DeepAI AI Chat
Log In Sign Up

Joint Design of Convolutional Code and CRC under Serial List Viterbi Decoding

by   Hengjie Yang, et al.

This paper studies the joint design of optimal convolutional codes (CCs) and CRC codes when serial list Viterbi algorithm (S-LVA) is employed in order to achieve the target frame error rate (FER). We first analyze the S-LVA performance with respect to SNR and list size, repsectively, and prove the convergence of the expected number of decoding attempts when SNR goes to the extreme. We then propose the coded channel capacity as the criterion to jointly design optimal CC-CRC pair and optimal list size and show that the optimal list size of S-LVA is always the cardinality of all possible CCs. With the maximum list size, we choose the design metric of optimal CC-CRC pair as the SNR gap to random coding union (RCU) bound and the optimal CC-CRC pair is the one that achieves a target SNR gap with the least complexity. Finally, we show that a weaker CC with a strong optimal CRC code could be as powerful as a strong CC with no CRC code.


page 1

page 2

page 3

page 4


CRC-Aided List Decoding of Convolutional Codes in the Short Blocklength Regime

This paper identifies convolutional codes (CCs) used in conjunction with...

List Decodability of Symbol-Pair Codes

We investigate the list decodability of symbol-pair codes in the present...

High-Rate Convolutional Codes with CRC-Aided List Decoding for Short Blocklengths

Recently, rate-1/ω zero-terminated and tail-biting convolutional codes (...

Optimal list decoding from noisy entropy inequality

A noisy entropy inequality for boolean functions by Samorodnitsky is app...

CRC-Aided Short Convolutional Codes and RCU Bounds for Orthogonal Signaling

We extend earlier work on the design of convolutional code-specific CRC ...

Design of Bilayer and Multi-layer LDPC Ensembles from Individual Degree Distributions

A new approach for designing bilayer and multi-layer LDPC codes is propo...

List and Certificate Complexities in Replicable Learning

We investigate replicable learning algorithms. Ideally, we would like to...