SM block as , and the received signal model can be expressed as
(13) 
where denotes the allone vector of size , denotes the Kronecker product, is the mmWave MIMO channel matrix of size , and is an AWGN vector at the receiver. Note that when , the system in (12) specializes to a SM aided mmWave system with only a single RF chain, which enjoys an extremely low cost of implementation and detection in mmWave communications. (G)SM aided mmWave scheme not only saves the transmitter cost by reducing the number of RF chains, but also achieves the precoding gains via the analog precoder, which meets the requirement of mmWave communications.
In [181], the implementation of analog beamformers in (G)SMbased mmWave systems is investigated, where the array of analog beamformers is optimally designed to attain the maximum rank of the equivalent channel matrix. Moreover, it is shown that with a reduced RFchain structure, the (G)SMbased mmWave scheme of [181] approaches the same constrained capacity as the fullRF spatialmultiplexing counterpart. The application of SSK and (G)SM in indoor mmWave communications at 60 GHz is studied in [182, 183, 184], where the channel is characterized as lineofsight (LOS) components due to the high reflection attenuation. Based on the maximum criterion of the minimum distance of received symbols, SSK [182] and (G)SM [183, 184] in LOS channels are designed to achieve optimal performance by orthogonalizing the channel matrix with proper antenna placement. By approximating the 3D statistical channel as a log–normal fading channel, [185] analyzes the capacity of QSM for outdoor mmWave communications. Assuming the CSI at the transmitter, [186] proposes an analog precoding aided (G)SM scheme for mmWave communications to improve the spectral efficiency. Specifically, after deriving a tight lower bound on the achievable spectral efficiency, the analog precoding aided (G)SM scheme in [186] is designed in an iterative way to meet the maximal spectral efficiency. Furthermore, hybrid precoding aided (G)SM schemes using hybrid digital and analog precoding are proposed for mmWave communications, which achieve an optimal spectral efficiency based on some approximations [187, 188].
To save the transmitter cost with a reduced number of antennas, (G)SM with a variable number of activated antennas is designed in [189] for indoor LoS mmWave communications to achieve higher transmission efficiency. Moreover, the capacity and energy efficiency are analyzed, and the optimization problems in terms of power allocation and antenna separation are investigated. To save energy consumption and allow robustness against pathloss attenuations, an orbital angular momentum aided SM scheme is proposed for mmWave communication systems in [190], where the capacity, error performance, and energy efficiency are also discussed. In [191]
, a SM aided receive antenna selection technology is developed for saving the RF cost at the transceiver in mmWave communications, where the combinatorial optimization problems of capacity and error performance are studied. To attain beamforming gain and reduce the cost of RF chain, an analog precodingaided virtual SM
[192] and its variant using multimode hybrid precorder [193]are proposed, which enhance the received SNR and spatial degrees of freedom with the low cost of single RF chain. Receive SM is also studied in indoor and outdoor mmWave communications
[194, 195, 196], where the performance is optimized with minimum symbol error probability in LOS channels [194].VB SM in Optical Wireless Communications (OWC)
In addition to mmWave communications discussed in the last subsection, OWC is another appealing complement to alleviate the shortage problem of wireless radio resources and achieve high data rate transmission [197], which can be categorized into indoor and outdoor implementations depending on the communication range. For example, by deploying lowcost LEDs for the indoor environment, OWC can provide simultaneous energyefficient lighting and high throughput transmission. As an efficient and inexpensive dualuse technology of illumination and communications, OWC enjoys a lot of advantages in both indoor and outdoor communications, such as cost efficiency, licensefree implementation, large bandwidth, and high reliability.
The primitive optical SM is developed in [198] for the indoor environment, which can be extended to various scenarios. In [199], an optical SM is developed for outdoor optical communications to achieve a higher spectral efficiency and energy efficiency, where the analytical and simulation results of both uncoded and coded systems are presented under the turbulence induced fading. Based on coherent detection, a theoretical framework for error performance analysis is presented for both uncoded and coded optical SM systems over atmospheric turbulence characterized by HK distribution in [200], which shows enhancement in performance, spectral efficiency and deployment cost. Under the consideration of atmospheric turbulence and pointing error, the error performance and mutual information of the optical SSK are studied over three different optical channel models [201]. Considering the intersymbol interference incurred by delay spread, [202] studies the error behavior of the optical SM over indoor multipath optical channels, whose upper bound on error performance is also derived. With CSI available at the transmitter, [203] proposes an optical SM based on the constellation optimization, whose minimum Euclidean distance is maximized at the receiver. To deal with the adverse effect on the detection of optical SM incurred by the high channel correlation, an optical SM based on the power imbalance is proposed for OWC in [204], which optimizes the constrained capacity using power allocation. In [205], the error performance of the SM aided OWC system is investigated over the indoor environment characterized by LOS components, which outperforms the repetition coding aided OWC system. In [206], the mutual information and its corresponding lower bound of the SM aided OWC system are theoretically calculated under the MISO and finite input alphabet setting. Based on these theoretical results, the optimal precorder is designed to maximize the minimum Euclidean distance. An enhanced optical SM scheme is proposed for indoor OWC in [207], which aligns multiple transmitreceive units to achieve higher SNR gains. Under the consideration of imperfect synchronization at the transmitter, the theoretical error performance of several variants of optical SM is derived for OWC in [208], which provides some insightful information about the effect of synchronization error.
For indoor environments, optical wireless communications can be deployed as VLC for simultaneous luminance and communication. In particular, as one indoor implementation of OWC, VLC is an attractive alternative to apply in some RFfree working environments, e.g., hospital, airplane, and gas station [209, 210]. Due to the wide application of LED array comprising a large number of LEDs for sufficient illumination, MIMO technology can be readily integrated into VLC to achieve high transmission rates, where various representative MIMO aided VLC schemes have been investigated and compared in [62]. Thanks to its capability of supporting fast switching in LEDs, SM aided VLC can be deployed, which activates a single LED only in the LED array for each channel use to convey index information additionally through the location of the active LED. Moreover, by adjusting the number of active LEDs and the intensities, both (G)SSK and (G)SM provide the flexibility to control the luminance of the LED array and the communication throughput. To attain higher power efficiency, an enhanced optical (G)SM scheme based on the collaborative constellation pattern is developed for indoor VLC in [211, 212, 213]. Using the MLbased photodetectors, the theoretical and experimental results of (G)SSK are presented for the shortrange indoor VLC in [214] and [215]. The error performance of (G)SM aided VLC systems is presented with an analytical upper bound on BER in [216]. Exploiting available CSI at the transmitter for interchannel interference reduction, an optical SM based on channel adaption is designed to apply in massive MIMOVLC in [217]. Moreover, the combination of optical SM and layered spacetime coding is designed for VLC by mapping the spatial information into different spacetime coding matrices. To achieve the control flexibility on luminance and the data throughput, a (G)SM aided hybrid dimming strategy using clipped optical OFDM is designed for VLC in [218], whose achievable rate is theoretically analyzed over channels with dense scattering and high correlation, respectively. To achieve a flexible design with an arbitrary number of transmit antennas for SM, a threedomain bit mapping solution based on channel adaptation is suggested in [219], whose lower bound on achievable rate is also obtained in closedform. Noticing the real and positive constraint on light intensity levels in intensity modulation/direct detection system, the authors of [220] and [221] successfully apply the (G)SM principle to the optical MIMOOFDM system, which exploits the advantage of SM based LED to avoid Hermitian symmetry and direct current (DC) bias. Specifically, each complex OFDM signal are split into four components consisting of the realimaginary and positivenegative parts and then encoded by four LEDs for transmission, which dispenses with DC biasing/asymmetrical clipping [220].
VC Hardware Implementation of SM
It is worth pointing out that most research works on SM make the impractical assumption of ideal hardware at the transceiver. Therefore, practical implementation of the SM family is an important and challenging research direction, since the channel estimation, transceiver movement, and hardware impairments due to phase noise, I/Q imbalance, and nonlinearity of the amplifier can have negative effects on the system performance. The first attempt on the hardware implementation and experimental assessment of the SM principle is reported in [222], where the testbed platform with two National Instruments (NI) PXIe devices is designed to verify the experimental performance of SM under the indoor propagation environment characterized by LoS wireless channels. In particular, the implementation and applicability of SM has been first approved, which reveals the impact of various hardware impairments at the transceiver, e.g., imperfect channel estimation, spatial correlation, and I/Q imbalance.
On the other hand, it is preferable to use lowresolution analogtodigital converters (ADCs) with fewbits quantization at the user terminal to achieve a much lower power consumption and reduce hardware cost significantly. In [223], the multiuser detection problem of the SM system is investigated for the largescale MIMO implementation employing lowresolution ADCs. It is shown that the SM system developed in [223] is robust to quantization error and channel correlation, and the proposed lowcomplexity detector achieves better performance than its linear counterparts. Noted that the antennaswitcher incurs time delays and is not costfree, [224] studies the impact and limitation of the antennaswitching time for SM, in which the spectral efficiency and achievable rate are analyzed under the practical hardware constraint of antennaswitching. In [225] and [226], the performance of SM based cognitive radio communications is investigated under the consideration of channel estimation error and hardware imperfections, where some theoretical results are derived for Rayleigh fading channels.
The hardware requirements and limitations of various SM members are discussed in [227], which shows that hardware impairments have a significant influence on practical performance. After developing an analytical framework under the transmit I/Q imbalance for SM, the authors of [228] show the robustness of SM to the I/Q imbalance. In [229], a practical hardware implementation of the SM detector is realized by using 87.4K logic gates in the 0.18m CMOS technology, which adopts the signalvectorbased list detection and dualdatapath architecture to achieve low hardware complexity and nearoptimal performance. Furthermore, hardware realization of the ML detection is implemented with 70.5K gates under the TSMC 90nm CMOS technology in [230], where the Givens rotation is applied to achieve stable division operation and lowcomplexity implementation.
VD SM Based Simultaneous Wireless Information and Power Transfer (SWIPT)
In all works mentioned above, the inactive antennas of an SM node are either kept idle or turned on for signal reception when operating in fullduplex mode. With the development of SWIPT technology, researchers begin to realize a new energy harvesting function of the inactive antennas in SM systems. SM based SWIPT has been therefore receiving considerable attention recently. Since the research topic is relatively new, there are only a few works in the literature. In summary, they can be still classified into three categories.
In the first category, the SM signal is exploited for SWIPT by power splitting [231], [232], [233], [140]. Specifically, in [231] a power splittingbased SWIPT is considered for an SM system with a multipleantenna receiver. The authors assume different power splitting factors for different receive antennas and develop an iterative algorithm to find the optimal values in maximizing the achievable rate under certain energy harvesting constraints. [232] develops a dualhop relaying network using SM at both sources and physicallayer network coding, where the relay obtains required energy from the RF SM signals via the power splitting protocol. Both fullduplex and halfduplex modes are considered for the network, and the fullduplex mode is shown to outperform its halfduplex counterpart as the quality of selfinterference cancellation and/or the spectral efficiency increases. While [231] and [232] consider transmit SM, [233] and [140] shed light on the application of receive SM to SWIPT. In [233], an energy pattern aided SWIPT system is proposed, where the information is conveyed not only by the indices of the activated receive antennas but also by the specific intensity of the delivered power. Results on BER demonstrate the beneficial immunity of the proposed system to power conversion. Unlike [233] that focuses on joint design, [140] proposes to transmit a separated energy and receive SM signals, such that the predefined energy signal can be removed at the information receiver to facilitate signal detection.
In the second category, the SM node participates in both information decoding and energy harvesting [234], [235]. This idea was first coined in [234], where a GSM node communicates to an information receiver through the active antennas while harvesting energy transmitted from an energy transmitter and recycling part of its selfenergy through the inactive antennas. Enhancement in terms of achievable rate over the existing time switchingbased and antenna selectionbased wireless powered communication protocols is verified therein. Later, the idea was extended to a dualhop relaying network in [235], where a GSMbased fullduplex relay without an external power supply is assumed, and three different realizations were proposed. While both the first and second methods are based on time switching, the third method is based on power splitting. The first method performs energy harvesting first, suffering from selfinterference in the following information decoding step, while by reversing the order the second method turns the adverse selfinterference into an advantage to energy harvesting as [234]. The third method carries out information decoding and energy harvesting simultaneously all the time, improving the transmission efficiency.
In the third category, the SWIPT capability is developed for distributed SM [236]. Assuming multiple geometrically separated singleantenna relays to form distributed SM, the authors of [236] propose two different protocols to enable SWIPT. The first protocol allows each relay node to harvest energy transmitted from the source, which is used by the activated relay to forward the source’s and its own data simultaneously. The second protocol exploits the inactive relays to recycle part of the transmitted energy in the network in addition to harvesting the transmitted energy from the source. Both protocols retain the original properties of DSM, but the second protocol enjoys a lower hardware complexity and better error performance than the first protocol in certain scenarios.
Vi Applications of SM in New Domains
Type  Entity  Representative Schemes  Main Achievements  

Antenna  SM[6], SSK[15], (G)SM[22], (G)SSK[34], QSM[26]  Higher energy efficiency & lower deployment cost  
Space Domain  LED  OSM[237]  Flexibility to control illumination & communications  
RF Mirror  RASSK[238], CM[239]  Better system performance & lower antenna cost  
Frequency Domain  Subcarriers  SIMOFDM [240], OFDMIM[98], OFDMI/QIM[241], GFDMIM[242]  Higher spectral efficiency& higher energy efficiency  
SingleDimension Entity  Subcarriers  CIOFDMIM[243], MIMOOFDMIM[244, 245, 246, 247]  Higher reliability& higher energy efficiency  
Time Domain  Time slot  SCIM[248]  Higher transmission efficiency for broadband systems  
Code Domain  Spreading Code  CIMSS[249], GCIMSS[250], IMOFDMSS[251]  Higher spectral efficiency & lower energy consumption  
Modulation Type  ESM[252], DMOFDM[253], MMOFDMIM[254]  Higher spectral efficiency  
Angle Domain  Polarization state  PolarSK[255], 3D PMod[256]  Higher spectral efficiency & lower hardware cost  
AoA  BACM[257]  Higher spectral efficiency  
SpaceTime  STSK[109]  Flexible tradeoff between diversity & multiplexing gain  
MultiDimension Entity  SpaceFrequency  GSFIM[258]  Better system performance  
SpaceTimeFrequency  STFSK[259], GSTFIM[260]  Higher spectral efficiency 
As a novel digital modulation technology for achieving both spectral efficiency and energy efficiency, the basic concept of SM is to convey additional information through the ON/OFF states of transmit antennas. Consequently, the concept of SM can be generalized and used in various applications, in which the transmission media for embedding additional information can be either physical (e.g., antenna) or virtual (e.g., spacetime matrix). Instead of using the fullactivation state, the distinguishing feature of SM is to embed the implicit information via the activation states of transmission media in addition to explicit information (e.g, constellation symbols), which is carried by the partially activated transmission media [261, 262, 263, 35, 264]. The information loss due to the deactivation of some transmission media can be compensated for by the implicit information based on the ON/OFF states, saving the cost of transmission media (the number of actual transmission media is significantly reduced) and striking an attractive compromise between energy efficiency and spectral efficiency.
Relying on the activation states of transmit antennas, SM provides a new perspective on the digital modulation, which is different from the conventional digital modulation using the amplitude, frequency, and phase on the sinusoidal carrier. Inspired by the concept of SM, researchers have proposed various forms and applications for SM, which alter the ON/OFF states of different transmission entities to embed additional information. In Table I, we summarize extensive representative forms of SM, which are classified according to their transmission entities that perform the ON/OFF keying mechanism. As Table I, the concept of SM can be performed in either a singledimension entity or a multidimension entity. So far, various singledimension entities have been explored to increase the spectral efficiency and energy efficiency of communication systems, which mainly includes spatial entities (e.g., antenna, RF mirror, and LED), frequency entities (e.g., subcarrier), time entities (e.g., time slot), code entities (e.g., spreading code and modulation type), angle entities (e.g., AoA and polarization state), etc. To further enhance the system performance and enjoy a more flexible system setting, several multidimension entities are also developed, which include more than one dimension for performing the ON/OFF keying mechanism.
Via SingleDimension Entity
After the pioneering works in [6] and [15], the emergence of SM and SSK schemes has sparked renewed and widespread interest in digital modulation, both of which require a single RF chain only and embed spatial information through the active antenna index. Afterwards, different variants of the SM technology have been developed to accommodate more than one active antennas and further enhance the spectral efficiency, which include (G)SM [22], (G)SSK[34] and QSM [26]. Besides the spatial entity of antenna, the implementation of SM can also be extended to communication systems deploying RF mirrors and LEDs. As presented in Section VB, the SM aided VLC, which conveys spatial information through the location of the active LED, is an efficient and inexpensive dualuse technology of illumination and communications [237]. Another remarkable implementation of SM in the space domain is the reconfigurable antenna (RA) based system, which controls the ON/OFF states of the RF mirrors surrounding a transmit antenna and further alters the radiation pattern to embed additional information [265, 266, 267, 268]. With the adjustable radiation characteristics of the transmit antenna, RASSK is proposed in [238] to improve the system performance and reduce the implementation cost in transmit antennas. Moreover, as the changing of radiation pattern results in the variations of the channel characteristics, it can be regarded as the modulation based on the channel state, which is also referred to as channel modulation (CM) schemes [239].
Although the SM technology originated in the space domain, SM is not exclusive to the space domain and can be easily transplanted into the frequency/time/code/angle domains. Inspiring by the SM concept, the authors of [240] introduce the subcarrierindex modulation (SIM)OFDM system, which is the first attempt to exploit the subcarrier indices for encoding information and alleviating the high peaktoaverage power ratio issue in OFDM systems. By replacing the transmission of antennas with subcarriers for performing the ON/OFF keying mechanism, orthogonal frequency division multiplexing with index modulation (OFDMIM) is proposed in [98] as a successful and inspiring representative for implementing the SM concept into multicarrier systems. Inspired by the design of the QSM scheme, OFDM with inphase/quadrature index modulation (OFDMI/QIM) is proposed in [241] to perform the ON/OFF keying mechanism independently in the inphase and quadrature dimensions, respectively, which further enhances the spectral efficiency and the error performance. In [269], the generalization of OFDMIM is proposed to further enhance the spectral efficiency and flexibility with a variable number of active subcarriers, which relaxes the constraint of the fixed number of active subcarriers in [98]. The concept of SM is also applied to generalized frequency division multiplexing (GFDM) in [242], which yields the GFDMIM scheme and achieves performance gain over conventional GFDM. To achieve higher diversity gain, a novel coordinate interleaved OFDMIM (CIOFDMIM) scheme is proposed in [243] by combining the SM concept and coordinate interleaved STBC in OFDM systems. The combination scheme of OFDMIM and MIMO, which referred to as MIMOOFDMIM, is investigated in [244] to achieve higher data rates and better error performance than the traditional MIMOOFDM scheme, where various lowcomplexity and nearoptimal algorithms are developed for the demodulation of MIMOOFDMIM [244, 245, 246, 247].
The successful implementation of OFDMIM has also propelled the spread of the SM concept to the time, code, and angle domains. In [248], a novel singlecarrier (SC) based IM (SCIM) scheme, which activates a subset of symbols in each timedomain subframe to convey additional information, is proposed to enhance the transmission efficiency for broadband systems. A code index modulationspread spectrum (CIMSS) is proposed in [249] to achieve better performance with reduced energy consumption, which carries additional information by choosing one of the two spreading Walsh codes. Moreover, a generalization scheme of CIMSS (called GCIMSS) is then proposed in [250] to further improve the spectral efficiency, which increases the set size of the spreading Walsh codes to bear more index information. By integrating the SM concept into OFDM spread spectrum (OFDMSS) systems, the authors of [251] propose an index modulated OFDMSS (IMOFDMSS) scheme to harvest the diversity gain and higher spectral efficiency, which embeds the index information through the selection of spreading codes and can be applied to multiuser systems. By enabling the flexibility of activating one or two antennas, an enhanced SM scheme is developed in [252] to overcome the compromised spectral efficiency of conventional SM, which jointly select the active antenna(s) and the constellations for embedding more index information. To mitigate the loss in spectral efficiency due to the inactive subcarriers, a dualmode OFDM (DMOFDM) is proposed in [253] to enhance the data transmission efficiency, in which the originally active subcarriers of OFDMIM carry the modulated symbols drawn from the primary constellation while those inactive subcarriers of OFDMIM are reactivated to carry new modulated symbols drawn from a secondary constellation. Furthermore, the generalization scheme of DMOFDM, which is referred to as multiplemode OFDMIM (MMOFDMIM) [254] and includes more constellation modes, is developed by exploiting the full permutation pattern to carry more index information. By exploiting polarization states for conveying index information through the angle domain, a generalized polarization shift keying (PolarSK) modulation scheme, is proposed for dualpolarization communication systems in [255] to achieve high spectral efficiency while retaining a low hardware cost. Moreover, a 3D polarized modulation (PMod) is developed for dualpolarization communication systems in [256], which selects both polarization state and the radiation phase according to the information bits to achieve better error performance. In [257], a beam angle channel modulation (BACM) scheme is introduced to convey index information in the angle domain, which embeds the information into the AoA of a transmission beam.
ViB MultiDimension Entity
As discussed in the last subsection, various applications based on the SM concept have been implemented in the space/frequency/time/code/angle domains, which exploit the ON/OFF states or the permutations of transmission entities in one domain to carry index information. To further enhance the system performance and enjoy a more flexible system setting, it is possible and beneficial to integrate resources across multiple dimensions to form a multidimension entity for the implementation of SM concept.
To unify the SM concept in both space and time domains, both coherent and differential STSK modulation schemes are developed via the activation mechanism on selecting one out of multiple spacetime dispersion matrices, which enables a flexible design to compromise between diversity and multiplexing gains [109]. By extending the SM concept to include both space and frequency dimensions, the authors of [258] develop a generalized space and frequency IM (GSFIM) scheme, which shows the potential to achieve better performance than the traditional MIMOOFDM scheme. Exploiting the ON/OFF keying mechanism across the space, time and frequency domains, spacetimefrequency shift keying (STFSK) scheme is also proposed in [259] to reap the potential gain from the three domains, which shows an improvement in the system performance and alleviation of intersymbol interference over frequencyselective fading channels. Based on the dispersion matrix activation over space, time and frequency domains, a generalized spacetimefrequency index modulation (GSTFIM) scheme is proposed in [260] to provide extra spectral efficiency.
Vii Conclusions and Future Directions
SM is a promising digital modulation technology to fulfill the requirements of emerging wireless systems due to its potential of achieving high energy efficiency, low deployment cost, free of interchannel interference, relaxed interantenna synchronization requirements, and compatibility with massive MIMO systems. As shown in this survey, relying on the activation states of transmit antennas to convey additional information, SM can achieve an attractive compromise between spectral efficiency and energy efficiency with simple design philosophy, which has been verified by extensive studies. We have first discussed the basic principles, variants, and enhancements of SM, and then shown the broad prospects of the SM concept in various implementations, including integration with other promising techniques, applications to emerging communication systems, and extensions to new domains. We hope that this survey and the research results in this special issue will be helpful to the readers to gain a better understanding and clearer picture on the advantages and opportunities of the SM family as well as its widerange applications.
Although considerable significant work has been studied on the research filed of SM, there are still many interesting as well as challenging research problems of SM to be tackled in order to further broaden its applications and fulfill the requirements of future wireless communications. Therefore, we close this survey with some promising directions worthy of investigation in future research, which are summarized as follows:

The scalability and integration of SM techniques to multiuser networks. In particular, scheduling and resource allocation of SM basednetworks is one of the indispensable tasks to maximize the system throughput and balance user fairness. Moreover, the antenna and power allocation problem of SM should be further investigated to achieve an effective multiuser interference management.

The highmobility management problem of the SM technology in dynamic and mobile networks such as vehicletoeverything (V2X) communications, unmanned aerial vehicles (UAV) communications, and underwater acoustic (UWA) communications. For example, since the occupation of pilots consumes a nonnegligible fraction of space/time/frequency/code resources, the number and pattern of pilots have a significant impact on the highmobility SM system, which should be carefully designed. Moreover, similar to the concept of SM, the distinguishable pilot patterns may be exploited to carry additional information bits and compensate for the overhead cost of the pilot occupation in the highmobility SM system.

Although SM is a promising candidate for massive MIMO, the channel acquisition/estimation can be a challenging task due to the massive number of antennas. In fact, the required channel training overhead is proportional to the number of antennas, which may be prohibitive or even unaffordable in practice. Furthermore, the channel feedback problem becomes even more prominent and challenging to the linkadaptive SM system under the massive MIMO setup.

Note that the error performance of the spatial information bits and the constellation information bits of SM can be significantly different under various wireless environments. This is attributed to the fact that the error performance of spatial information bits highly depends on the differentiability of the channel signatures. Therefore, how to balance the error performance difference and schedule the information bits is an essential problem of SM.

Relying on the ON/OFF state for embedding information, SM can be an energy efficient and costeffective technique for the IoT to connect a massive number of machinetype communication (MTC) devices. Since MTC devices are usually low data rate requirements, periodic data traffic arrivals, limited signal processing capability, and strict energy constraints, the specific design and integration of SM to the IoT is also a promising research direction.

The fundamental tradeoff problem of SM between energy/spectral efficiency and the other communication requirements such as high reliability, high capacity, flexible design, and low latency.

The application/generalization of the SM concept to new emerging transmission entities such as orbital angular momentum can also bring new challenging problems.
References
 [1] L. Lu, G. Y. Li, A. L. Swindlehurst, A. Ashikhmin, and R. Zhang, “An overview of massive MIMO: Benefits and challenges,” IEEE J. Sel. Areas Commun., vol. 8, no. 5, pp. 742–758, Oct. 2014.
 [2] J. G. Andrews, S. Buzzi, W. Choi, S. V. Hanly, A. Lozano, A. C. K. Soong, and J. C. Zhang, “What will 5G be?” IEEE J. Sel. Areas Commun., vol. 32, no. 6, pp. 1065–1082, Jun. 2014.
 [3] V. W. Wong, R. Schober, D. W. K. Ng, and L.C. Wang, Key technologies for 5G wireless systems. Cambridge university press, 2017.
 [4] E. Dahlman, G. Mildh, S. Parkvall, J. Peisa, J. Sachs, Y. Selén, and J. Sköld, “5G wireless access: Requirements and realization,” IEEE Commun. Mag., vol. 52, no. 12, pp. 42–47, Dec. 2014.
 [5] R. Mesleh, H. Haas, C. W. Ahn, and S. Yun, “Spatial modulation  a new low complexity spectral efficiency enhancing technique,” in Proc. IEEE Int. Conf. Commun. Netw. in China, Beijing, China, Oct. 2006, pp. 1–5.
 [6] R. Y. Mesleh, H. Haas, S. Sinanovic, C. W. Ahn, and S. Yun, “Spatial modulation,” IEEE Trans. Veh. Technol., vol. 57, no. 4, pp. 2228–2241, Jul. 2008.
 [7] M. D. Renzo, H. Haas, A. Ghrayeb, S. Sugiura, and L. Hanzo, “Spatial modulation for generalized MIMO: Challenges, opportunities, and implementation,” Proc. IEEE, vol. 102, no. 1, pp. 56–103, Jan. 2014.
 [8] M. Di Renzo, H. Haas, and P. M. Grant, “Spatial modulation for multipleantenna wireless systems: A survey,” IEEE Commun. Mag., vol. 49, no. 12, pp. 182–191, Dec. 2011.
 [9] P. Yang, M. Di Renzo, Y. Xiao, S. Li, and L. Hanzo, “Design guidelines for spatial modulation,” IEEE Commun. Surveys Tuts., vol. 17, no. 1, pp. 6–26, First quarter 2015.
 [10] P. Yang, Y. Xiao, Y. L. Guan, K. V. S. Hari, A. Chockalingam, S. Sugiura, H. Haas, M. Di Renzo, C. Masouros, Z. Liu, L. Xiao, S. Li, and L. Hanzo, “Singlecarrier SMMIMO: A promising design for broadband largescale antenna systems,” IEEE Commun. Surveys Tuts., vol. 18, no. 3, pp. 1687–1716, Third quarter 2016.
 [11] A. Stavridis, S. Sinanovic, M. Di Renzo, and H. Haas, “Energy evaluation of spatial modulation at a multiantenna base station,” in Proc. IEEE Veh. Technol. Conf. (VTC Fall), Las Vegas, NV, USA, Sept. 2013, pp. 1–5.
 [12] A. Stavridis, S. Sinanovic, M. Di Renzo, H. Haas, and P. Grant, “An energy saving base station employing spatial modulation,” in Proc. IEEE Int. Workshop on Comput. Aided Modeling and Design of Commun. Links and Netw. (CAMAD), Barcelona, Spain, Sept. 2012, pp. 231–235.
 [13] D. A. Basnayaka, M. Di Renzo, and H. Haas, “Massive but few active MIMO,” IEEE Trans. Veh. Technol., vol. 65, no. 9, pp. 6861–6877, Sept. 2016.
 [14] Y. Cui and X. Fang, “Performance analysis of massive spatial modulation MIMO in highspeed railway,” IEEE Trans. Veh. Technol., vol. 65, no. 11, pp. 8925–8932, Nov. 2016.
 [15] J. Jeganathan, A. Ghrayeb, L. Szczecinski, and A. Ceron, “Space shift keying modulation for MIMO channels,” IEEE Trans. Wireless Commun., vol. 8, no. 7, pp. 3692–3703, Jul. 2009.
 [16] M. D. Renzo, D. D. Leonardis, F. Graziosi, and H. Haas, “Space shift keying (SSK) MIMO with practical channel estimates,” IEEE Trans. Commun., vol. 60, no. 4, pp. 998–1012, Apr. 2012.
 [17] J. Choi, “Sparse signal detection for space shift keying using the Monte Carlo EM algorithm,” IEEE Signal Process. Lett., vol. 23, no. 7, pp. 974–978, Jul. 2016.

[18]
H. W. Liang, W. H. Chung, and S. Y. Kuo, “Codingaided Kmeans clustering blind transceiver for space shift keying MIMO systems,”
IEEE Trans. Wireless Commun., vol. 15, no. 1, pp. 103–115, Jan. 2016.  [19] M. Di Renzo and H. Haas, “Improving the performance of space shift keying (SSK) modulation via opportunistic power allocation,” IEEE Commun. Lett., vol. 14, no. 6, pp. 500–502, Jun. 2010.
 [20] ——, “Bit error probability of spaceshift keying MIMO over multipleaccess independent fading channels,” IEEE Trans. Veh. Technol., vol. 60, no. 8, pp. 3694–3711, Oct. 2011.
 [21] ——, “Space shift keying (SSK) modulation with partial channel state information: Optimal detector and performance analysis over fading channels,” IEEE Trans. Commun., vol. 58, no. 11, pp. 3196–3210, Nov. 2010.
 [22] A. Younis, N. Serafimovski, R. Mesleh, and H. Haas, “Generalised spatial modulation,” in Proc. Conf. Rec. 44th Asilomar Conf. Signals, Syst. Comput, Pacific Grove, CA, USA, Nov. 2010, pp. 1498–1502.
 [23] J. Fu, C. Hou, W. Xiang, L. Yan, and Y. Hou, “Generalised spatial modulation with multiple active transmit antennas,” in Proc. IEEE Globecom Workshops (GC Wkshps), Miami, FL, USA,, Dec. 2010, pp. 839–844.
 [24] J. Wang, S. Jia, and J. Song, “Generalised spatial modulation system with multiple active transmit antennas and low complexity detection scheme,” IEEE Trans. Wireless Commun., vol. 11, no. 4, pp. 1605–1615, Apr. 2012.
 [25] B. Zheng, X. Wang, M. Wen, and F. Chen, “Soft demodulation algorithms for generalized spatial modulation using deterministic sequential monte carlo,” IEEE Trans. Wireless Commun., vol. 16, no. 6, pp. 3953–3967, Jun. 2017.
 [26] R. Mesleh, S. S. Ikki, and H. M. Aggoune, “Quadrature spatial modulation,” IEEE Trans. Veh. Technol., vol. 64, no. 6, pp. 2738–2742, Jun. 2015.
 [27] Y. Bian, X. Cheng, M. Wen, L. Yang, H. V. Poor, and B. Jiao, “Differential spatial modulation,” IEEE Trans. Veh. Technol., vol. 64, no. 7, pp. 3262–3268, Jul. 2015.
 [28] N. Ishikawa and S. Sugiura, “Unified differential spatial modulation,” IEEE Wireless Commun. Lett., vol. 3, no. 4, pp. 337–340, Aug. 2014.
 [29] M. Wen, X. Cheng, Y. Bian, and H. V. Poor, “A lowcomplexity nearML differential spatial modulation detector,” IEEE Signal Process. Lett., vol. 22, no. 11, pp. 1834–1838, Nov. 2015.
 [30] L. Yang, “Transmitter preprocessing aided spatial modulation for multipleinput multipleoutput systems,” in Proc. IEEE Veh. Technol. Conf. (VTC Spring), Yokohama, Japan, May 2011, pp. 1–5.
 [31] A. Stavridis, S. Sinanovic, M. Di Renzo, and H. Haas, “Transmit precoding for receive spatial modulation using imperfect channel knowledge,” in Proc. IEEE Veh. Technol. Conf. (VTC Spring), Yokohama, Japan, May 2012, pp. 1–5.
 [32] R. Zhang, L. Yang, and L. Hanzo, “Generalised precoding aided spatial modulation,” IEEE Transactions on Wireless Communications, vol. 12, no. 11, pp. 5434–5443, Nov. 2013.
 [33] ——, “Error probability and capacity analysis of generalised precoding aided spatial modulation,” IEEE Transactions on Wireless Communications, vol. 14, no. 1, pp. 364–375, Jan. 2015.
 [34] J. Jeganathan, A. Ghrayeb, and L. Szczecinski, “Generalized space shift keying modulation for MIMO channels,” in Proc. IEEE Int. Symp. Personal, Indoor, Mobile Radio Commun. (PIMRC), Cannes, France, Sept. 2008, pp. 1–5.
 [35] E. Basar, M. Wen, R. Mesleh, M. D. Renzo, Y. Xiao, and H. Haas, “Index modulation techniques for nextgeneration wireless networks,” IEEE Access, vol. 5, pp. 16 693–16 746, 2017.
 [36] B. Zheng, M. Wen, F. Chen, N. Huang, F. Ji, and H. Yu, “The Kbest sphere decoding for soft detection of generalized spatial modulation,” IEEE Trans. Commun., vol. 65, no. 11, pp. 4803–4816, Nov. 2017.
 [37] P. Yang, Y. Xiao, Y. Yu, and S. Li, “Adaptive spatial modulation for wireless MIMO transmission systems,” IEEE Commun. Lett., vol. 15, no. 6, pp. 602–604, Jun. 2011.
 [38] P. Yang, Y. Xiao, L. Li, Q. Tang, Y. Yu, and S. Li, “Link adaptation for spatial modulation with limited feedback,” IEEE Trans. Veh. Technol., vol. 61, no. 8, pp. 3808–3813, Oct. 2012.
 [39] P. Yang, Y. Xiao, Y. Yu, L. Li, Q. Tang, and S. Li, “Simplified adaptive spatial modulation for limitedfeedback MIMO systems,” IEEE Trans. Veh. Technol., vol. 62, no. 6, pp. 2656–2666, Jul. 2013.
 [40] P. Yang, Y. Xiao, B. Zhang, S. Li, M. ElHajjar, and L. Hanzo, “Power allocationaided spatial modulation for limitedfeedback MIMO systems,” IEEE Trans. Veh. Technol., vol. 64, no. 5, pp. 2198–2204, May 2015.
 [41] Y. Fu, C. Wang, A. Ghazal, e. M. Aggoune, and M. M. Alwakeel, “Performance investigation of spatial modulation systems under nonstationary wideband highspeed train channel models,” IEEE Trans. Wireless Commun., vol. 15, no. 9, pp. 6163–6174, Sept. 2016.
 [42] M. D. Renzo and H. Haas, “Space shift keying (SSK—) MIMO over correlated rician fading channels: Performance analysis and a new method for transmitdiversity,” IEEE Trans. Commun., vol. 59, no. 1, pp. 116–129, Jan. 2011.
 [43] R. Mesleh, M. D. Renzo, H. Haas, and P. M. Grant, “Trellis coded spatial modulation,” IEEE Trans. Wireless Commun., vol. 9, no. 7, pp. 2349–2361, Jul. 2010.
 [44] E. Basar, U. Aygolu, E. Panayirci, and H. V. Poor, “New trellis code design for spatial modulation,” IEEE Trans. Wireless Commun., vol. 10, no. 8, pp. 2670–2680, Aug. 2011.
 [45] ——, “Spacetime block coded spatial modulation,” IEEE Trans. Commun., vol. 59, no. 3, pp. 823–832, Mar. 2011.
 [46] M. Le, V. Ngo, H. Mai, X. N. Tran, and M. D. Renzo, “Spatially modulated orthogonal spacetime block codes with nonvanishing determinants,” IEEE Trans. Commun., vol. 62, no. 1, pp. 85–99, Jan. 2014.
 [47] X. Li and L. Wang, “High rate spacetime block coded spatial modulation with cyclic structure,” IEEE Commun. Lett., vol. 18, no. 4, pp. 532–535, Apr. 2014.
 [48] C. Jeon and J. W. Lee, “Multistrata spacetime coded spatial modulation,” IEEE Commun. Lett., vol. 19, no. 11, pp. 1945–1948, Nov. 2015.
 [49] M. Di Renzo and H. Haas, “On transmit diversity for spatial modulation MIMO: Impact of spatial constellation diagram and shaping filters at the transmitter,” IEEE Trans. Veh. Technol., vol. 62, no. 6, pp. 2507–2531, Jul. 2013.
 [50] C. Yu, S. Hsieh, H. Liang, C. Lu, W. Chung, S. Kuo, and S. Pei, “Compressed sensing detector design for space shift keying in MIMO systems,” IEEE Commun. Lett., vol. 16, no. 10, pp. 1556–1559, Oct. 2012.
 [51] W. Liu, N. Wang, M. Jin, and H. Xu, “Denoising detection for the generalized spatial modulation system using sparse property,” IEEE Commun. Lett., vol. 18, no. 1, pp. 22–25, Jan. 2014.
 [52] L. Xiao, P. Yang, Y. Xiao, S. Fan, M. D. Renzo, W. Xiang, and S. Li, “Efficient compressive sensing detectors for generalized spatial modulation systems,” IEEE Trans. Veh. Technol., vol. 66, no. 2, pp. 1284–1298, Feb. 2017.
 [53] C. Wang, P. Cheng, Z. Chen, J. A. Zhang, Y. Xiao, and L. Gui, “NearML lowcomplexity detection for generalized spatial modulation,” IEEE Commun. Lett., vol. 20, no. 3, pp. 618–621, Mar. 2016.
 [54] S. Hur, T. Kim, D. J. Love, J. V. Krogmeier, T. A. Thomas, and A. Ghosh, “Millimeter wave beamforming for wireless backhaul and access in small cell networks,” IEEE Trans. Commun., vol. 61, no. 10, pp. 4391–4403, Oct. 2013.
 [55] X. zhu, Z. Wang, and J. Cao, “NOMAbased spatial modulation,” IEEE Access, vol. 5, pp. 3790–3800, 2017.
 [56] P. Yang, Y. Xiao, M. Xiao, and Z. Ma, “NOMA aided precoded spatial modulation for downlink MIMO transmissions,” IEEE J. Sel. Areas Commun., to be published, 2019.
 [57] Q. Li, M. Wen, E. Basar, H. V. Poor, and F. Chen, “Spatial modulationaided cooperative NOMA: Performance analysis and comparative study,” IEEE J. Sel. Areas Commun., to be published, 2019.
 [58] C. Zhong, X. Hu, X. Chen, D. W. K. Ng, and Z. Zhang, “Spatial modulation assisted multiantenna nonorthogonal multiple access,” IEEE Wireless Commun., vol. 25, no. 2, pp. 61–67, Apr. 2018.
 [59] S. Sinanovic, N. Serafimovski, M. D. Renzo, and H. Haas, “Secrecy capacity of space keying with two antennas,” in Proc. IEEE Veh. Technol. Conf. (VTC Fall), Quebec City, QC, Canada, Sept. 2012, pp. 1–5.
 [60] S. R. Aghdam, T. M. Duman, and M. D. Renzo, “On secrecy rate analysis of spatial modulation and space shift keying,” in Proc. IEEE Int. Black Sea Conf. Commun. Netw. (BlackSeaCom), Constanta, Romania, May 2015, pp. 63–67.
 [61] X. Guan, Y. Cai, and W. Yang, “On the secrecy mutual information of spatial modulation with finite alphabet,” in Proc. IEEE Int. Conf. Wireless Commun. Signal Process. (WCSP), Huangshan, China, Oct. 2012, pp. 1–4.
 [62] T. Fath and H. Haas, “Performance comparison of MIMO techniques for optical wireless communications in indoor environments,” IEEE Trans. Commun., vol. 61, no. 2, pp. 733–742, Feb. 2013.
 [63] N. Ishikawa, S. Sugiura, and L. Hanzo, “50 years of permutation, spatial and index modulation: From classic RF to visible light communications and data storage,” IEEE Commun. Surveys Tuts., vol. 20, no. 3, pp. 1905–1938, Third quarter 2018.
 [64] S. Sugiura, T. Ishihara, and M. Nakao, “Stateoftheart design of index modulation in the space, time, and frequency domains: Benefits and fundamental limitations,” IEEE Access, vol. 5, pp. 21 774–21 790, 2017.
 [65] K. Ntontin, M. Di Renzo, A. I. PerezNeira, and C. Verikoukis, “A lowcomplexity method for antenna selection in spatial modulation systems,” IEEE Commun. Lett., vol. 17, no. 12, pp. 2312–2315, Dec. 2013.
 [66] J. Jeganathan, A. Ghrayeb, and L. Szczecinski, “Spatial modulation: Optimal detection and performance analysis,” IEEE Commun. Lett., vol. 12, no. 8, pp. 545–547, Aug. 2008.
 [67] N. R. Naidoo, H. J. Xu, and T. A. Quazi, “Spatial modulation: Optimal detector asymptotic performance and multiplestage detection,” IET Commun., vol. 5, no. 10, pp. 1368–1376, Jul. 2011.
 [68] S. Sugiura, C. Xu, S. X. Ng, and L. Hanzo, “Reducedcomplexity coherent versus noncoherent QAMaided spacetime shift keying,” IEEE Trans. Commun., vol. 59, no. 11, pp. 3090–3101, Nov. 2011.
 [69] Q. Tang, Y. Xiao, P. Yang, Q. Yu, and S. Li, “A new lowcomplexity nearML detection algorithm for spatial modulation,” IEEE Wireless Commun. Lett., vol. 2, no. 1, pp. 90–93, Feb. 2013.
 [70] J. Zheng, “Signal vector based list detection for spatial modulation,” IEEE Wireless Commun. Lett., vol. 1, no. 4, pp. 265–267, Aug. 2012.
 [71] C. Li, Y. Huang, M. D. Renzo, J. Wang, and Y. Cheng, “Lowcomplexity ML detection for spatial modulation MIMO with APSK constellation,” IEEE Trans. Veh. Technol., vol. 64, no. 9, pp. 4315–4321, Sept. 2015.
 [72] M. Maleki, H. R. Bahrami, and A. Alizadeh, “On MRCbased detection of spatial modulation,” IEEE Trans. Wireless Commun., vol. 15, no. 4, pp. 3019–3029, Apr. 2016.
 [73] A. GarciaRodriguez and C. Masouros, “Lowcomplexity compressive sensing detection for spatial modulation in largescale multiple access channels,” IEEE Trans. Commun., vol. 63, no. 7, pp. 2565–2579, Jul. 2015.
 [74] K. Ishibashi and S. Sugiura, “Effects of antenna switching on bandlimited spatial modulation,” IEEE Wireless Commun. Lett., vol. 3, no. 4, pp. 345–348, Aug. 2014.
 [75] M. Arisaka and S. Sugiura, “Energyversusbandwidthefficiency tradeoff in spatially modulated massive MIMO downlink,” IEEE Wireless Commun. Lett., vol. 8, no. 1, pp. 197–200, Feb. 2019.
 [76] M. D. Renzo and H. Haas, “Performance comparison of different spatial modulation schemes in correlated fading channels,” in Proc. IEEE Int. Conf. Commun. (ICC), Cape Town, South Africa, May 2010, pp. 1–6.
 [77] H. Chang, Y. Liu, and Y. T. Su, “Detection of spatialmodulated signals in the presence of CSI error and timespatial correlation,” in Proc. IEEE Globecom Workshops (GC Wkshps), Atlanta, GA, USA, Dec. 2013, pp. 82–86.
 [78] R. Rajashekar, K. V. S. Hari, and L. Hanzo, “Reducedcomplexity ML detection and capacityoptimized training for spatial modulation systems,” IEEE Trans. Commun., vol. 62, no. 1, pp. 112–125, Jan. 2014.
 [79] S. Sugiura and L. Hanzo, “Effects of channel estimation on spatial modulation,” IEEE Signal Process. Lett., vol. 19, no. 12, pp. 805–808, Dec. 2012.
 [80] R. Mesleh and S. S. Ikki, “On the effect of Gaussian imperfect channel estimations on the performance of space modulation techniques,” in Proc. IEEE Veh. Technol. Conf. (VTC Spring), Yokohama, Japan, May 2012, pp. 1–5.
 [81] X. Wu, H. Claussen, M. Di Renzo, and H. Haas, “Channel estimation for spatial modulation,” IEEE Trans. Commun., vol. 62, no. 12, pp. 4362–4372, Dec. 2014.
 [82] M. Di Renzo and H. Haas, “Bit error probability of SMMIMO over generalized fading channels,” IEEE Trans. Veh. Technol., vol. 61, no. 3, pp. 1124–1144, Mar. 2012.

[83]
——, “Bit error probability of space modulation over Nakagamim fading: Asymptotic analysis,”
IEEE Commun. Lett., vol. 15, no. 10, pp. 1026–1028, Oct. 2011.  [84] P. Yang, Y. L. Guan, Y. Xiao, M. Di Renzo, S. Li, and L. Hanzo, “Transmit precoded spatial modulation: Maximizing the minimum euclidean distance versus minimizing the bit error ratio,” IEEE Trans. Wireless Commun., vol. 15, no. 3, pp. 2054–2068, Mar. 2016.
 [85] A. Younis, W. Thompson, M. Di Renzo, C.X. Wang, M. A. Beach, H. Haas, and P. M. Grant, “Performance of spatial modulation using measured realworld channels,” in Proc. IEEE Veh. Technol. Conf. (VTC Fall), Las Vegas, NV, USA, Sept. 2013, pp. 1–5.
 [86] S. Narayanan, M. Di Renzo, F. Graziosi, and H. Haas, “Distributed spatial modulation: A cooperative diversity protocol for halfduplex relayaided wireless networks,” IEEE Trans. Veh. Technol., vol. 65, no. 5, pp. 2947–2964, May 2016.
 [87] J. Zhang, Q. Li, K. J. Kim, Y. Wang, X. Ge, and J. Zhang, “On the performance of fullduplex twoway relay channels with spatial modulation,” IEEE Trans. Commun., vol. 64, no. 12, pp. 4966–4982, Dec. 2016.
 [88] Y. Yang, “Spatial modulation exploited in nonreciprocal twoway relay channels: Efficient protocols and capacity analysis,” IEEE Trans. Commun., vol. 64, no. 7, pp. 2821–2834, Jul. 2016.
 [89] G. Xiang, G. Li, Y. Xu, Y. Ou, J. Wang, and H. Zhu, “A lowcomplexity soft output detection algorithm for spatial modulation systems,” in Proc. IEEE Int. Conf. Commun. Technol. (ICCT), Chengdu, China, Oct. 2017, pp. 33–37.
 [90] C. Li, Y. Cheng, Y. Zhang, and Y. Huang, “Lowcomplexity softoutput detectors for LDPC coded spatial modulation systems,” in Proc. IEEE Int. Conf. Wireless Commun. Signal Process. (WCSP), Nanjing, China, Oct. 2015, pp. 1–6.
 [91] N. Serafimovski, M. Di Renzo, S. Sinanovic, R. Y. Mesleh, and H. Haas, “Fractional bit encoded spatial modulation (FBESM),” IEEE Commun. Lett., vol. 14, no. 5, pp. 429–431, May 2010.
 [92] B. Jiao, M. Wen, M. Ma, and H. V. Poor, “Spatial modulated full duplex,” IEEE Wireless Commun. Lett., vol. 3, no. 6, pp. 641–644, Dec. 2014.
 [93] P. Raviteja, Y. Hong, and E. Viterbo, “Spatial modulation in fullduplex relaying,” IEEE Commun. Lett., vol. 20, no. 10, pp. 2111–2114, Oct. 2016.
 [94] Y. Yang and S. Aissa, “Bitpadding information guided channel hopping,” IEEE Commun. Lett., vol. 15, no. 2, pp. 163–165, Feb. 2011.
 [95] Yuli Yang and S. Aissa, “Information guided channel hopping with an arbitrary number of transmit antennas,” IEEE Commun. Lett., vol. 16, no. 10, pp. 1552–1555, Oct. 2012.
 [96] L. Xiao, Y. Xiao, Y. Zhao, P. Yang, M. Di Renzo, S. Li, and W. Xiang, “Timedomain turbo equalization for singlecarrier generalized spatial modulation,” IEEE Trans. Wireless Commun., vol. 16, no. 9, pp. 5702–5716, Sept. 2017.
 [97] P. Patcharamaneepakorn, S. Wu, C.X. Wang, e. M. Aggoune, M. M. Alwakeel, X. Ge, and M. Di Renzo, “Spectral, energy, and economic efficiency of 5G multicell massive MIMO systems with generalized spatial modulation,” IEEE Trans. Veh. Technol., vol. 65, no. 12, pp. 9715–9731, Dec. 2016.
 [98] E. Basar, U. Aygolu, E. Panayirci, and H. V. Poor, “Orthogonal frequency division multiplexing with index modulation,” IEEE Trans. Signal Process., vol. 61, no. 22, pp. 5536–5549, Nov. 2013.
 [99] D. E. Knuth, The Art of Computer Programming, Volume 4, Fascicle 3: Generating All Combinations and Partitions. AddisonWesley Professional, 2005.
 [100] T. Lakshmi Narasimhan and A. Chockalingam, “On the capacity and performance of generalized spatial modulation,” IEEE Commun. Lett., vol. 20, no. 2, pp. 252–255, Feb. 2016.
 [101] Y. Xiao, Z. Yang, L. Dan, P. Yang, L. Yin, and W. Xiang, “Lowcomplexity signal detection for generalized spatial modulation,” IEEE Commun. Lett., vol. 18, no. 3, pp. 403–406, Mar. 2014.
 [102] C. E. Chen, C. H. Li, and Y. H. Huang, “An improved orderedblock MMSE detector for generalized spatial modulation,” IEEE Commun. Lett., vol. 19, no. 5, pp. 707–710, May 2015.
 [103] C. T. Lin, W. R. Wu, and C. Y. Liu, “Lowcomplexity ML detectors for generalized spatial modulation systems,” IEEE Trans. Commun., vol. 63, no. 11, pp. 4214–4230, Nov. 2015.
 [104] L. He, J. Wang, W. Ding, and J. Song, “Sparse Bayesian learning based symbol detection for generalised spatial modulation in largescale MIMO systems,” in Proc. IEEE GLOBECOM, San Diego, USA, Dec. 2015, pp. 1–6.
 [105] J. A. CalBraz and R. SampaioNeto, “Lowcomplexity sphere decoding detector for generalized spatial modulation systems,” IEEE Commun. Lett., vol. 18, no. 6, pp. 949–952, Jun. 2014.
 [106] A. Younis, S. Sinanovic, M. D. Renzo, R. Mesleh, and H. Haas, “Generalised sphere decoding for spatial modulation,” IEEE Trans. Commun., vol. 61, no. 7, pp. 2805–2815, Jul. 2013.
 [107] A. Younis, M. D. Renzo, R. Mesleh, and H. Haas, “Sphere decoding for spatial modulation,” in Proc. IEEE Int. Conf. Commun. (ICC), Kyoto, Japan, Jun. 2011, pp. 1–6.
 [108] L. Xiao, P. Yang, Y. Xiao, J. Liu, S. Fan, B. Dong, and S. Li, “An improved softinput softoutput detector for generalized spatial modulation,” IEEE Signal Process. Lett., vol. 23, no. 1, pp. 30–34, Jan. 2016.
 [109] S. Sugiura, S. Chen, and L. Hanzo, “Coherent and differential spacetime shift keying: A dispersion matrix approach,” IEEE Trans. Commun., vol. 58, no. 11, pp. 3219–3230, Nov. 2010.
 [110] P. A. Martin, “Differential spatial modulation for APSK in timevarying fading channels,” IEEE Commun. Lett., vol. 19, no. 7, pp. 1261–1264, Jul. 2015.
 [111] J. Liu, L. Dan, P. Yang, L. Xiao, F. Yu, and Y. Xiao, “Highrate APSKaided differential spatial modulation: Design method and performance analysis,” IEEE Commun. Lett., vol. 21, no. 1, pp. 168–171, Jan. 2017.
 [112] M. Wen, Z. Ding, X. Cheng, Y. Bian, H. V. Poor, and B. Jiao, “Performance analysis of differential spatial modulation with two transmit antennas,” IEEE Commun. Lett., vol. 18, no. 3, pp. 475–478, Mar. 2014.
 [113] N. Ishikawa and S. Sugiura, “Rectangular differential spatial modulation for openloop noncoherent massiveMIMO downlink,” IEEE Trans. Wireless Commun., vol. 16, no. 3, pp. 1908–1920, Mar. 2017.
 [114] C. Wu, Y. Xiao, L. Xiao, P. Yang, and X. Lei, “Spacetime block coded rectangular differential spatial modulation,” in IEEE Int. Conf. Commun. (ICC), Kansas City, MO, USA, May 2018, pp. 1–6.
 [115] C. Wu, Y. Xiao, L. Xiao, P. Yang, X. Lei, and W. Xiang, “Spacetime block coded rectangular differential spatial modulation: System design and performance analysis,” IEEE Trans. Commun., to be published, 2019.
 [116] L. Xiao, P. Xiao, Y. Xiao, C. Wu, D. Mi, and I. A. Hemadeh, “Rectangular differential ofdm with index modulation,” in Proc. IEEE Veh. Technol. Conf. (VTC Spring), Kuala Lumpur, Malaysia, May 2019, pp. 1–5.
 [117] N. Ishikawa, R. Rajashekar, C. Xu, S. Sugiura, and L. Hanzo, “Differential spacetime coding dispensing with channel estimation approaches the performance of its coherent counterpart in the openloop massive mimoofdm downlink,” IEEE Trans. Commun., vol. 66, no. 12, pp. 6190–6204, Dec. 2018.
 [118] M. Zhang, M. Wen, X. Cheng, and L. Yang, “A dualhop virtual MIMO architecture based on hybrid differential spatial modulation,” IEEE Trans. Wireless Commun., vol. 15, no. 9, pp. 6356–6370, Sept. 2016.
 [119] J. Li, M. Wen, X. Cheng, Y. Yan, S. Song, and M. H. Lee, “Differential spatial modulation with gray coded antenna activation order,” IEEE Commun. Lett., vol. 20, no. 6, pp. 1100–1103, Jun. 2016.
 [120] W. Zhang, Q. Yin, and H. Deng, “Differential full diversity spatial modulation and its performance analysis with two transmit antennas,” IEEE Commun. Lett., vol. 19, no. 4, pp. 677–680, Apr. 2015.
 [121] R. Rajashekar, N. Ishikawa, S. Sugiura, K. V. S. Hari, and L. Hanzo, “Fulldiversity dispersion matrices from algebraic field extensions for differential spatial modulation,” IEEE Trans. Veh. Technol., vol. 66, no. 1, pp. 385–394, Jan. 2017.
 [122] N. Ishikawa and S. Sugiura, “Single and multiplerf aided noncoherent generalized spatial modulation,” in Proc. IEEE Veh. Technol. Conf. (VTC Spring), Seoul, South Korea, May 2014, pp. 1–5.
 [123] A. G. Helmy, M. Di Renzo, and N. AlDhahir, “Differential spatially modulated spacetime block codes with temporal permutations,” IEEE Trans. Veh. Technol., vol. 66, no. 8, pp. 7548–7552, Aug. 2017.
 [124] R. Zhang, L. Yang, and L. Hanzo, “Performance analysis of nonlinear generalized precoding aided spatial modulation,” IEEE Trans. Wireless Commun., vol. 15, no. 10, pp. 6731–6741, Oct. 2016.
 [125] C. Liu, L. Yang, and W. Wang, “Transmitterprecodingaided spatial modulation achieving both transmit and receive diversity,” IEEE Trans. Veh. Technol., vol. 67, no. 2, pp. 1375–1388, Feb 2018.
 [126] M. Maleki, K. MohamedPour, and M. Soltanalian, “Receive spatial modulation in correlated massive mimo with partial CSI,” IEEE Trans. Signal Process., vol. 67, no. 5, pp. 1237–1250, Mar. 2019.
 [127] Y. Cao and T. Ohtsuki, “Orthogonality structure designs for generalized precoding aided spatial modulation,” IEEE Wireless Commun. Lett., to be published, 2019.
 [128] A. Mokh, M. Crussiére, M. Hélard, and M. Di Renzo, “Theoretical performance of coherent and incoherent detection for zeroforcing receive antenna shift keying,” IEEE Access, vol. 6, pp. 39 907–39 916, 2018.
 [129] Y. Li, X. Lei, W. Tang, D. He, Y. Xiao, and W. Xiang, “Performance analysis of MMSE precoding aided spatial modulation,” IEEE Access, vol. 6, pp. 44 835–44 845, 2018.
 [130] P. Yang, J. Zhu, Z. Liu, Y. Xiao, S. Li, and W. Xiang, “Unified power allocation for receive spatial modulation based on approximate optimization,” IEEE Access, vol. 6, pp. 49 450–49 459, 2018.
 [131] P. Yang, Y. Xiao, M. Xiao, J. Zhu, S. Li, and W. Xiang, “Enhanced receive spatial modulation based on power allocation,” IEEE J. Sel. Areas Commun., to be published, 2019.
 [132] Y. Li, X. Lei, Y. Xiao, P. Yang, C. Zheng, and W. Xiang, “Power allocation for precodingaided spatial modulation,” IEEE Commun. Lett., vol. 22, no. 5, pp. 1094–1097, May 2018.
 [133] A. Stavridis, M. Di Renzo, P. M. Grant, and H. Haas, “Performance analysis of receive space modulation in the shadowing MIMO broadcast channel,” IEEE Trans. Commun., vol. 65, no. 5, pp. 1972–1983, May 2017.
 [134] ——, “On the asymptotic performance of receive space modulation in the shadowing broadcast channel,” IEEE Commun. Lett., vol. 20, no. 10, pp. 2103–2106, Oct. 2016.
 [135] A. Stavridis, M. D. Renzo, and H. Haas, “Performance analysis of multistream receive spatial modulation in the mimo broadcast channel,” IEEE Trans. Wireless Commun., vol. 15, no. 3, pp. 1808–1820, Mar. 2016.
 [136] J. Li, M. Wen, X. Cheng, Y. Yan, S. Song, and M. H. Lee, “Generalized precodingaided quadrature spatial modulation,” IEEE Trans. Veh. Technol., vol. 66, no. 2, pp. 1881–1886, Feb. 2017.
 [137] Y. Chen, L. Wang, Z. Zhao, M. Ma, and B. Jiao, “Secure multiuser MIMO downlink transmission via precodingaided spatial modulation,” IEEE Commun. Lett., vol. 20, no. 6, pp. 1116–1119, Jun. 2016.
 [138] F. Wu, L. Yang, W. Wang, and Z. Kong, “Secret precodingaided spatial modulation,” IEEE Commun. Lett., vol. 19, no. 9, pp. 1544–1547, Sept. 2015.
 [139] F. Wu, R. Zhang, L. Yang, and W. Wang, “Transmitter precodingaided spatial modulation for secrecy communications,” IEEE Trans. Veh. Technol., vol. 65, no. 1, pp. 467–471, Jan. 2016.
 [140] C. Cheng, M. Di Renzo, F. Graziosi, and A. Zappone, “On simultaneous wireless information and power transfer for receive spatial modulation,” IEEE Access, vol. 5, pp. 23 204–23 211, 2017.
 [141] Z. Bouida, A. Ghrayeb, and K. A. Qaraqe, “Adaptive spatial modulation for spectrum sharing systems with limited feedback,” IEEE Trans. Commun., vol. 63, no. 6, pp. 2001–2014, Jun. 2015.
 [142] W. Wang and W. Zhang, “Huffman codingbased adaptive spatial modulation,” IEEE Trans. Wireless Commun., vol. 16, no. 8, pp. 5090–5101, Aug. 2017.
 [143] M. Maleki and H. R. Bahrami, “Adaptive space modulation with partial CSIT in spatially correlated fading channels,” IEEE Trans. Veh. Technol., vol. 63, no. 7, pp. 3184–3195, Sept. 2014.
 [144] P. Cheng, Z. Chen, J. A. Zhang, Y. Li, and B. Vucetic, “A unified precoding scheme for generalized spatial modulation,” IEEE Trans. Commun., vol. 66, no. 6, pp. 2502–2514, Jun. 2018.
 [145] S. Luo, X. T. Tran, K. C. Teh, and K. H. Li, “Adaptive spatial modulation for uplink mmWave communication systems,” IEEE Commun. Lett., vol. 21, no. 10, pp. 2178–2181, Oct. 2017.
 [146] M. Koca and H. Sari, “Precoding for spatial modulation against correlated fading channels,” IEEE Trans. Wireless Commun., vol. 17, no. 9, pp. 5857–5870, Sept. 2018.
 [147] A. G. Helmy, M. Di Renzo, and N. AlDhahir, “Enhancedreliability cyclic generalized spatialandtemporal modulation,” IEEE Commun. Lett., vol. 20, no. 12, pp. 2374–2377, Dec. 2016.
 [148] L. Xiao, Y. Xiao, P. Yang, J. Liu, S. Li, and W. Xiang, “Spacetime block coded differential spatial modulation,” IEEE Trans. Veh. Technol., vol. 66, no. 10, pp. 8821–8834, Oct. 2017.
 [149] L. Wang, Z. Chen, Z. Gong, and M. Wu, “Diversityachieving quadrature spatial modulation,” IEEE Trans. Veh. Technol., vol. 66, no. 12, pp. 10 764–10 775, Dec. 2017.
 [150] P. Chen, Z. Cao, Z. Chen, and X. Wang, “Offgrid DOA estimation using sparse bayesian learning in MIMO radar with unknown mutual coupling,” IEEE Trans. Signal Process., vol. 67, no. 1, pp. 208–220, Jan. 2019.
 [151] Z. Gao, L. Dai, Z. Wang, S. Chen, and L. Hanzo, “Compressivesensingbased multiuser detector for the largescale SMMIMO uplink,” IEEE Trans. Veh. Technol., vol. 65, no. 10, pp. 8725–8730, Oct. 2016.
 [152] T. Wang, S. Liu, F. Yang, J. Wang, J. Song, and Z. Han, “Generalized spatial modulationbased multiuser and signal detection scheme for terrestrial return channel with NOMA,” IEEE Trans. Broadcast., vol. 64, no. 2, pp. 211–219, Jun. 2018.
 [153] L. Xiao, Y. Xiao, C. Xu, X. Lei, P. Yang, S. Li, and L. Hanzo, “Compressedsensing assisted spatial multiplexing aided spatial modulation,” IEEE Trans. Wireless Commun., vol. 17, no. 2, pp. 794–807, Feb. 2018.
 [154] L. Xiao, P. Xiao, Y. Xiao, I. Hemadeh, A. Mohamed, and L. Hanzo, “Bayesian compressive sensing assisted space–time block coded quadrature spatial modulation,” IEEE Trans. Veh. Technol., vol. 67, no. 10, pp. 10 044–10 048, Oct. 2018.
 [155] L. Dai, B. Wang, Y. Yuan, S. Han, C.L. I, and Z. Wang, “Nonorthogonal multiple access for 5G: solutions, challenges, opportunities, and future research trends,” IEEE Commun. Mag., vol. 53, no. 9, pp. 74–81, Sept. 2015.
 [156] Z. Ding, Y. Liu, J. Choi, Q. Sun, M. Elkashlan, C. L. I, and H. V. Poor, “Application of nonorthogonal multiple access in LTE and 5G networks,” IEEE Commun. Mag., vol. 55, no. 2, pp. 185–191, Feb. 2017.
 [157] N. Serafimovski, S. Sinanović, M. Di Renzo, and H. Haas, “Multiple access spatial modulation,” EURASIP J. Wireless Commun. Netw., vol. 2012, no. 1, p. 299, 2012.
 [158] T. Wang, S. Liu, F. Yang, J. Wang, J. Song, and Z. Han, “Blocksparse compressive sensing based multiuser and signal detection for generalized spatial modulation in NOMA,” in IEEE Int. Wireless Commun. Mobile Comput. Conf. (IWCMC), Valencia, Spain, Jun. 2017, pp. 1992–1997.
 [159] X. Wang, J. Wang, L. He, and J. Song, “Spectral efficiency analysis for downlink NOMA aided spatial modulation with finite alphabet inputs,” IEEE Trans. Veh. Technol., vol. 66, no. 11, pp. 10 562–10 566, Nov. 2017.
 [160] X. Wang, J. Wang, L. He, Z. Tang, and J. Song, “On the achievable spectral efficiency of spatial modulation aided downlink nonorthogonal multiple access,” IEEE Commun. Lett., vol. 21, no. 9, pp. 1937–1940, Sept. 2017.
 [161] J. W. Kim, M. Irfan, S. M. AL, and S. Y. Shin, “Selective nonorthogonal multiple access (NOMA) and spatial modulation (SM) for improved spectral efficiency,” in IEEE Int. Symp. Intelligent Signal Process. Commun. Syst. (ISPACS), Nusa Dua, Indonesia, Nov. 2015, pp. 552–555.
 [162] R. F. Siregar, F. W. Murti, and S. Y. Shin, “Combination of spatial modulation and nonorthogonal multiple access using hybrid detection scheme,” in IEEE Int. Conf. Ubiquitous and Future Netw. (ICUFN), Milan, Italy, Jul. 2017, pp. 476–481.
 [163] Y. Chen, L. Wang, Y. Ai, B. Jiao, and L. Hanzo, “Performance analysis of NOMASM in vehicletovehicle massive MIMO channels,” IEEE J. Sel. Areas Commun., vol. 35, no. 12, pp. 2653–2666, Dec. 2017.
 [164] ——, “NOMASM for cooperatively enhancing vehicletovehicle transmissions,” in IEEE Globecom Workshops (GC Wkshps), Singapore, Dec 2017, pp. 1–6.
 [165] Y. Liu, L. Yang, and L. Hanzo, “Spatial modulation aided sparse codedivision multiple access,” IEEE Trans. Wireless Commun., vol. 17, no. 3, pp. 1474–1487, Mar. 2018.
 [166] Y. Zheng, H. Duan, and C. Wang, “Learning the truth privately and confidently: Encrypted confidenceaware truth discovery in mobile crowdsensing,” IEEE Trans. Inf. Forensics Security, vol. 13, no. 10, pp. 2475–2489, Oct. 2018.
 [167] L. Wang, S. Bashar, Y. Wei, and R. Li, “Secrecy enhancement analysis against unknown eavesdropping in spatial modulation,” IEEE Commun. Lett., vol. 19, no. 8, pp. 1351–1354, Aug. 2015.
 [168] F. Wu, W. Wang, C. Dong, and L. Yang, “Performance analysis of secret precodingaided spatial modulation with finitealphabet signaling,” IEEE Access, vol. 6, pp. 29 366–29 381, Jun. 2018.
 [169] Y. Huang, B. Zheng, M. Wen, F. Ji, F. Chen, and H. Yu, “Improving physical layer security via random precoding,” in Proc. IEEE Globecom Workshops (GC Wkshps), Singapore, Singapore, Dec. 2017, pp. 1–6.
 [170] F. Shu, Z. Wang, R. Chen, Y. Wu, and J. Wang, “Two highperformance schemes of transmit antenna selection for secure spatial modulation,” IEEE Trans. Veh. Technol., vol. 67, no. 9, pp. 8969–8973, Sept. 2018.
 [171] C. Liu, L. Yang, and W. Wang, “Secure spatial modulation with a fullduplex receiver,” IEEE Wireless Commun. Lett., vol. 6, no. 6, pp. 838–841, Dec. 2017.
 [172] Z. Huang, Z. Gao, and L. Sun, “Antieavesdropping scheme based on quadrature spatial modulation,” IEEE Commun. Lett., vol. 21, no. 3, pp. 532–535, Mar. 2017.
 [173] H. Taha and E. Alsusa, “Secret key exchange and authentication via randomized spatial modulation and phase shifting,” IEEE Trans. Veh. Technol., vol. 67, no. 3, pp. 2165–2177, Mar. 2018.
 [174] X. Wang, X. Wang, and L. Sun, “Spatial modulation aided physical layer security enhancement for fading wiretap channels,” in Proc. IEEE Int. Conf. Wireless Commun. Signal Process. (WCSP), Yangzhou, China, Oct. 2016, pp. 1–5.
 [175] X. Jiang, M. Wen, H. Hai, J. Li, and S. Kim, “Secrecyenhancing scheme for spatial modulation,” IEEE Commun. Lett., vol. 22, no. 3, pp. 550–553, Mar. 2018.
 [176] Y. Yang and M. Guizani, “Mappingvaried spatial modulation for physical layer security: Transmission strategy and secrecy rate,” IEEE J. Sel. Areas Commun., vol. 36, no. 4, pp. 877–889, Apr. 2018.
 [177] Z. Gao, Z. Huang, and L. Sun, “Secure cooperative communication with nondedicated relays,” IEEE Trans. Veh. Technol., vol. 67, no. 8, pp. 7811–7815, Aug. 2018.
 [178] Z. Bouida, A. Stavridis, A. Ghrayeb, H. Haas, and M. Hasna, “Precoded spatial modulation for the wiretap channel with relay selection and cooperative jamming,” in Proc. IEEE Wireless Commun. Netw. Conf. (WCNC), San Francisco, CA, USA, Mar. 2017, pp. 1–6.
 [179] S. Rangan, T. S. Rappaport, and E. Erkip, “Millimeterwave cellular wireless networks: Potentials and challenges,” Proc. IEEE, vol. 102, no. 3, pp. 366–385, Mar. 2014.
 [180] W. Wu, N. Zhang, N. Cheng, Y. Tang, K. Aldubaikhy, and X. Shen, “Beef up mmWave dense cellular networks with D2Dassisted cooperative edge caching,” IEEE Trans. Veh. Technol., to be published, 2019.
 [181] N. Ishikawa, R. Rajashekar, S. Sugiura, and L. Hanzo, “Generalizedspatialmodulationbased reducedRFchain millimeterwave communications,” IEEE Trans. Veh. Technol., vol. 66, no. 1, pp. 879–883, Jan. 2017.
 [182] P. Liu and A. Springer, “Space shift keying for LOS communication at mmWave frequencies,” IEEE Wireless Commun. Lett., vol. 4, no. 2, pp. 121–124, Apr. 2015.
 [183] P. Liu, M. D. Renzo, and A. Springer, “Lineofsight spatial modulation for indoor mmWave communication at 60 GHz,” IEEE Trans. Wireless Commun., vol. 15, no. 11, pp. 7373–7389, Nov. 2016.
 [184] P. Liu, J. Blumenstein, N. S. Perović, M. D. Renzo, and A. Springer, “Performance of generalized spatial modulation MIMO over measured 60GHz indoor channels,” IEEE Trans. Commun., vol. 66, no. 1, pp. 133–148, Jan. 2018.
 [185] A. Younis, N. Abuzgaia, R. Mesleh, and H. Haas, “Quadrature spatial modulation for 5G outdoor millimeter–wave communications: Capacity analysis,” IEEE Trans. Wireless Commun., vol. 16, no. 5, pp. 2882–2890, May 2017.
 [186] L. He, J. Wang, and J. Song, “Spectralefficient analog precoding for generalized spatial modulation aided mmWave MIMO,” IEEE Trans. Veh. Technol., vol. 66, no. 10, pp. 9598–9602, Oct. 2017.
 [187] ——, “On generalized spatial modulation aided millimeter wave MIMO: Spectral efficiency analysis and hybrid precoder design,” IEEE Trans. Wireless Commun., vol. 16, no. 11, pp. 7658–7671, Nov. 2017.
 [188] ——, “Spatial modulation for more spatial multiplexing: RFchainlimited generalized spatial modulation aided mmWave MIMO with hybrid precoding,” IEEE Trans. Commun., vol. 66, no. 3, pp. 986–998, Mar. 2018.
 [189] P. Liu, M. D. Renzo, and A. Springer, “Variable generalized spatial modulation for indoor LOS mmwave communication: Performance optimization and novel switching structure,” IEEE Trans. Commun., vol. 65, no. 6, pp. 2625–2640, Jun. 2017.
 [190] X. Ge, R. Zi, X. Xiong, Q. Li, and L. Wang, “Millimeter wave communications with OAMSM scheme for future mobile networks,” IEEE J. Sel. Areas Commun., vol. 35, no. 9, pp. 2163–2177, Sept. 2017.
 [191] P. Yang, Y. Xiao, Y. L. Guan, Z. Liu, S. Li, and W. Xiang, “Adaptive SMMIMO for mmWave communications with reduced RF chains,” IEEE J. Sel. Areas Commun., vol. 35, no. 7, pp. 1472–1485, Jul. 2017.
 [192] M. Lee and W. Chung, “Transmitter design for analog beamforming aided spatial modulation in millimeter wave MIMO systems,” in Proc. IEEE Int. Symp. Personal, Indoor, Mobile Radio Commun. (PIMRC), Valencia, Spain, Sept. 2016, pp. 1–6.
 [193] ——, “Adaptive multimode hybrid precoding for singleRF virtual space modulation with analog phase shift network in MIMO systems,” IEEE Trans. Wireless Commun., vol. 16, no. 4, pp. 2139–2152, Apr. 2017.
 [194] N. S. Perović, P. Liu, M. D. Renzo, and A. Springer, “Receive spatial modulation for LOS mmwave communications based on TX beamforming,” IEEE Commun. Lett., vol. 21, no. 4, pp. 921–924, Apr. 2017.
 [195] A. Raafat, A. Agustin, and J. Vidal, “Receive spatial modulation for massive MIMO systems,” in Proc. IEEE Globecom, Singapore, Singapore, Dec. 2017, pp. 1–6.
 [196] J. Zhu, P. Yang, Y. Xiao, X. Lei, and Q. Chen, “Low RFcomplexity receive spatial modulation for millimeterwave MIMO communications,” IEEE Commun. Lett., vol. 22, no. 7, pp. 1338–1341, Jul. 2018.
 [197] A. Sevincer, A. Bhattarai, M. Bilgi, M. Yuksel, and N. Pala, “LIGHTNETs: Smart LIGHTing and mobile optical wireless NETworks — a survey,” IEEE Commun. Surveys Tuts., vol. 15, no. 4, pp. 1620–1641, Fourth Quarter 2013.
 [198] R. Mesleh, R. Mehmood, H. Elgala, and H. Haas, “Indoor MIMO optical wireless communication using spatial modulation,” in Proc. IEEE Int. Conf. Commun. (ICC), Cape Town, South Africa, May 2010, pp. 1–5.
 [199] T. Özbilgin and M. Koca, “Optical spatial modulation over atmospheric turbulence channels,” J. Lightw. Technol., vol. 33, no. 11, pp. 2313–2323, Jun. 2015.
 [200] K. P. Peppas and P. T. Mathiopoulos, “Freespace optical communication with spatial modulation and coherent detection over HK atmospheric turbulence channels,” J. Lightw. Technol., vol. 33, no. 20, pp. 4221–4232, Oct. 2015.
 [201] A. Jaiswal, M. Abaza, M. R. Bhatnagar, and V. K. Jain, “An investigation of performance and diversity property of optical space shift keyingbased FSOMIMO system,” IEEE Trans. Commun., vol. 66, no. 9, pp. 4028–4042, Sept. 2018.
 [202] H. G. Olanrewaju, J. Thompson, and W. O. Popoola, “Performance of optical spatial modulation in indoor multipath channel,” IEEE Trans. Wireless Commun., vol. 17, no. 9, pp. 6042–6052, Sept. 2018.
 [203] J. Wang, J. Wang, Y. Wu, M. Lin, and M. Chen, “Constellation optimization for spatial modulation based indoor optical wireless communications,” in Proc. IEEE Globecom Workshops (GC Wkshps), Singapore, Singapore, Dec. 2017, pp. 1–6.
 [204] N. Ishikawa and S. Sugiura, “Maximizing constrained capacity of powerimbalanced optical wireless MIMO communications using spatial modulation,” J. Lightw. Technol., vol. 33, no. 2, pp. 519–527, Jan. 2015.
 [205] T. Fath, H. Haas, M. D. Renzo, and R. Mesleh, “Spatial modulation applied to optical wireless communications in indoor LOS environments,” in Proc. IEEE Globecom, Kathmandu, Nepal, Dec. 2011, pp. 1–5.
 [206] J. Wang, Z. Yang, Y. Wang, and M. Chen, “On the performance of spatial modulationbased optical wireless communications,” IEEE Photon. Technol. Lett., vol. 28, no. 19, pp. 2094–2097, Oct. 2016.
 [207] R. Mesleh, H. Elgala, R. Mehmood, and H. Haas, “Performance of optical spatial modulation with transmittersreceivers alignment,” IEEE Commun. Lett., vol. 15, no. 1, pp. 79–81, Jan. 2011.
 [208] H. G. Olanrewaju and W. O. Popoola, “Effect of synchronization error on optical spatial modulation,” IEEE Trans. Commun., vol. 65, no. 12, pp. 5362–5374, Dec. 2017.
 [209] D. Karunatilaka, F. Zafar, V. Kalavally, and R. Parthiban, “LED based indoor visible light communications: State of the art,” IEEE Commun. Surveys Tuts., vol. 17, no. 3, pp. 1649–1678, Third Quarter 2015.
 [210] W. O. Popoola, “Impact of VLC on light emission quality of white LEDs,” J. Lightw. Technol., vol. 34, no. 10, pp. 2526–2532, May 2016.
 [211] C. R. Kumar and R. K. Jeyachitra, “Power efficient generalized spatial modulation MIMO for indoor visible light communications,” IEEE Photon. Technol. Lett., vol. 29, no. 11, pp. 921–924, Jun. 2017.
 [212] ——, “Dualmode generalized spatial modulation MIMO for visible light communications,” IEEE Commun. Lett., vol. 22, no. 2, pp. 280–283, Feb. 2018.
 [213] ——, “Improved joint generalized spatial modulations for MIMOVLC systems,” IEEE Commun. Lett., vol. 22, no. 11, pp. 2226–2229, Nov. 2018.
 [214] W. O. Popoola, E. Poves, and H. Haas, “Error performance of generalised space shift keying for indoor visible light communications,” IEEE Trans. Commun., vol. 61, no. 5, pp. 1968–1976, May 2013.
 [215] W. O. Popoola and H. Haas, “Demonstration of the merit and limitation of generalised space shift keying for indoor visible light communications,” J. Lightw. Technol., vol. 32, no. 10, pp. 1960–1965, May 2014.
 [216] S. P. Alaka, T. L. Narasimhan, and A. Chockalingam, “Generalized spatial modulation in indoor wireless visible light communication,” in Proc. IEEE Globecom, San Diego, CA, USA, Dec. 2015, pp. 1–7.
 [217] K. Xu, H. Yu, and Y. Zhu, “Channeladapted spatial modulation for massive MIMO visible light communications,” IEEE Photon. Technol. Lett., vol. 28, no. 23, pp. 2693–2696, Dec. 2016.
 [218] T. Wang, F. Yang, L. Cheng, and J. Song, “Spectralefficient generalized spatial modulation based hybrid dimming scheme with LACOOFDM in VLC,” IEEE Access, vol. 6, pp. 41 153–41 162, 2018.
 [219] J. Wang, J. Zhu, S. Lin, and J. Wang, “Adaptive spatial modulation based visible light communications: SER analysis and optimization,” IEEE Photon. J., vol. 10, no. 3, pp. 1–14, Jun. 2018.
 [220] A. Yesilkaya, E. Basar, F. Miramirkhani, E. Panayirci, M. Uysal, and H. Haas, “Optical MIMOOFDM with generalized LED index modulation,” IEEE Trans. Commun., vol. 65, no. 8, pp. 3429–3441, Aug. 2017.
 [221] E. Başar, E. Panayirci, M. Uysal, and H. Haas, “Generalized led index modulation optical OFDM for MIMO visible light communications systems,” in IEEE Int. Conf. Commun. (ICC), Kuala Lumpur, Malaysia, May 2016, pp. 1–5.
 [222] N. Serafimovski, A. Younis, R. Mesleh, P. Chambers, M. Di Renzo, C. Wang, P. M. Grant, M. A. Beach, and H. Haas, “Practical implementation of spatial modulation,” IEEE Trans. Veh. Technol., vol. 62, no. 9, pp. 4511–4523, Nov. 2013.
 [223] S. Wang, Y. Li, and J. Wang, “Multiuser detection in massive spatial modulation MIMO with lowresolution ADCs,” IEEE Trans. Wireless Commun., vol. 14, no. 4, pp. 2156–2168, Apr. 2015.
 [224] E. Soujeri and G. Kaddoum, “The impact of antenna switching time on spatial modulation,” IEEE Wireless Commun. Lett., vol. 5, no. 3, pp. 256–259, Jun. 2016.
 [225] A. Afana, N. A. Ali, and S. Ikki, “Performance of cognitive spatial modulation MIMO systems under transceiver hardware impairments,” in Proc. IEEE Int. Symp. Personal, Indoor, Mobile Radio Commun. (PIMRC), Montreal, QC, Canada, Oct. 2017, pp. 1–5.
 [226] A. Afana, N. AbuAli, and S. Ikki, “On the joint impact of hardware and channel imperfections on cognitive spatial modulation MIMO systems: Cramer–rao bound approach,” IEEE Syst. J., pp. 1–12, 2018.
 [227] R. Mesleh, O. Hiari, A. Younis, and S. Alouneh, “Transmitter design and hardware considerations for different space modulation techniques,” IEEE Trans. Wireless Commun., vol. 16, no. 11, pp. 7512–7522, Nov. 2017.
 [228] A. G. Helmy, M. Di Renzo, and N. AlDhahir, “On the robustness of spatial modulation to I/Q imbalance,” IEEE Commun. Lett., vol. 21, no. 7, pp. 1485–1488, Jul. 2017.
 [229] G. Lee and T. Kim, “Implementation of a nearoptimal detector for spatial modulation MIMO systems,” IEEE Trans. Circuits Syst. II, Exp. Briefs, vol. 63, no. 10, pp. 954–958, Oct. 2016.
 [230] T. Liu, Y. Ye, C. Huang, C. Chen, C. Chen, Y. Hwang, and Y. Chu, “A lowcomplexity maximum likelihood detector for the spatially modulated signals: Algorithm and hardware implementation,” IEEE Trans. Circuits Syst. II, Exp. Briefs, to be published, 2019.
 [231] S. Guo, H. Zhang, Y. Wang, and D. Yuan, “Spatial modulated simultaneous wireless information and power transfer,” in Proc. IEEE GLOBECOM, Washington, DC, USA, Dec. 2016, pp. 1–6.
 [232] A. Koc, I. Altunbas, and E. Basar, “Twoway fullduplex spatial modulation systems with wireless powered AF relaying,” IEEE Wireless Commun. Lett., vol. 7, no. 3, pp. 444–447, Jun. 2018.
 [233] R. Zhang, L. Yang, and L. Hanzo, “Energy pattern aided simultaneous wireless information and power transfer,” IEEE J. Sel. Areas Commun., vol. 33, no. 8, pp. 1492–1504, Aug. 2015.
 [234] M. Zhang and X. Cheng, “Spatialmodulationbased wirelesspowered communication for achievable rate enhancement,” IEEE Commun. Lett., vol. 21, no. 6, pp. 1365–1368, Jun. 2017.
 [235] W. Qu, X. Cheng, C. Chen, C. Chen, and L. Yang, “Fullduplex SWIPT relaying based on spatialmodulation,” IEEE Trans. Green Commun. Netw., to be published, 2019.
 [236] S. Narayanan, M. ShikhBahaei, J. Hou, and M. F. Flanagan, “Wirelesspowered distributed spatial modulation with energy recycling and finiteenergy storage,” IEEE Trans. Wireless Commun., vol. 17, no. 10, pp. 6645–6662, Oct. 2018.
 [237] R. Mesleh, H. Elgala, and H. Haas, “Optical spatial modulation,” IEEE/OSA J. Opt. Commun. Netw., vol. 3, no. 3, pp. 234–244, Mar. 2011.
 [238] Z. Bouida, H. ElSallabi, A. Ghrayeb, and K. A. Qaraqe, “Reconfigurable antennabased spaceshift keying (SSK) for MIMO Rician channels,” IEEE Trans. Wireless Commun., vol. 15, no. 1, pp. 446–457, Jan. 2016.
 [239] E. Basar and I. Altunbas, “Spacetime channel modulation,” IEEE Trans. Veh. Technol., vol. 66, no. 8, pp. 7609–7614, Aug. 2017.
 [240] R. Abualhiga and H. Haas, “Subcarrierindex modulation OFDM,” in Proc. IEEE Int. Symp. Personal, Indoor, Mobile Radio Commun. (PIMRC), Tokyo, Japan, Sept. 2009, pp. 177–181.
 [241] B. Zheng, F. Chen, M. Wen, F. Ji, H. Yu, and Y. Liu, “Lowcomplexity ML detector and performance analysis for OFDM with inphase/quadrature index modulation,” IEEE Commun. Lett., vol. 19, no. 11, pp. 1893–1896, Nov. 2015.
 [242] E. Ozturk, E. Basar, and H. A. Cirpan, “Generalized frequency division multiplexing with index modulation,” in Proc. IEEE Globecom Workshops (GC Wkshps), Washington, DC, USA, Dec. 2016, pp. 1–6.
 [243] E. Başar, “OFDM with index modulation using coordinate interleaving,” IEEE Wireless Commun. Lett., vol. 4, no. 4, pp. 381–384, Aug. 2015.
 [244] ——, “Multipleinput multipleoutput OFDM with index modulation,” IEEE Signal Process. Lett., vol. 22, no. 12, pp. 2259–2263, Dec. 2015.
 [245] E. Basar, “On multipleinput multipleoutput OFDM with index modulation for next generation wireless networks,” IEEE Trans. Signal Process., vol. 64, no. 15, pp. 3868–3878, Aug. 2016.
 [246] B. Zheng, M. Wen, E. Basar, and F. Chen, “Multipleinput multipleoutput OFDM with index modulation: Lowcomplexity detector design,” IEEE Trans. Signal Process., vol. 65, no. 11, pp. 2758–2772, Jun. 2017.
 [247] ——, “Lowcomplexity nearoptimal detector for multipleinput multipleoutput OFDM with index modulation,” in Proc. IEEE Int. Conf. Commun. (ICC), Paris, France, May 2017, pp. 1–6.
 [248] M. Nakao, T. Ishihara, and S. Sugiura, “Singlecarrier frequencydomain equalization with index modulation,” IEEE Commun. Lett., vol. 21, no. 2, pp. 298–301, Feb. 2017.
 [249] G. Kaddoum, M. F. A. Ahmed, and Y. Nijsure, “Code index modulation: A high data rate and energy efficient communication system,” IEEE Commun. Lett., vol. 19, no. 2, pp. 175–178, Feb. 2015.
 [250] G. Kaddoum, Y. Nijsure, and H. Tran, “Generalized code index modulation technique for highdatarate communication systems,” IEEE Trans. Veh. Technol., vol. 65, no. 9, pp. 7000–7009, Sept. 2016.
 [251] Q. Li, M. Wen, E. Basar, and F. Chen, “Index modulated OFDM spread spectrum,” IEEE Trans. Wireless Commun., vol. 17, no. 4, pp. 2360–2374, Apr. 2018.
 [252] C. Cheng, H. Sari, S. Sezginer, and Y. T. Su, “Enhanced spatial modulation with multiple signal constellations,” IEEE Trans. Commun., vol. 63, no. 6, pp. 2237–2248, Jun. 2015.
 [253] T. Mao, Z. Wang, Q. Wang, S. Chen, and L. Hanzo, “Dualmode index modulation aided OFDM,” IEEE Access, vol. 5, pp. 50–60, 2017.
 [254] M. Wen, E. Basar, Q. Li, B. Zheng, and M. Zhang, “Multiplemode orthogonal frequency division multiplexing with index modulation,” IEEE Trans. Commun., vol. 65, no. 9, pp. 3892–3906, Sept. 2017.
 [255] J. Zhang, Y. Wang, J. Zhang, and L. Ding, “Polarization shift keying (PolarSK): System scheme and performance analysis,” IEEE Trans. Veh. Technol., vol. 66, no. 11, pp. 10 139–10 155, Nov. 2017.
 [256] P. Henarejos and A. I. PérezNeira, “3d polarized modulation: System analysis and performance,” IEEE Trans. Commun., vol. 66, no. 11, pp. 5305–5316, Nov. 2018.
 [257] J. Hoseyni and J. Ilow, “Beam angle channel modulation,” in Proc. IEEE Veh. Technol. Conf. (VTC Fall), Toronto, ON, Canada, Sept. 2017, pp. 1–6.
 [258] T. Datta, H. S. Eshwaraiah, and A. Chockalingam, “Generalized spaceandfrequency index modulation,” IEEE Trans. Veh. Technol., vol. 65, no. 7, pp. 4911–4924, Jul. 2016.
 [259] H. A. Ngo, C. Xu, S. Sugiura, and L. Hanzo, “Spacetimefrequency shift keying for dispersive channels,” IEEE Signal Process. Lett., vol. 18, no. 3, pp. 177–180, Mar. 2011.
 [260] M. I. Kadir, “Generalized spacetimefrequency index modulation,” IEEE Commun. Lett., to be published, 2018.
 [261] E. Basar, “Index modulation techniques for 5G wireless networks,” IEEE Commun. Mag., vol. 54, no. 7, pp. 168–175, Jul. 2016.
 [262] X. Cheng, M. Zhang, M. Wen, and L. Yang, “Index modulation for 5G: Striving to do more with less,” IEEE Wireless Commun., vol. 25, no. 2, pp. 126–132, Apr. 2018.
 [263] M. Wen, X. Cheng, and L. Yang, Index Modulation for 5G Wireless Communications. Wireless Networks Series, Springer, 2017.
 [264] P. Yang, Y. Xiao, Y. L. Guan, M. Di Renzo, S. Li, and L. Hanzo, “Multidomain index modulation for vehicular and railway communications: A survey of novel techniques,” IEEE Veh. Technol. Mag., vol. 13, no. 3, pp. 124–134, Sept. 2018.
 [265] M. Di Renzo, “Spatial modulation based on reconfigurable antennas—a new air interface for the IoT,” in IEEE Military Commun. Conf. (MILCOM), Baltimore, MD, USA, Oct. 2017, pp. 495–500.
 [266] D.T. PhanHuy, Y. Kokar, K. Rachedi, P. Pajusco, A. Mokh, T. Magounaki, R. Masood, C. Buey, P. Ratajczak, N. MalhourouxGaffet, J.M. Conrat, J.C. Prévotet, A. Ourir, J. De Rosny, M. Crussière, M. Hélard, A. Gati, T. Sarrebourse, and M. Di Renzo, “Singlecarrier spatial modulation for the internet of things: Design and performance evaluation by using real compact and reconfigurable antennas,” IEEE Access, vol. 7, pp. 18 978–18 993, 2019.
 [267] D. N. Viet, M. Di Renzo, V. Basavarajappa, B. B. Exposito, J. Basterrechea, and D.T. PhanHuy, “Spatial modulation based on reconfigurable antennas: Performance evaluation by using the prototype of a reconfigurable antenna,” arXiv preprint arXiv:1901.01752, 2019.
 [268] D. PhanHuy, Y. Kokar, J. Rioult, N. MalhourouxGaffet, J. Prevotet, C. Buey, P. Ratajczak, M. Helard, K. Rachedi, A. Ourir, C. Leray, J. de Rosny, A. Le Cornec, A. Gati, T. Sarrebourse, P. Pajusco, and M. Di Renzo, “First visual demonstration of transmit and receive spatial modulations using the “radio wave display”,” in Int. ITG Workshop on Smart Antennas, Berlin, Germany, Mar. 2017, pp. 1–7.
 [269] R. Fan, Y. J. Yu, and Y. L. Guan, “Generalization of orthogonal frequency division multiplexing with index modulation,” IEEE Trans. Wireless Commun., vol. 14, no. 10, pp. 5350–5359, Oct. 2015.
Comments
There are no comments yet.