| Popis: |
Cognitive radio networks (CRNs) is an emerging paradigm to exploit the spectrum holes very intelligently. In CRNs, it is important to use the under-utilized spectrum very effectively without affecting the primary users. However, the sudden appearance of the primary user forces the secondary user (SU) to hand-off to another available channel to continue it's transmission. CR system performances highly depend on the spectrum access behaviour of licensed users. Minimizing spectrum hand-off, blocking probability and dropping probability of secondary users (SUs) are the most challenging task in CRNs. Most of the existing dynamic spectrum access (DSA) schemes consider only licensed spectrum for the secondary users traffic without considering the transmissions also in the unlicensed bands. Thus to reduce dropping and blocking probabilities of secondary users, we extended classical schemes i.e. random access as well as reservation based channel access which consider unlicensed bands also. An analytical model based on 4-dimensional Markov chains is developed for random access as well as reservation access schemes that consider channel access opportunities in both licensed and unlicensed band. We studied opportunistic channel access problem for finite number of primary and secondary users. In addition, the hand-off, dropping and blocking probabilities of secondary users are derived as performance metrics and compared with the random scheme and reservation based schemes that do not considers unlicensed bands. |
