Cell size and shape adjustment depending on call traffic distribution

In this paper the potential of a smart antenna based dynamic cell size and shape control scheme is assessed. By intelligent control of antenna pattern, the whole cellular network performance can be improved by contracting the antenna pattern around the source of peak traffic and expanding adjacent cells coverage to fill in the coverage loss. We perform a constrained optimization of antenna patterns by using real-coded genetic algorithms (RCGA), and develop a CDMA cellular system simulator to evaluate the overall improvement of the system performance. A transformation of the problem space is used to remove the principal power constraint. A problem with the intuitive transformation is shown and a revised one is presented. This highlights a problem with transformation-based methods in genetic algorithms. While the aim of transformation is to speed convergence, a bad transformation can be counter-productive. Optimization results for two scenarios show potential capacity improvement exceeding 20%.