Joint Semi-blind Equalization And Decoding Of LDPC Coded ISI/MIMO Signals

In digital wireless communication systems,signals at the receiver suffer serious inter-symbol interference(ISI)caused by the multi-path effect.In order to overcome the ISI problem,an equalizer is implemented to correct or compensate the system.Semi-blind equalization algorithm improves the performance significantly with fewer pilot symbols,which has become a hot research topic.Traditional receivers often optimize the different modules separately,ignoring the intrinsic relationship between the parameters of these modules.In order to ensure the correct transmission as much as possible,a joint detection and decoding technology gradually has proposed by combination of channel detection and decoding,which makes full use of the constraint relationship between the various modules,thereby reducing the error rate and improving system performance.Based on a series of research results in the field of communication by linear programming(LP)method,a joint detection and decoding of the system receiver is proposed for ISI / MIMO(Multiple-Input Multiple-Output)system encoded by LDPC(Low Density Parity Check)in this paper.To achieve these research objectives,the decoding algorithm is integrated into different detection algorithms to get a unified linear programming problem.The main research objectives include two aspects.Firstly,a detection algorithm was implemented which approximates the performance of the optimal detection algorithm at higher communication transmission rate.And on this basis,by introducing the channel estimation algorithm we extended the limitations of the original algorithm.Secondly,for the flat fading MIMO system,we use a small number of pilot symbols to design the semi-blind equalization algorithm,further a high performance receiver based on the joint detection and decoding algorithm is achieve by optimize equalization algorithm.The main contributions of this paper include the following aspects:1.After the study of the performance of the maximum likelihood detection,the optimal detection algorithm for QPSK modulation system is derived based on the original detection algorithm.By using the linear relationship between the QPSK symbol and the BPSK symbol,the maximum likelihood detection algorithm is separated from the real and imaginary parts and processed linearly to achieve the optimal detection algorithm under QPSK modulation.At the same time,for this detection algorithm,the limitation is that the transmission channel is known to the receiver,and the application requirement under the unknown channel is extended by introducing the MMSE channel estimation algorithm.2.For the MIMO system with LDPC code,the LP semi-blind equalization algorithm is proposed by designing the effective equalizer structure and using the characteristics of pilot symbols and transmitting data.Furthermore,in order to obtain better detection performance,the decoding constraint of LDPC code is introduced into the detection algorithm,and a joint detection and decoding algorithm is proposed to realize the design of the optimal receiver under the system.