AN IMPROVED NEURAL-NETWORK FOR CHANNEL ASSIGNMENT PROBLEMS IN CELLULAR MOBILE COMMUNICATION-SYSTEMS

This paper presents an improved neural network for channel assignment problems in cellular mobile communication systems in the new co-channe l interference model. Sengoku et al. first proposed the neural network for the same problem, which can find solutions only in small size cel lular systems with up to 40 cells in our simulations. For the practica l use in the next generation's cellular systems, the performance of ou r improved neural network is verified by large size cellular systems w ith up to 500 cells. The newly defined energy function and the motion equation with two heuristics in our neural network achieve the goal of finding optimum or near-optimum solutions in a nearly constant time.