Framework for Dynamic Resource Allocation to Avoid Intercell Interference by Evidence Theory

Abstract

The paper proposes a framework to allocate the radio resources dynamically to avoid intercell interference (ICI) using evidence theory in MIMO-OFDMA (Multiple Input Multiple Output–Orthogonal Frequency Division Multiple Access) downlink system. In order to utilize the available radio resources such as bandwidth and power for improving cell throughput, the neighboring cells need to share the common resources. Hence, user terminals (UTs) at cell boundary experience severe ICI and contribute to the reduced overall cell throughput. We propose evidence-based Dempster–Shafer Combination Rule (DSCR) to estimate the confidence factor (CF) based on evidence parameters. We estimate the confidence interval (CI) for the bifurcation of cell users based on either decision ratio or CF as cell center and cell edge. With DSCR, an uncertainty contained in the hypothesis is avoided. The obtained evidences are the decisive factors to allocate resources dynamically to avoid ICI. The simulation results demonstrate the improvement in cell edge performance and in turn enhance the overall cell capacity.