Research On The Low-precision Receiving Technologies For MmWave Communication System

With the rapid development of wireless communication and the emergence of new application scenarios,millimeter-wave(mmWave)communication technology has been developed rapidly to significantly increase the data transmission rate.Moreover,mmWave technology and massive multiple-input multiple-output(MIMO)can be perfectly combined,making the mmWave massive MIMO system an attractive technology for 5G and future wireless communications.However,the potential high power consumption and high cost of mmWave massive MIMO system may prevent its commercialization.Receivers with low-precision ADCs will greatly reduce system cost and power consumption.Therefore,the research on receiving technology for mmWave massive MIMO systems with low-precision quantization is of great significance.This paper focuses on key receiver design challenges,including channel estimation and signal detection.Firstly,this paper studies low-precision channel estimation techniques.With a certain Signal Noise Ratio(SNR)level,the sparsity adaptive orthogonal matching pursuit(SAOMP)algorithm with the abiblity to learn sparsity adaptively can achieve better channel estimation performance than traditional orthogonal matching pursuit(OMP)algorithms.In this paper,the multi-bit expectation maximization(EM)algorithm is deduced and combined with the compressed sensing(CS)algorithm.Compared with traditional EM and CS algorithms,the novel algorithm proposed in this paper can improve the channel estimation performance.In the wideband model,the random Gaussian sequence and shifted Zadoff-Chu or Gray sequence can provide measurement matrix with a good Restricted Isometry Property(RIP)for channel estimation with CS.Then this paper makes deep research on the mmWave low-precision MIMO signal detection technology,and derives the maximum likelihood detection(MLD)formula for low-precision quantization.Applying optimization method and sphere decoding,the complexity of the receiver can be reduced significantly.The suboptimal receiver using the optimization method for multi-bit systems has complexity that is only 0.1%-10% of the MLD.The complexity can be further decreased as the number of bits increases,and its performance can approach the performance of the optimal receiver.The complexity of sphere decoding(SD)decreases as the number of bits decrease.Compared with the optimal receiver,SD can achieve almost the same performance with only 1% of the complexity,and thus achieve a good tradeoff between system complexity and performance.Finally,this paper combines the oversampling technique with the EM algorithm to boost system performance and reduce system complexity,especially in some cases when the number of user equipments and the modulation order are large.we can achieve better performance than the optimal receiver using simple linear detection techniques.As the number of users and the modulation order increase,the increment of complexity is acceptable.