Fuzzy effect based dynamic channel assignment

Abstract

A class of dynamic channel assignment (DCA) algorithms referred to as the fuzzy distributed DCA (FDDCA) is presented. From decision point of view, conventional DCA algorithms are global and crisp decision makers. The decision of to or not to assign a channel for an incoming call is a crisp logic evaluation over a set of crisp system constraints. Complex logic operations are normally required to achieve the performance bound, resulting in increased cost and delay. In contrast, FDDCA algorithms are soft decision makers and the decision making is distributed. Every base station is responsible for making decisions of assigning channels for incoming calls. Fuzzy logic is employed as the decision making logic. Interference constraints are softened and treated as fuzzy sets characterized by membership functions. The algorithm was tested on the example by Sivarajan et. al. (see Proc 40th IEEE Vehicular Tech Conf., p. 631-637, May 1990), and is shown to outperform crisp DCA algorithms in throughput capacities, especially within the practical range of 1-5% blocking.<>