This paper describes a novel programmatic approach to efficiently distribute resources in a dynamic cellular network, using local negotiations. Our proposed mechanism is reactive and facilitates parallel self-adaptation efforts, leading to dynamics that improve overall network performance. The local nature of the negotiations being performed as part of the adaptation process enables frequent changes in the network's parameters with a negligible coordination overhead. The results of our experiments suggest rapid adjustment to changes and overall improvement over time in the number of users served by the network. We evaluate our algorithm based on the service level index, measured by the number of covered handsets. Nevertheless, the proposed algorithm supports any set of parameters and any combination of performance measures supplied by service providers.
展开▼
