摘要： 提出一种前馈神经网络盲多用户检测算法，利用前馈神经网络替代原有检测器中的滤波器，通过惩罚函数对约束恒模代价函数进行求解，获得前馈神经网络权值和参数的迭代公式，实现了盲多用户检测。Matlab仿真结果表明，该算法改善了系统的误码率性能，加快了算法的收敛速度。

关键词: 盲多用户检测, 前馈神经网络, 恒模算法, 惩罚函数

Abstract: A blind multi-user detection algorithm based on feed-forward neural network is proposed. A feed-forward neural network replaces the filter of the original detector. A constrained constant modulus algorithm cost function is obtained by a penalty function. Iterative formula of weights and parameters of neural networks are acquired. The blind multi-user detection algorithm is realized. MATLAB simulation indicates that the new algorithm improves system performance such as bit error rate and convergence speed.

Key words: blind multi-user detection, feed-forward neural network, constant module algorithm, penalty function

中图分类号: 

• TP18