Research on Minimizing PCI Interference Based on Intelligent Optimization Algorithms

Abstract

This study focuses on optimizing the allocation of Physical Cell Identities (PCIs) in wireless communication networks to minimize Measurement Report (MR) conflicts, confusion, and mod-3 interference. Starting with a raw dataset containing conflicting and confusing MR data, as well as mod-3 interference, the dataset is cleaned and refined to retain 2,890 communities with MR-related data. A model is then developed to allocate PCIs to 2,067 core communities, aiming to minimize the total MR conflicts, confusion, and mod-3 interference. The problem is formulated as an integer linear programming (ILP) model, where the decision variable indicates whether a specific PCI is assigned to a community. The objective function minimizes the sum of MR conflicts, confusion, and mod-3 interference, subject to constraints such as unique PCI allocation and mod-3 interference conditions. To solve the problem, two optimization techniques are explored: the Monte Carlo method and simulated annealing. While the Monte Carlo method was less effective due to the complexity of the relationships between variables, the simulated annealing algorithm provided better results for PCI reallocation. This study demonstrates an effective approach for PCI allocation and highlights the benefits of using simulated annealing for complex network optimization.