Analysis of cognitive radio spectrum access based on fractional channel reservation

Abstract

In cognitive radio networks, channel reservation is an effective measure to solve the trade-off problem between dropping probability and blocking probability. Because of the integer nature of channel, most of the current reservation schemes were based on integer. However, the number of reserved channels of integer reservation schemes can't be adequately adjusted as a result of optimizing the system performance. In this paper, we model the cognitive radio network as a two-dimensional Markov process, which overcomes the restriction of integer nature of channel. Then two fractional reservation schemes are realized. One is fractional channel reservation for primary user (FCRPU), the other is fractional channel reservation for handoff secondary user (FCRHSU), both of which can solve the trade-off problem above very well. The precision of analytical results is validated by Monte Carlo Simulations, on the strength of which we come to the conclusion that fractional channel reservation scheme and the FCRHSU scheme can achieve better performances than the integer one and the FCRPU one, respectively.