作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2007, Vol. 33 ›› Issue (10): 196-198. doi: 10.3969/j.issn.1000-3428.2007.10.070

• 人工智能及识别技术 • 上一篇    下一篇

基于神经网络量子算法的多用户检测器

高洪元,刁 鸣,贾宗圣   

  1. (哈尔滨工程大学信息与通信工程学院,哈尔滨 150001)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-05-20 发布日期:2007-05-20

Multiuser Detector Based on Neural Network Quantum Algorithm

GAO Hongyuan, DIAO Ming, JIA Zongsheng   

  1. (School of Information and Communication Engineering, Harbin Engineering University, Harbin 150001)
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-05-20 Published:2007-05-20

摘要: 利用遗传量子算法和Hopfield神经网络,提出了一种融合两种算法优点的神经网络量子算法,并将其应用到CDMA通信系统的多用户检测问题中。所提算法把神经网络嵌入到遗传量子算法的每一代中,可进一步提高量子种群的适应度函数值。通过混合神经网络到GQA中,还可加快GQA的收敛速度进而减少算法的计算复杂度。另外,GQA所提供的良好初值改善了HNN的性能,嵌入的HNN也提高了GQA的性能。仿真结果证明了该方法的抗多址干扰能力和抗远近效应能力都优于传统检测器和一些应用智能算法的多用户检测器。

关键词: 遗传量子算法, Hopfield神经网络, 多用户检测, CDMA

Abstract: A neural network quantum algorithm (NNQA) that employs a genetic quantum algorithm (GQA) and a Hopfield neural network (HNN) for the multiuser detection problem in code-division multiple-access (CDMA) communications system is proposed. Using this algorithm, the Hopfield neural network is embedded into the GQA to improve further the fitness of the population at each generation. Such a hybridization of the GQA with the Hopfield neural network reduces its computational complexity by providing faster convergence. A better initial data estimation supplied by the GQA improves the performance of the HNN, and the embedded Hopfield neural network improves the performance of the GQA. Simulation results show that the proposed approach of multiuser detection has significant performance improvements over conventional detector and the previous detectors based on intelligent algorithms in terms of bit-error-rate, multiple access interference and near-far resistance.

Key words: Genetic quantum algorithm, Hopfield neural network, Multiuser detection, CDMA

中图分类号: