基于神经网络离散混合蛙跳算法的多用户检测

为进一步提高基于离散混合蛙跳算法（DSFLA）的多用户检测性能，提出一种基于DSFLA和神经网络相结合的神经网络离散混合蛙跳算法，并用于多用户检测。在DSFLA的每一族内更新中，随机选择若干只“青蛙”采用Hopfield神经网络的寻优更新策略，进行快速迭代，寻找全局最优。仿真结果证明，基于神经网络离散混合蛙跳算法的多用户检测器在误码率、收敛速度、系统容量、抗远近能力等方面都优于传统方法和一些应用优化算法的多用户检测器。

Multi-user Detection Based on Discrete Shuffled Frog Leaping Algorithm with Neural Network

To improve further performance of multi-user detection based on Discrete Shuffled Frog Leaping Algorithm(DSFLA), a novel hybrid algorithm that employs DSFLA and Hopfield Neural Network(HNN) for the multi-user detection is presented. Using this approach, updating each family in DSLA, a number of frogs are chosen at random using the updating strategy of HNN to find the global optimum. Simulation results show that the HDSFLA-MUD has significant performance improvement over conventional receivers and previous multi-user detectors based on previous optimum algorithm in terms of convergence, bit-error-rate, capacity of system and near-far resistance.