基于免疫克隆选择算法的多用户检测技术研究

为了解决最佳检测器计算复杂度较高的难题，提出一种精简有效的克隆选择算法。把人工免疫系统和神经网络系统的信息处理机制引入到CSA提出了免疫克隆选择算法。所提ICSA通过使用随机Hopfield神经网络制备疫苗构成新的免疫算子，把新的免疫算子结合到克隆选择算法中，不仅加快了克隆选择算法的收敛速度，并提高了克隆选择算法的全局收敛能力。然后在CDMA系统利用此算法设计了新的多用户检测器。仿真结果证明了ICSA检测器能够快速收敛到全局最优解，并且无论抗多址干扰和抗远近效应能力都优于传统方法和一些应用优化算法的多用户检测器。

Study of Multiuser Detection Technology Based on Immune Clonal Selection Algorithm

To resolve high complexity of optimum multiuser detector,a simple and effective clonal selection algorithm(CSA)was proposed.By introducing the information processing mechanism of artificial immune systems and neural network to CSA,an immune clonal selection algorithm(ICSA)was proposed.In ICSA,a new method of preparing vaccines using stochastic Hopfield neural network was introduced to form immune operator.The proposed ICSA is a hybridization of the CSA with the immune operator that reduces the computational complexity by providing faster convergence and improves the performance of CSA.Then a novel multiuser detector based on ICSA was designed for code division multiple access communication systems.Simulation results show that the proposed detector not only can achieve the global optimization values in fast convergence rate,but also is superior to the conventional receiver and the multiuser detectors based on the previous optimal algorithms in the cancellation of multiple-access-interference and near-far effect.