Spectrum Allocation in Cognitive Networks with Learning Automata

Cognitive radio networks (CRNs) involve extensive exchange of control messages, which are used to coordinate critical network functions such as distributed spectrum sensing, medium access, and routing, to name a few. Frequent channel-switching will bring many problems such as delay, packet loss and communication cost. To mitigate the influence of these problems, it is necessary to reduce the channel-switching times. After reviewing the prior works about spectrum allocation we propose a LAGSA (Learning Automata based Global Spectrum Allocation) algorithm in this paper. It can give guidance to the next allocation process by using the information obtained from the historical data transmission results. By the simulation we have discussed the relationship between algorithm astringency andspectrum idle probability, learning pace respectively. Comparing with Greedy allocation algorithm, fixed allocation algorithm and random allocation algorithm in terms of average successful transmission ratio and channel-switching times, AIGOSA has obvious advantage for improving the global spectrum utilization ratio.

Keywords:

Channel assignment , Ad-hoc networks , Learning automata , TDMA