Loading...

Spectrum Mobility Analysis in Cognitive Radio Networks

Sheikholeslami, Fatemeh | 2012

483 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 43300 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Nasiri Kenari, Masoumeh
  7. Abstract:
  8. Opportunistic spectrum access in cognitive radio networks has introduced multiple challenges in the realization of this technology. One of these key challenges is spectrum mobility, in which a secondary user (SU) must be able to vacate channels upon the appearance of a primary user (PU) and reestablish a communication link to avoid service termination, known as spectrum handoff. For a connection-based modeling, multiple interruptions can occur, which lead to multiple handoffs; hence, in a proactive approach, a sequence of target channels, rather than a single target channel must be determined. The first parts of this thesisis dedicated to the analysis of a time-slotted cognitive radio, in which SUs are forced to stay on their current channels upon PU appearance, and pause their transmission until the channel becomes idle again. To this end, considering practical limitations such as imperfect sensing and maximum allowed extended service time, parameters such as extended data delivery time, maximum achievable throughput, and packet loss probability are derived. The results are also extended to the continuous spectrum access of primary users. In this case, it is shown that the optimal duration for the time slots can provide a higher performance for the secondary user. In the second part, more advanced networks, capable of channel switching and parallel transmissions on multiple channels are investigated. In such networks, because of the important role of the secondary traffic, load balancing techniques are considered jointly with the target channel selection schemes required in spectrum handoff procedures. To this end, by utilizing queuing theory, a new network model is introduced and the joint aforementioned problem is explored. Parameters such as total sojourn time of secondary packets and maximum achievable throughput are also evaluated. Since the optimal solution has a high computational complexity, a suboptimal solution is presented. It is shown that the suboptimal solution also leads to a considerable decrease in the average sojourn time of the packets, and enables the secondary user to exploit the transmission opportunities in the network more efficiently.
  9. Keywords:
  10. Cognitive Radio ; Queuing Theory ; Spectrum Mobility ; Connection-based Analysis

 Digital Object List

 Bookmark