IRS-Assisted Ambient Backscatter Communications Utilizing Deep Reinforcement Learning
share
We consider an ambient backscatter communication (AmBC) system aided by an intelligent reflecting surface (IRS). The optimization of the IRS to assist AmBC is extremely difficult when there is no prior channel knowledge, for which no design solutions are currently available. We utilize a deep reinforcement learning-based framework to jointly optimize the IRS and reader beamforming, with no knowledge of the channels or ambient signal. We show that the proposed framework can facilitate efficient AmBC communication with a detection performance comparable to several benchmarks under full channel knowledge.
READ FULL TEXT