An Efficient Mobile Data Offloading Solution for Opportunistic Networks

Abstract

Mobile data offloading will be the need for upcoming networking technologies like 5G and D2D, due to its significant contribution towards efficiently utilizing the network bandwidth with limited energy usages. There are several networking alternatives like Bluetooth, WiFi, and opportunistic communications, which may help deliver data that would typically be required from cellular operators. The utilization of network bandwidth can be significantly improved by data offloading. Mobile data offloading plays a significant role in saving cellular network bandwidths. It helps the network to increase its throughput as a large number of users can receive the data. The problem of mobile data offloading is characterized by identifying the NP-hard target set selection (TSS) problem. TSS is the method of selecting a minimal set of nodes for sharing duplicate data. The collaborative viewpoints of users along with the ISPs may propose optimizations for identifying efficient target sets. At the same time, the presence of malicious users in the target set may hinder the network activities either by spoofing or by reducing the offloading potential of the network. These malicious users may result in decreasing the overall performance of the network. In this thesis, we propose a hybrid communication mechanism for determining the target set of users. We use the properties of opinion dynamics amongst users to address the issue of overlapping community detection. The security aspect of the target set is addressed using the trust-based metrics of the users. Additionally, the thesis also proposes a methodology to distribute incentives to the users in the target sets, addressing the concerns related to partial incentive distribution or third-party intervention. newline