An Efficient Radio Admission Control Algorithm for 2.5G/3G Cellular Networks

We design an efficient radio admission control algorithm that minimizes blocking probability subject to the condition that the overload probability is smaller than a pre-specified threshold. Our algorithm is quite general and can be applied to both TDMA-based cellular technologies, such as GPRS and EDGE, and CDMA-based technologies, such as UMTS and CDMA2000. We extend prior work in measurement-based admission control in wireline networks to wireless cellular networks and to heterogeneous users. We take the variance of the resource requirement into account while making the admission decision. Using simulation results, we show that our admission control algorithm is able to meet the target overload probability over a range of call arrival rates and radio conditions. We also compare our scheme with a simple admission control algorithm and also show how to use our approach for the carrier selection problem