Research On Joint Detection And Decoding For Uplink System Of Iot Communications

The next generation of Intern et of Things(IoT)has the characteristics of high connection density,low deployment cost,delay and error sensitivity.It puts forward the technical requirements of high frequency spectrum efficiency,low processing energy consumption,low delay and high reliability for develoment of IoT.The polar-coded sparse code multiple access(PC-SCMA)effectively improve the system capacity,and has the superior performance of short code,for which PC-SCMA is regarded as a candidate for IoT communications in 6G systems.However,it processes receiving signals at the cost of high complexity,while the performance is unsatisfactory in overload scenarios.Therefore,the thesis focuses on the high performance and low complexity detection and decoding technology of uplink PC-SCMA system to improve the practicability of PC-SCMA for short packet communication of IoT.Based on the concept of joint detection and decoding,the thesis designs a sparse factor graph-based expanded iterative detection and decoding(S-EIDD)algorithm.The algorithm iteratively updates the reliability information on the scale expanded factor graph composed of the SCMA factor graph and the sparse factor graph of the Polar code to complete the detection and decoding.Considering that the information update efficiency of the two local factor graphs of S-EIDD is not consistent,the thesis proposes a method to design the iterative parameters of the joint structure using the Extrinsic Information Transfer Chart(EXIT chart),and determines the iterative parameters of the S-EIDD receiver.Simulation results show that S-EIDD receiver outperforms the traditional separate detection-decoding receiver of 2dB at BER=0.01 with the code length of 16,the code rate of 1/2,and the overload rate of 150%.Moreover,the computational complexity is less than 50%.To enhance reliability and reduce processing consumption,the thesis introduces Cyclic Redundancy Check(CRC)to the PC-SCMA system,and presents an upgraded version of the S-EIDD algorithm,called Improved SEIDD(I-S-EIDD).On the one hand,I-S-EIDD algorithm evaluates the reliability of information during the iteration process,suppresses the propagation of unreliable information and strenthens the propagation of reliable information by means of post-processing,thereby accelerating iteration convergence and improving decoding performance.On the other hand,I-S-EIDD algorithm prunes the factor graph during the iteration process,deleting redundant calculations and decreasing the parallel computing scale of the receiver,so the computational complexity of the IS-EIDD algorithm in a single iteration is dynamically reduced.The simulation results show that in the uplink PC-SCMA system with the code length of 16,the code rate of 1/2,and the overload rate of 150%,the I-SEIDD receiver outperforms S-EIDD of about 0.5 dB at BER=0.01,and its computational complexity is reduced by about 15%.