Dynamic resource allocation in wireless ad hoc networks based on QS-CDMA

We consider a wireless ad-hoc network for data applications using an orthogonal quasi-synchronous CDMA air interface with a linear receiver structure. The throughput performance of such networks is highly affected by a limited number of signature sequences and by individual power constraints on each node. This paper deals with strategies for allocating powers and sequences to nodes. These strategies exploit the relative delay tolerance of data applications and the random packet arrivals to improve the throughput performance. We provide an upper bound of the optimal centralized strategy, and then propose a centralized heuristic algorithm and two distributed approximations of this strategy based on clustering.