应用神经网络粒子群算法的多用户检测

为了减少最优多有户检测器的计算复杂度,提出了一种融合粒子群优化算法和神经网络的神经网络粒子群优化算法,并设计了一种解决CDMA通信系统的多用户检测问题的新方法。该方法是把神经网络嵌入到粒子群优化算法的每一代中以改进算法性能。通过混合神经网络到PSO中,还可以加快PSO的收敛速度,减少计算复杂度。仿真结果证明了所设计的检测器无论抗多址干扰能力和抗远近效应能力都优于应用Hopfield神经网络、遗传算法和粒子群算法的多用户检测器。

Multi-User Detection Based on Particle Swarm Optimization Algorithm with Neural Network

To reduce computational complexity of the optimal multi-user detector, a novel hybrid algorithm that employs particle swarm optimization algorithm (PSO) and Hopfield neural network is presented. Then we design a novel multi-user detector in code-division multiple-access (CDMA) communication systems. Using this approach, the Hopfield neural network is embedded into the PSO to improve further the performance of the population at each generation. Such a hybridization of the PSO with the Hopfield neural network reduces its computational complexity by providing faster convergence. Simulation results are provided to show that the proposed detector has significant performance improvements over the detectors based on Hopfield neural network, genetic algorithm, and particle swarm optimization in terms of multiple access interference and near-far resistance.