A hybrid distributed channel allocation algorithm using traffic prediction

In this paper we investigate the effect of including a traffic prediction module in a channel allocation algorithm. The presented algorithm uses a hybrid channel assignment scheme where each cell is in charge of a group of local channels and shares a single pool of global channels. The traffic prediction module incorporated is used to calculate the potential number of local channels a cell will need in the following time periods. It takes into account recent traffic history of cells in addition to their diurnal traffic periodicities. A performance comparison showed that incorporating such a prediction module into the hybrid allocation algorithm yielded significant decrease in failure rate, message complexity and acquisition delay. Simulation results showed also that a prediction module that considers both traffic periodicities and cells recent history outperformed that considering only traffic periodicities.