Improved Uplink Channel Detection Method And Application In Large Scale MIMO System

With the spectrum resources become increasingly tense, to satisfy the need of high-speed data transfer rates, the transmitter and receiver with multiple-input multiple-output(MIMO, Multiple Input Multiple Output) technology which can get very high band utilization has been widely used in wireless communication systems, and become very promising technology in the future mobile communication. The introduction of MIMO technology in wireless communication system is a great challenge to the complex signal processing at receiver, especially the design and implementation of MIMO signal detector, which requires a detector with low complexity, low power and hardware overhead to achieve high-speed high-quality signal detection. Signal detection for MIMO systems can be Sphere decoder, V-BLAST algorithm, mesh reduction algorithms, maximum likelihood detection methods. Although the maximum likelihood detection algorithm has the optimal detection performance, its high complexity makes it difficult to apply in reality. Therefore, how to reduce detection complexity and improve detection performance has been a hot issue in the research scholars. Detection algorithm for large MIMO systems are studied in this paper.This paper designs low complexity algorithms based on Markov chain Monte Carlo(MCMC) technique for signal detection and channel estimation on the uplink in large scale multiuser multiple input multiple output(MIMO) systems with tens to hundreds of antennas at the base station(BS) and similar number of uplink users. A BS receiver that employs a randomized sampling method(which makes a probabilistic choice between Gibbs sampling and random sampling in each iteration) for detection and a Gibbs sampling based method for channel estimation is designed. The algorithm designed for detection alleviates the stalling problem encountered at high SNRs in conventional MCMC algorithm and achieves near-optimal performance in large systems. A novel ingredient in the detection algorithm that is responsible for achieving near-optimal performance at low complexities is the use of a randomized MCMC(R-MCMC) strategy. Near-optimal detection performance is demonstrated for large number of BS antennas and users(e.g., 64, 128, 256 BS antennas/users). The MCMC based channel estimation algorithm refines an initial estimate of the channel obtained during pilot phase through iterations with R-MCMC detection during data phase. In time division duplex(TDD) systems where channel reciprocity holds, these channel estimates can be used for multiuser MIMO precoding on the downlink. The receiver designed in the paper achieves performance which is near optimal and close to that with perfect channel knowledge.The main contents of this paper are summarized as follows:1. The channel characteristics and capacity of MIMO is analysed. Meanwhile, detection algorithms ML, ZF, MMSE and their complexity is analysed..2. This paper designs R-MCMC algorithm and MCMC-based channel estimation algorithm. When the antenna number is up to hundreds, the algorithm can get close to the performance of Sphere decoder. At the same time, the simulation results are given to compare performance.3. The paper analyses the application of R-MCMC detection/estimation algorithm in the high-speed railway channel model. According to the simulation result, R-MCMC algorithm not only has low complexity, but also can achieve better performance.