Abstract
Inclusion of statistical knowledge of the primary user (PU) channel usage had shown to be beneficial in dynamic spectrum access. Motivated by this fact, this paper investigated the importance of collecting and using statistics on neighboring secondary users (SUs) in selecting channels in addition to the knowledge of PU channel usage. The paper assumed that PU traffic characteristics of the channels are included in the radio environment map in the form of probabilistic suffix trees, which is a sequence predictor based on Markov property. In the proposed method, an intelligent sequence hopping-based common control channel and a carrier sense multiple access (CSMA)/collision avoidance (CA)-based medium access control (MAC) protocol were introduced. As shown in the paper, selecting channels using statistics of both the neighboring SUs and PUs reduced the number of packet collisions compared to a scheme which only uses PU statistics. Furthermore, the simulation results showed that the scheme proposed had better throughput performance with respect to both the random channel selection scheme and the scheme which only uses PU statistics while having less training complexity.
Highlights
The vast growth in wireless applications in the past decade has pushed the FCC to search for efficient spectrum allocation mechanisms to avoid spectrum scarcity [1]
We presented in this paper, the radio environment map (REM) learned the primary user (PU) channel usage distributions from the past channel usage data using a scheme called the probabilistic suffix tree (PST) algorithm [6], which belongs to the class of variable-order Markov models (VMMs)
6 Conclusion In this paper, we presented a proactive channel access medium access control (MAC) for an ad hoc cognitive radio (CR) network having a REM for initial setup and showed how this MAC can be used to keep the interference experienced by the PUs at a low level while providing better throughput for the secondary users (SUs)
Summary
The vast growth in wireless applications in the past decade has pushed the FCC to search for efficient spectrum allocation mechanisms to avoid spectrum scarcity [1]. We presented in this paper, the REM learned the PU channel usage distributions from the past channel usage data using a scheme called the PST algorithm [6], which belongs to the class of variable-order Markov models (VMMs). The heuristic schemes proposed in the literature being simple can lead to more SU packet collisions, because all the SUs use the same schedule only varied because of imperfect sensing They lead to inefficient channel usage in a CSMA/CA-based protocol. To find out the states that matter in the variable-order Markov model, we used an algorithm called the PST algorithm, which was introduced in [6] To use this algorithm, we needed a training sequence of sufficient length which represents channel behavior under normal circumstances, in our case, it was the channel state data gathered by the REM which are binary sequences. The downside to this channel selection method is not having a guaranteed rendezvous time as the schemes [24] and [23]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
