基于差分进化算法的MU-MIMO用户调度算法

MU-MIMO user scheduling algorithm based on differential evolution algorithm

为了使MU-MIMO系统的用户得到更公平的服务,提出了一种块对角化预编码下基于差分进化算法的用户调度算法.该算法将差分进化算法进行改进,使其可对任意可行大小的用户子集进行调度,并进行参数自适应,与改进的比例公平算法相结合进行用户调度.实验结果表明,在不同的信噪比和用户数下,该算法较改进遗传算法在效率方面有大幅度提高,且公平度、系统和数据率相对于改进遗传算法均有所提高,能有效地解决块对角化预编码下的用户调度问题.

To make the users of multi-user multiple input multiple output （MU-MIMO） system get more fair service, a user scheduling algorithm based on differential evolution algorithm under the block diagonalization pre-coding condition was proposed. The differential evolution algorithm was improved in the proposed algorithm so that the subset of users with arbitrary possible size could be scheduled, and the self-adaptive parameters were obtained. In addition, the proposed algorithm could perform the user scheduling in combination with the proportional fair algorithm. The experimental results show that with different signal to noise ratio （SNR） and different number of users, the efficiency of the proposed algorithm gets substantially improved compared with the improved genetic algorithm, and the fairness as well as the sum data rate of system are also enhanced. And thus, the problem of user scheduling under the block diagonalization pre-coding condition can be effectively solved.