Capacity Balancing Between the Reverse and Forward Links in Multiservice CDMA Cellular Networks With Cross-Layer Design

Abstract

Capacity unbalance is a practical problem for future cellular networks where the forward link traffic and the reverse link traffic are asymmetric. In this paper, an analytical framework for balancing the reverse and forward link capacities with an adaptive soft handoff probability (SHP) scheme in multiservice code-division multiple-access cellular networks is presented. With the proposed adaptive SHP scheme, system capacity can be balanced under various reverse and forward link traffic volumes, and system performance is optimized at the optimal SHP. A cross-layer model involving the physical layer, the link layer, and the network layer is presented. SHP in the physical layer, the outage probability in the link layer, and connection admission control (CAC) schemes, including complete sharing and virtual partitioning, in the network layer are jointly considered. The quality of service metrics in the link layer, including SIR and the outage probability, is derived with the information from the physical layer. The admission region is obtained by satisfying the outage probability requirements in both the forward and reverse links. Based on the admission region, the network layer grade of service, including the new connection blocking probability, the handoff connection dropping probability, and the throughput, is formulated as the performance metrics. The optimal SHP is selected by looking for the lowest penalty of connection blocking, which counts in both new connection blocking and handoff connection dropping probabilities. The cross-influences between the selection of the optimal SHP and the CAC schemes are also investigated. Numerical results show that the maximum achievable gain in throughput for the high-revenue services is about 80% by combining the benefits from the adaptive SHP and dynamic CAC schemes as compared to when none of these schemes is used.