Loading...

Optimization of Random Access for Wireless Networks with Delay Sensitive Traffic

Khodaian, Amir Mahdi | 2012

515 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 43077 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Hossein Khalaj, Babak
  7. Abstract:
  8. Since random access networks can work in distributed manner they have found variety of applications in wireless local area networks and wireless sensor networks. Throughput maximization and calculation of optimal parameters of such networks are subject of many research but most of these researches have ignored delay and suffer from high packet delay. We investigate and analyze delay in random access and solve the delay constrained utility maximization problem. We define network utility as a combination of throughput and cost of energy consumption and solve “access network optimization with link delay constraint”, and “cross layer optimization of transport and access layer with end to end delay constraint”. We also investigate the optimal tradeoff between energy consumption, throughput and delay. We use sub-gradient and Newton like algorithms in distributed optimization algorithms and compare their convergence rates through simulations.
    We also consider adaptive queue management along with optimal access control and introduce cross layer adaptive algorithms. We consider a dynamic programming model and define the per stage utility function as a concave function of throughput, energy consumption, and average queue length. We then propose algorithms that can maximize average of utility per stage in finite or infinite time horizon. We show that the optimal transmission strategy has a threshold structure versus the channel state so the transmitter should send its packets in case channel is in a state better than that threshold. We also propose a method for queue management in which packet arrival is controlled based on the link state vector.
    Finally we model delay optimization in energy harvesting cognitive radio networks and use dynamic programming to find optimal transmission strategy of the secondary user. We take into account that the interference of the secondary user should not surpass the permitted level and control quality of service for the primary user of the channel
  9. Keywords:
  10. Random Access ; Dynamic Programming ; Adaptive Control ; Cognitive Radio ; Utility Maximization

 Digital Object List

 Bookmark

No TOC