Data detection method based on simulated annealing neural network and interference elimination

Abstract

The invention discloses a data detection method based on a simulated annealing neural network and interference elimination. The method comprises the following steps: acquiring spread spectrum data; modulating the spread spectrum data; judging whether 10 end users have been selected; generating mixed data; demodulating the mixed data; generating training sample data; constructing the simulated annealing neural network; training the simulated annealing neural network; judging whether the number of training iterations is less than 1000; acquiring simulated annealing neural network optimization data; acquiring interference elimination data; setting an initial value; calculating the total number of wrong elements; judging whether the current number of iterations is equal to the total number ofinterference elimination data elements; and detecting a data bit error rate. The data detection method based on the simulated annealing neural network and the interference elimination provided by theinvention can be used for detecting data sent by a user from the mixed data on a same frequency band in a Gaussian channel; and compared with the prior art, the detected data sent by the user have thelowest bit error rate.

Description

technical field [0001] The invention belongs to the technical field of communication, and further relates to a data detection method based on simulated annealing neural network and interference elimination in the technical field of wireless communication. The invention can be used to detect the data sent by the user from the mixed data on the same frequency band in the Gaussian channel with the lowest error rate. Background technique [0002] The data sent by the user is processed by Multicarrier direct sequence code division multiple access technology and then transmitted on the same frequency band of the Gaussian channel in wireless communication. The data sent by the user is coded and modulated with a spreading code, and the receiving end According to the orthogonality of the spreading code, the user's data detection is completed. Multi-carrier direct sequence code division multiple access MC-DS-CDMA technology combines code division multiple access CDMA (code division m...

AI Technical Summary

Problems solved by technology

However, the disadvantage of this method is that because the iterative process of the quantum neural network adopts the gradient descent algorithm, the obtained weights and thresholds are not optimal, which affects the classification and detection performance of the neural network and leads to The difference between the data output by the network and the data sent by the user is too large, and the bit error rate of the detected user data is high

The disadvantage of this method is that it only considers the influence of the incomplete orthogonality of the spreading code, and does not take into account the further destruction of the orthogonality of the spreading code due to OFDM modulation. The influence of complete orthogonality has not been completely eliminated, and the detected data is quite different from the data sent by the user.

Images