Title: A hyper-spherical surface search optimisation algorithm for practical cellular wireless network

Authors: Wenxiang Wang; Zhiping Tan; Lanlan Kang; Yanxin Lai; Jialin Li

Addresses: Gannan University of Science and Technology, No. 156 Hakka Avenue, Zhanggong District, Ganzhou, China ' Guangdong Polytechnic Normal University, No. 293 Zhongshan Avenue West, Tianhe District, Guangzhou, China ' Gannan University of Science and Technology, No. 156 Hakka Avenue, Zhanggong District, Ganzhou, China ' Gannan University of Science and Technology, No. 156 Hakka Avenue, Zhanggong District, Ganzhou, China ' School of Science and Technology, Gannan Normal University, No. 61 Hongqi Avenue, Ganzhou, China

Abstract: The optimisation of practical cellular wireless networks is a complex mixed-variable large-scale multi-objective optimisation problem. Traditional multi-objective algorithms encounter problems such as the curse of dimensionality, slow convergence speed, and quantisation errors when solving such problems. To address this problem, a large-scale multi-objective hyper-spherical surface search optimisation algorithm, named HSSOF-BHA is proposed in this paper. Firstly, a hyper-spherical surface search method is used to compress the high-dimensional search space. Secondly, a micro-hyper-spherical surface local search mechanism is performed to enhance the coverage performance. Thirdly, a mixed-variable evolutionary operator is employed to overcome quantisation errors. The experiments in the practical cellular wireless network optimisation model show that the proposed algorithm has significant advantages over the other four state-of-the-art comparison algorithms, and the indicators can increase RSRP by 5.72 dBm, SINR by 3.15 dB, and reduce OCR by 9.56% compared to manual optimisation indicators of the current network.

Keywords: hyper-spherical surface; large-scale optimisation; mixed-variable optimisation; evolutionary algorithm; wireless network optimisation.

DOI: 10.1504/IJBIC.2025.150636

International Journal of Bio-Inspired Computation, 2025 Vol.26 No.4, pp.220 - 231

Received: 16 Oct 2024
Accepted: 25 Jan 2025
Published online: 18 Dec 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article