Toward Optimally Efficient Search With Deep Learning for Large-Scale MIMO Systems

This paper investigates the optimal signal detection problem with a particular interest in large-scale multiple-input multiple-output (MIMO) systems. The is NP-hard and can be solved optimally by searching shortest path on decision tree. Unfortunately, existing search algorithms often involve prohibitively high complexities, which indicates that they are infeasible MIMO To address this issue, we propose general heuristic algorithm, namely, hyper-accelerated tree (HATS) algorithm. proposed algorithm employs deep neural network (DNN) to estimate heuristic, then use estimated speed up underlying memory-bounded idea inspired fact reaches efficiency function. Simulation results show almost bit error rate (BER) performance systems, while memory size bounded. In meanwhile, it visits nearly fewest nodes. practical scenarios, thereby applicable for Besides, code available at \url{https://github.com/skypitcher/hats}.