The Research Of Detector For Massive GSSK-MIMO Systems

Massive Multiple Input Multiple Output(MIMO)systems with tens or hundreds of antennas at both transmitter and receiver sides have recently attracted considerable attention due to the several advantages including high date rates,transmit diversity and link reliability,which make it constitutes a promising technology for the design of future wireless communication systems(5G).However too large number of antennas results in increasing the complexity of hardware,energy consumption and the complexity of the detector at both ends.Generalized Space Shift Keying(GSSK)allows several transmit antennas to be active simultaneously during transmission and encodes the information bits into various combinations of multiple active transmit antennas.Since the phase and amplitude of the transmitted symbols do not convey any information,thus GSSK can reduce Inter Channel Interference(ICI),transceiver overhead and dramatically improve the energy efficiency.At the receiver side,the GSSK detection only needs to determine the indices of activated antennas via detection algorithm,which reduce the complexity of detector.GSSK resolve the problem of Inter Channel Interference(ICI)and Inter Antenna Synchronization(IAS),which caused by large number of active RF chains in massive MIMO systems.Therefore,GSSK has important value in the publication and research for massive MIMO systems.This study aims to research the detection algorithm for massive GSSK-MIMO systems,and propose three new algorithms.The main contents and innovations of this thesis are as follows:Firstly,an algorithm based on penalty function is proposed in this thesis.A lemma is proposed to show that the ML detection of GSSK can be converted into an equivalent 0-1 quadratic programming if the penalty factor is greater than a small constant.At last,an algorithm is proposed to determine the penalty factor.Secondly,a partial optimal detector based on the necessary global optimality condition(NGOC)of binary quadratic programming(BQP)is presented in the thesis.First,an optimal decision rule is derived out from the(NGOC)of(BQP).Furthermore,a decision feedback method which can iteratively decide the entries of global optimal solution is proposed to increase the number of entries which can be determined.Taking advantage of these determined entries,the original MIMO detection can be reduced to a smaller scale one for the undetermined entries.Finally,coupled with some efficient conventional detectors,the remained small-scale problem can be solved efficiently.Thirdly,a decision probability ordered successive interference cancellation(SIC)detector is presented in this thesis.Using Lyapunov central limit theorem,an algorithm derived from the global optimal condition of binary quadratic programming is presented to calculate the probability that the inputs can be correctly decided.Then,the SIC detection can be optimally ordered by such decision probability.Thus,the error propagation of SIC can be overcome efficiently.The three proposed detection algorithm are simulated.The simulation results show that the proposed algorithms can achieve efficient performance.So the study proposes a new way for the publication and research for massive GSSK-MIMO systems.