Research On Parallel Implementation Of High-Rate Channel Encoding And Decoding

Millimeter wave communication has become a new research direction of mobile communication because of its higher communication rate.It is applied in 5G mobile communication systems to achieve higher transmission rates.With the research and development of 5G mobile communication systems,high-speed data transmission and ultra-low end-to-end transmission delays place higher requirements on channel coding technology.In modern communication systems,turbo codes are a commonly used channel coding method.By using an iterative decoding algorithm based on soft information exchange between two component decoders,it has strong error correction capabilities,which helps to approach wireless channel capacity transmission throughput for reliable communication.However,due to the complexity of the iterative decoding algorithm of the turbo code,the serial decoding of the log-map decoding algorithm has a limited decoding rate and a large decoding delay.Therefore,improving the turbo decoding algorithm through parallel decoding,increasing the processing rate of turbo decoding,reducing the decoding delay,helping to achieve high-rate channel coding and decoding in high-frequency and low-frequency mixed networking systems,so as to achieve higher communication rate.This paper studies the use of channel encoders and multiple turbo decoders to achieve high-rate channel coding and parallel decoding.A fully parallel turbo code decoding algorithm is studied,which can process all the bits in the decoded information frame in parallel.Each iterative decoding only needs two clock cycles,and the entire decoding process only needs dozens of clock cycles.Therefore,the throughput of turbo decoding can be significantly improved,and the decoding delay can be reduced.After MATLAB simulation,it is proved that when using 36 iterations,the fully parallel turbo decoding algorithm can achieve the error correction decoding performance which is achieved when using the log-map algorithm with 6 iterations.By comparison,it shows that it can achieve a decoding rate much higher than the log-map algorithm under the various frame lengths of the LTE standard turbo code at the cost of higher hardware resource requirements.Based on the fully parallel turbo decoding algorithm,this paper presents an implementation scheme of a fully parallel turbo decoder on FPGA.The overall structure and processing process of the hardware implementation of the decoder,the design of the algorithm block processing unit and the interleaving module are introduced in detail.Through fixed-point simulations,decoder performance tests were performed for different frame lengths of the turbo code in the LTE standard,and it was proved that the fully parallel turbo decoder can be applied to turbo code of various frame lengths.