This dissertation develops an analytic framework to undertake cell design in a cellular system. The cell is modeled in a broader sense than ever done before. In our analytical model, we incorporated the use of guard channels, queueing of new calls, and hybrid channel allocation. A numerically stable and efficient solution to a queueing system with two arrival streams having reserved and borrowable servers has been developed. This queueing system is used to model the cell behavior. The model provides valuable insights into the behavior of the cell, and this in turn has been used to devise an efficient stochastic optimization algorithm for determining the minimum number of channels required by the cell. Our techniques are stable, easy to implement for practical systems and produce optimized solutions quickly. This is particularly useful because we expect that future designs of cellular systems may execute such algorithms on cell-site processors. / Ph. D.