Loading...

VLC Access Networks: Optimizing Access Point Placement and User Association

Dastgheib, Mohammad Amir | 2023

200 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 56670 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Salehi, Jawad; Beyranvand, Hamzeh
  7. Abstract:
  8. In this dissertation, the optimization problem of visible light communication (VLC) access networks is discussed and investigated. Firstly, by examining the fundamental concepts in VLC systems, the channel capacity is analyzed, and the concept of capacity per unit cost is introduced to improve energy consumption. The appropriate relationship for channel capacity in optimization studies is also explored. In this regard, we reached a new distribution for transmitted signals, which, by reducing up to approximately 92% of the additional energy required for data communication, increases energy efficiency in visible light communication systems up to 98%, bringing the energy consumption of the combined communication and lighting system very close to the energy consumption of the lighting-only system. Next, the impact of access point placement on network performance is examined, and geometric structures representing coverage regions are utilized for optimizing the placement strategy. The results indicate that the optimal layout of access points can increase the average provided rate to users by approximately 70%. After determining the placement, the user association problem is studied, and a method is proposed to transform the allocation objective into a convex function. A statistical formulation is presented for optimizing this problem over the geometric structures. We demonstrated that with an increase in the number of users and, consequently, an increase in the complexity of the problem, the statistical form yields results similar to the actual system behavior. To overcome computational challenges, a method utilizing stochastic gradients for optimization is proposed, leading to a novel machine learning algorithm. Numerical results show that the use of this algorithm can significantly improve the speed of the discussed optimizations by about two orders of magnitude. Overall, this dissertation contributes insights and methodologies for the design of more efficient and effective VLC access networks, and it can be applied to achieve better system design and solve optimization problems in this context
  9. Keywords:
  10. Optical Wireless Communication Networks ; Visible Light Communication ; Optimal Placement ; Machine Learning ; Geometric Structural ; Spatial Tessellations ; Access Points Placement ; User Association

 Digital Object List

 Bookmark

...see more