Optimization for Multiple-Beams Through K-Means Clustering Algorithm

This paper deals with the problem of determining the optimal number and angle among the multiple beams in a single-cell scenario. In order to solve this problem, we proposed a novel interference algorithm called the K-means Interference Avoidance Algorithm (KIA) in Conventional One-beam Tilting (COT), Conventional Double-beams Tilting (CDT) and Fractional Frequency Double-beams Tilting (FFDT). Furthermore, we propose Practical K-means Interference Avoidance Algorithm (P-KIA) to select the number and angle of beams adaptively for alleviating the Inter-Beam interference (IBI) in each environment of User Equipment (UE) and gNB. Finally, this paper shows both cell throughput and throughput distribution over System Level Simulation (SLS) for performance evaluation of the proposed algorithm.