Spectrum sharing in a cellular system

Abstract

The cost of long term leasing of spectrum has proven to be a major road block in broad deployment of 3rd generation cellular radio systems. Similar economical road blocks can be predicted for systems beyond 3G. In order to alleviate the burden of spectrum cost incurred by service providers, there must be a paradigm shift from fixed allocations to a more flexible spectrum management. Barring practical implementations, this is indeed a feasible economical solution since the spectrum as a commodity does not depreciate with use. The search for novel spectrum management techniques has been ongoing for a decade or so and has taken many different forms from dynamic channel assignment to dynamic spectrum allocation, software defined radio, and cognitive radio. These different methodologies attack a common problem, efficient spectrum management, with various tools and complexities and have had varying degrees of success. This paper introduces a simple concept of spectrum sharing in a cellular radio system. We propose two protocols to manage spectrum sharing by redistributing excess users to spectrum bands with excess user capacity. We analyze the performance of ideal sharing by evaluating the probability of call block. We also simulate the performance of the two protocols in a realistic cellular environment. Both analyses show great gains in performance when the base-stations decide to share the spectrum as opposed to only assign users to their licensed band.