NeXt generation/dynamic spectrum access/cognitive radio wireless networks: a survey
Computer Networks: The International Journal of Computer and Telecommunications Networking
Analysis of cognitive radio spectrum access with finite user population
IEEE Communications Letters
VoIP capacity analysis in cognitive radio system
IEEE Communications Letters
Voice service support over cognitive radio networks
ICC'09 Proceedings of the 2009 IEEE international conference on Communications
Proceedings of the 6th International Conference on Queueing Theory and Network Applications
| Hi-index | 0.00 |
In this paper, we analyze an unslotted carrier sense multiple access(CSMA) in multi-channel cognitive radio networks. The usage pattern of a channel by the primary users(PUs) follows On/Off process and is independent with other's channel activities. The multi-channel based unslotted CSMA is operated as follows. Each secondary user(SU) senses all channels(called all sensing) when SU wants to send a packet. If a SU finds idle channels, then the SU selects randomly one among those channels and transmits a packet. After a packet transmission, the SU goes to exponential distributed vacation in order to prevent the channel being exclusively used by one SU. If the SU does not find any idle channels, then the SU goes to vacation and then senses all channels again. When a PU returns to the channel before the SU finishes its packet transmission on the channel, then the SU vacates the channel to the PU and senses all channels immediately. If the SU finds idle channels, then the SU continues to transmit its current packet on one of the idle channels, otherwise, the SU goes to vacation and the SU's packet transmission is regarded as unsuccessful. We model this unslotted CSMA with all sensing by continuous time Markov chain(CTMC) with Level Dependent Finite QBD structure and obtain the steady state probability of the Markov chain using matrix analytic method. We obtain several performance measures such as the normalized throughput of SUs per a channel and the forced termination probability of SUs. Next, we introduce a concept of random m-sensing which generalizes all sensing and investigate an unslotted CSMA with random m-sensing. Random m-sensing means that a SU senses m channels by random selection among number M of channels. Numerical results show that the throughput of SUs in unslotted CSMA with random m-sensing increases as the number m of sensing channels increases. In particular, the throughput of SUs with random M-sensing(i.e., all sensing) is much larger than that with random 1-sensing(i.e., random sensing).