Abstract
In this paper, we consider the dynamic resource allocation problem for orthogonal frequency division multiple access being used in modern broadband wireless systems such as IEEE 802.16 (WiMax). The objective is to allocate the network resources in an efficient manner in order to maximize the system throughput while guaranteeing the minimum level of quality-of- service to all users under practical conditions such as intercell interference. In order to reduce the communication overhead and computational complexity, we propose a suboptimal two-step approach. Firstly, the radio network controller solves the large scaled network planning problem based on aggregated information about all users in the network. Given the network planning results, each base station solves the small scaled cell throughput maximization problem based on its full knowledge of all users within its cell. We demonstrate the improvements of our scheme using the IEEE 802.16 (WiMax) system as an example. We show that with our proposed dynamic allocation scheme, it is possible to achieve the significant throughput gain with less complexity and signaling overhead.