Fairness optimization of thinning Call Admission Control in wireless networks

Abstract

Resource reservation Call Admission Control (CAC) schemes constitute an efficient solution for prioritizing high priority calls. Issues concerning fairness among calls arise during the implementation of CAC. In this paper, a new optimization model selecting the admission probability in thinning schemes is proposed to provide fairness in resource sharing among different input call stream flows. The existence of an optimal admission probability value is proven based on users satisfaction under different traffic input load maintaining users Quality of Service (QoS) levels. Based on this analysis, an effective fairness and thinning CAC policy is proposed, where the admission decision is dynamically adjusted based on appropriately defined users utility. The analysis is performed under different degrees of users risk aversion, reflecting their behavior towards the perceived QoS levels. The optimal values are determined through an iterative algorithm. The performance results obtained verify the improvement achieved by the integrated model, especially when networks operate near congestion.