Improving Capacity and Robustness of Wireless Networks
Open Access
- Author:
- Qiu, Li
- Graduate Program:
- Computer Science
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- June 27, 2018
- Committee Members:
- Guohong Cao, Dissertation Advisor/Co-Advisor
Guohong Cao, Committee Chair/Co-Chair
Bhuvan Urgaonkar, Committee Member
Sencun Zhu, Committee Member
Zhibiao Zhao, Outside Member - Keywords:
- Cache
Network capacity
Popularity-aware caching
Social links
Amplify-and-forward
Direct link placement - Abstract:
- Wireless ad hoc networks enable communications among mobile nodes without any infrastructure support, as nodes themselves relay and forward packets for each other. Due to wireless interference and network dynamicity, they suffer from low network capacity and unreliable network connections. Specifically, due to the interference between concurrent transmissions, the per-node capacity decreases with the increasing number of nodes. Moreover, it is difficult to maintain stable end-to-end connections, especially between faraway nodes, since the communication links may experience high dynamicity. To address these challenges, we propose resource management strategies to improve the network capacity and the link robustness. To improve the network capacity, we design caching techniques, where nodes retrieve desired contents from close neighbors rather than the faraway server, to reduce the transmission distance and improve the network capacity. To maintain reliable communication between important social pairs, we devote the network resources to these important social links, rather than focus on improving the performance of data forwarding for all nodes in the network. We propose two strategies, amplify-and-forward and direct link placement, to connect faraway nodes and improve the data forwarding performance. The goal of this dissertation is to design efficient resource management strategies such as content caching, transmission scheduling and link placement to improve the network capacity and maintain important social links. First, we utilize caching techniques to improve the network capacity. When the content popularity follows a uniform distribution, we study the capacity of wireless networks with caching considering the cache size of each node, the total size of unique content in the network, and the number of nodes in the network. We present an upper bound on network capacity, and present an achievable capacity lower bound. Our results suggest that the capacity of wireless ad hoc networks with caching can remain constant even as the number of nodes in the network increases. Second, when the contents have skewed popularity, we evaluate how the distribution of the content popularity affects the per-node capacity, and design popularity-aware caching strategies that maximizes the per-node capacity for wireless ad hoc networks. Based on the popularity-aware caching strategies, we derive different capacity scaling laws based on the skewness of the content popularity. Our results suggest that for wireless networks with caching, when contents have skewed popularity, increasing the number of nodes monotonically increases the per-node capacity. Third, to maintain the communication links between important social pairs, we adopt a cooperative amplify-and-forward strategy, where nodes (relays) cooperate to improve the signal strength at the destination. We formulate and study two optimization problems for maintaining the required link throughput: Min-Energy and Min-Relay, where the goal of Min-Energy is to minimize the power consumption of the relays, and the goal of Min-Relay is to minimize the number of active relays. Evaluation results show that Min-Relay can significantly reduce the number of active relays compared to Min-Energy, while achieving comparable power consumption. Forth, to improve the robustness of the social links, we propose to proactively place some reliable communication links (e.g., satellite links) to the given network to improve the social link quality. We formulate and study the problem of maintaining social links through direct link placement (called MSL). Due to the complexity and the NP-hardness of the MSL problem, we propose solutions to bound it by two submodular functions, and derive an approximation algorithm with provable approximation ratio based on the sandwich approximation strategy. Evaluation results based on both synthetic graphs and a real-world social network dataset demonstrate the effectiveness of our proposed algorithms.