Improved AWGN channel implementation method and device based on Polar method

Abstract

The invention relates to an improved AWGN channel implementation method and an improved AWGN channel implementation device based on a Polar method. The method comprises the steps of generating two uniformly distributed random sequence sets through combining with an LFSR method and a CASR method; and looking up a pre-stored look-up table according to the random sequences and computing, and thus acquiring Gaussian white noise. The device comprises a random sequence generation module, an index generation module, a clock signal generation module, a delay module and a Gaussian white noise generation module. The random sequences generated according to the improved AWGN channel implementation method and device based on the Polar method provided by the invention have smaller continuous output correlation, the problems of data loss and cross block domain are avoided during the process of computing the look-up table index, and meanwhile the output speed is improved.

Description

technical field [0001] The invention belongs to the technical field of wireless communication, and in particular relates to an improved AWGN channel realization method and device based on the Polar method. Background technique [0002] In a real communication system, the transmission channel of the communication system is not ideal and is always affected by various interferences and noises, among which Gaussian white noise is one of the most harmful interferences. In order to accurately test and analyze the performance of real communication systems, technicians usually simulate the real communication environment under laboratory conditions and artificially generate Gaussian white noise. The existing hardware Gaussian noise generation methods usually include two types: physical noise generation method and digital synthetic noise method. The physical noise generator has high precision, but the implementation circuit is relatively complicated, so the digital noise generation m...

AI Technical Summary

Problems solved by technology

The CLT method requires a large number of parallel random index factors to operate, which seriously reduces the running speed; the two sets of PN sequences of the Box–Muller method can only obtain a set of Gaussian noise output, and to realize complex Gaussian noise, it is necessary to add two additional sets of PN sequences. And to achieve a more accurate output, the bit width of the PN sequence needs to be very high, which will cause unnecessary waste of resources

[0004] The Polar method is an improved version of the Box–Muller method. It mainly improves the utilization rate of the PN sequence. The complex Gaussian noise output can be obtained by using two sets of PN sequences. However, during the implementation process, the intermediate index factors that do not meet the requirements will be discarded. According to the probability calculation, about 21% of the data will be discarded. When the hardware is implemented, this will affect the continuity of the generated white noise. In order to ensure that the final Gaussian noise can be continuously output, it is necessary to operate across clock domains during hardware implementation. Makes high demands on the system clock when generating high rate noise factors

At the same time, in order to synchronize the data of the two clock domains, it is often necessary to add an additional FIFO (First Input First Output, first-in-first-out queue), resulting in a waste of system resources

Images