Research On Channel Data Processing Of Channel Simulator Based On Multi-core DSP

The channel simulator can simulate the propagation characteristics of the channel in the real environment,so it is widely used in various occasions such as communication technology research and communication devices testing.In the channel simulator,the DSP chip is responsible for processing the channel parameters generated by the channel model in real time and transmitting the calculation results to the FPGA chip for convolution operations.With the development of mobile communication technologies such as Massive MIMO technology,the number of sub-channels supported by the channel model is increasing,and the amount of data of channel parameters generated by the channel model is also increased.This puts forward higher requirements for the DSP channel data processing algorithms in the DSP chip.Based on the multi-core DSP chip of the channel simulator,this paper implements the existing single-core DSP channel data processing algorithm in the DSP chip.Through the verification and analysis of the DSP channel data processing algorithm,the performance of the DSP channel data processing algorithm is improved from two aspects of accuracy and operating efficiency.In order to further improve the operation efficiency of the DSP channel data processing algorithm,the design and performance verification of the multi-core parallel channel data processing algorithm were completed based on the multi-core parallel architecture model supported by the multi-core DSP chip.The main contents of paper are as followings:1?The existing channel data processing algorithm is implemented and verified in the DSP chip.The existing channel data processing algorithm is implemented in the DSP chip according to the algorithm implementation process of the DSP software development environment,and then the correctness of the implementation algorithm is verified by comparing the running results of the implementation algorithm and the simulation algorithm.The memory usage of the DSP channel data processing algorithm is tested,and the result shows that the memory usage of the algorithm is low,which reserves sufficient space for further optimization of the algorithm in the future.Finally,the error of the implementation algorithm is tested,and the results show that the overall error of the algorithm is large,and the accuracy of the algorithm needs to be optimized.2?The DSP channel data processing algorithm is optimized.First,by optimizing the error steps in the DSP channel data processing algorithm,the accuracy of the DSP channel data processing.algorithm is improved.Then,by using the built-in instructions of the DSP to optimize the implementation code of the DSP channel data processing algorithm,the operating efficiency of the DSP channel data processing algorithm is improved.3?Completed the design and verification of the multi-core parallel channel data processing algorithm.In order to further improve the operating efficiency of the DSP channel data processing algorithm,two multi-core parallel design schemes based on different multi-core parallel architecture models are proposed.And by comparing the performance of the two schemes,the final multi-core parallel channel data processing algorithm is determined.