Adaptive admission control for non-homogeneous handover requests in wireless networks

In practical wireless networks, many users move in a batch, such as by bus and by subway, and this causes a non-homogeneous handover request pattern. In such a nonhomogeneous pattern, the conventional method using the (homogeneous) Poisson assumptions for bandwidth reservation and allocation may cause invalid resource allocation planning. We propose an adaptive admission control scheme, based on the estimation of near future traffic load and probabilistic bandwidth reservation. The amount of bandwidth for admitting handover requests in the near future is estimated, based on a probability distribution estimated from historic data. An optimization model is developed, in a distributed manner, to determine an adequate amount of bandwidth to reserve according to realtime network traffic conditions. With this, an efficient and exact algorithm, using the optimality condition, is also studied, based on mathematical programming. Our scheme could be utilized as a guideline for cost-effective radio resource management in mobile multimedia networks.