Improvement of Borrowing Channel Assignment by Using Cellular Probabilistic Self-organizing Map

Abstract

Borrowing channel assignment (BCA) is used for efficient spectrum allocation in cellular systems. However, when many cells with heavy traffic cluster together and there are still some cells with light or medium traffic, i.e., patterned traffic load, BCA is not effective for spectrum allocation. In this paper, an improvement of BCA for patterned traffic load, based on traffic prediction by cellular probabilistic self-organizing map (CPSOM), is proposed. Upon the prediction of traffic that does not vary fast, the proposed method has better performance for patterned traffic load than the traditional BCA methods. Experimental results corroborate that the proposed method achieves higher satisfaction probability than the traditional FCA and BCA techniques.