Journal of Information Science and Engineering, Vol. 34 No. 4, pp. 851-867

Analysis of Spectrum Handoff under General Residual Time Distributions of Spectrum Holes in Cognitive Radio Networks

Spectrum handoff probability, spectrum handoff rate or average number of spectrum handoffs are important parameters in the evaluation of performance metrics of Cognitive Radio Networks. Due to the mobility of spectrum holes, a Cognitive Radio user needs to perform spectrum handoff successfully to complete its data transmission. This paper scrupulously investigates and analyzes the probability of spectrum handoff under diverse realistic primary and cognitive user traffic models. A theoretical analysis of spectrum handoff probability and average number of spectrum handoffs is presented here under the assumption that the residual time of spectrum holes are exponential, Erlangian and Weibull distributed. We also present impact of departure rate of cognitive users (*µ*) and spectrum holes (*λ*) on the probability of spectrum handoff and rate of spectrum handoffs. The accuracy of the derived distribution is validated by simulation results for the Weibull (*k*=3) distribution function and a detailed comparison with exponential, Erlang-2 and Erlang-3, 23 distribution of residual time of spectrum holes is presented.

Keywords:
cognitive radio, residual time, spectrum handoff, spectrum mobility, spectrum hole, Weibull distribution