Mobile Small Cell Deployment in Wireless Communication Systems

• Main Authors: Shih-Fan Chou, 周詩梵
• Other Authors: Ai-Chun Pang
• Format: Others
• Language: en_US
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/8z39m9

博士 === 國立臺灣大學 === 資訊工程學研究所 === 106 === One viable and low-cost method of accommodating the explosive growth of mobile broadband traffic is to introduce small cells for next generation cellular networks. However, static small cells cannot be flexibly placed to meet the demand of time/space-varying traffic, and idle or under-utilized cells would result in resource wastage and system performance degradation. Therefore, this dissertation adopts the mobile small cell concept and seeks to optimize the deployment of mobile small cells. If a finite number of mobile small cells can serve more users for more time, the mobile small cell deployment will have more gains. To reveal the performance gains from proper deployment strategies, this dissertation uses ground and airborne vehicles respectively to serve as the carriers for mobile small cells. We first target the deployment problem on the ground with the objective of maximizing total service time of all users. Specifically, service time maximization exhibits an interesting trade-off between user density and the travel time of mobile small cells. We prove that our target problem is NP-hard and cannot be approximated in polynomial time with a ratio better than (1 – 1/e), unless P =NP. To solve the problem, we propose a polynomial time (1 – 1/e)-approximation algorithm, and the proposed algorithm is one of the best approximation algorithms based on the inapproximability ratio. Next, we extend our preliminary results on 2D deployment to further accommodate the flexible deployment of flying unmanned aerial vehicles (UAVs), with the goal of maximizing the total throughput of all users. The problem is formulated as a non-convex non-linear program and its convexified reformulation can be solved by Lagrangian dual relaxation and subgradient projection methods. We then propose a heuristic algorithm to deal with the trade-off among flight altitude, travel time and battery life. The capabilities of the above-mentioned proposed algorithms are evaluated by conducting a series of simulations with realistic parameter settings, providing insightful and encouraging results in mobile small cell deployment for wireless communication systems.