الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 103 |  |  |
| | | from | 103 | | |
Abstract
Cognitive radio networks enhance the spectrum utilization by allowing an unlicensed (secondary) user to access the licensed frequency band while not being used by the licensed (primary) user. Challenges arise due to the non-homogeneous delay requirements for secondary users and the temporary use of the spectrum by the secondary users as they have to vacate the channel upon the appearance of the primary user and select a target channel to continue their unfinished transmission in a process called spectrum handoff. This study is concerned with the proactive target channel selection approach, where the target channel is predetermined before starting the transmission based on long term statistics. The IEEE 802.22 wireless regional area networks standard suggests two proactive target channel selection approaches, namely, the always stay and the always change. The former states that the SU stays at his current channel every time he gets interrupted by the PU. However, the latter states that the SU has to change his current channel after each interruption. Moreover, we consider prioritized secondary users networks in which the secondary user traffic is classified into M prioritized classes based on delay requirements in order to support the delay-sensitive secondary user applications Firstly, an interruption-based priority mechanism is introduced. In this mechanism, after each interruption from the primary user the priority of the secondary user increases within the same class in order to reduce the handoff delay. A mixed preemptive/non-preemptive resume priority M/G/1 queuing model is used to analyze multiple spectrum handoffs. The analytical results are applied to evaluate the latency performance of the proposed queuing model and to compare it with existing spectrum handoff systems for prioritized and unitary secondary user networks. This comparison is performed based on the target channel sequences mentioned in the IEEE 802.22 wireless regional area networks standard. Results show that the proposed system outperforms existing systems in terms of expected handoff delay and extended data delivery time under various network conditions. Secondly, a general target sequence, where a secondary user can select either to stay at his current channel or change his operating channel after each interruption is proposed. Analytical formulae for both the extended data delivery time and the dropping probability are derived. Moreover, this general target sequence is optimized using a genetic algorithm in the seek of minimizing the average extended data delivery time. The performance of the optimized target sequence is compared with the target channel sequences mentioned in the IEEE802.22 wireless regional area networks standard for different network conditions of homogeneous and non-homogeneous loads. Based on the numerical results, a sub-optimal target sequence in case of a homogeneously loaded network is suggested. Finally, a hybrid priority discipline which attempts to relax the extreme conditions of the preemptive and non-preemptive priority disciplines is proposed. The proposed hybrid discipline is based on the number of interruptions that a SU has encountered during his transmission time. A comparison between the proposed hybrid discipline, non-preemptive discipline and the classical preemptive disciplines in terms of system time and extended data delivery time is presented. Numerical results show that, the proposed hybrid discipline reduces the extended data delivery time for lower priority users while preserves the extended data delivery time for the higher priority users.