Maximization of sum-rate capacity in MU-MIMO Broadcast system with computationally efficient algorithms

Abstract

The aim of this paper is to optimize the sum-rate in a wireless system with multiple users and multiple inputs and multiple outputs (MU-MIMO). The optimal value of the channel rate is achieved using the dirty paper coding (DPC) technique. But the DPC is accompanied by a large computational complexity and a requirement for complete channel state information (CSI) at the transmitter. The complexity of the system grows exponentially with the number of users. So, the DPC is practically prohibitive in networks with low latency and high data rate requirements. To achieve near-optimal channel capacity in a MU-MIMO broadcast (BC) system, we apply meta-heuristic optimization methods to select an optimal set of antennas from both the base station (BS) and receiver side. The selection process is accomplished by recently developed arithmetic optimization algorithm (AOA) and golden jackal optimization (GJO) algorithms, and it is seen that these optimization strategies provide almost the same throughput as DPC. Along with that, the computational and temporal complexity of the optimization algorithms is insignificant in comparison to the complexity of DPC. We propose modified versions of the GJO and AOA algorithms and utilize them in the MU-MIMO and also in massive MIMO systems. The results show that, in terms of throughput, the AOA algorithm outperforms the GJO algorithm in the MIMO BC system as well as in massive MIMO system.