On the Effect of a Priority Algorithm and Backoff Policies in the Performance of a CDMA Cellular System with Integrated Services

In this work, we evaluate the performance of a CDMA cellular system with integrated services: voice and data. The activity of voice user is an exponential ON/OFF model. The data user session is a WWW browsing type. When a base station transmits information to a user it does it through supplemental code aggregation. We evaluated the impact of data users on the system capacity and in the forced termination calls. With integrated services the dropout probability is greater than the blocking probability, this degrade the QoS of the system. In this sense, we proposed a priority algorithm. The algorithm is based on the unconstrained delay of data service and the objective is to reduce the forced call terminations. It shows that with our priority algorithm the dropout probability is smaller than the blocking probability, and then there is an improved QoS support in the system. The access protocols use backoff strategies like a constant backoff, uniform backoff, geometric backoff and binary exponential backoff; we apply these to generate backoff delay for data users when the supplemental channel request is rejected or the burst is interrupted. Call admission control (CAC) is used to regulate the traffic load in the system. To maintain QoS requirements power adjust is considered. We use a static simulation to evaluate the performance of the system (call blocking and dropout probabilities, packet request delay, coverage, mean time of data session and burst admission).