Multicriteria dynamic routing in communication networks based on F learning automata

To cope with rapidly changing network conditions traffic routing methods must be adaptive, flexible and intelligent for efficient network management. The use of intelligent algorithms based on learning automata can be efficient for traffic control. However, thus far these learning schemes have been focused only to unimodal routing problems in connection oriented or packet oriented networks. We propose a novel routing scheme based on the new hybrid approach to multicriteria routing problem that combines the theory of learning automata with fuzzy logic theory. We call these automata the F type learning automata. Well known learning automata of P, Q and S types are special cases of this F type automata. We have suggested general reinforcement scheme for these automata and defined performance criteria. These automata are feasible, nonabsorbing and strictly distance diminishing. We provide conditions in order to be ergodic and expedient. Finally, we provide simulation results of the circuit switched telecommunication network whereby two criteria- quality and price, have been taken into account simultaneously.