Research On APSK Modulation Symbol Shaping And Low-complexity Decoding Algorithm For LDPC

With the continuous evolution of communication technology,how to achieve the improvement of channel capacity and low complexity processing algorithm has always been an important issue in the development of communication technology.In the communication system,the low-density parity check code has become an important part of the receiver system because of its good error correction performance and hardware realizability,and the low-complexity decoding has always been one of the research hotspots;The joint optimization of encoding and modulation technology is now the mainstream direction of signal system research.The existing new encoding and modulation technology cannot further approach the Shannon tolerance without the support of constellation shaping technology.The existing constellation shaping methods mainly have two limitations.Firstly,the new probability shaping techniques mainly focus on ASK and QAM modulation,with little involvement in the optimization and implementation of APSK modulation.Secondly,the encoding and modulation techniques based on APSK are relatively old,complex,and have unexpected performance gains.Therefore,the main innovation points of this article in constellation formation are as follows:firstly,a new constellation probability distribution scheme is proposed,which is different from traditional grouping schemes.Based on the MB distribution,the APSK constellation is probability optimized;Secondly,a new probability amplitude shaping technique based on APSK is proposed,which achieves probability mapping through a constant component distribution matcher and is simulated and verified in AWGN channels.The experimental results show that the shaping technique proposed in this paper can achieve a maximum gain of 0.66dB compared to uniform signal transmission at 2.4bits2-D,which is superior to traditional constellation shaping architectures.In the second part of this article,based on the current LDPC decoding algorithm,a fixed-point analysis is conducted from the perspective of uniform quantization.Firstly,the basic principle of quantization and the iterative process of decoding are introduced,and then the quantization formula and specific implementation of the layered minimum sum algorithm are proposed.Finally,the optimal quantization bit width and correction factor are determined through floating-point and fixed-point simulations,achieving a low complexity decoding algorithm.