User mobility-aware device-to-device and aerial communication networks

Abstract

Due to the enormous number of cellular users, the traffic load on the core network is increasing rapidly. Hence, it is difficult to maintain the required quality of service in the cellular network. Further, the normal operations of a cellular network is also hampered in wake of a natural or man-made disaster. To support cellular network in such scenarios, recent studies have proposed the use of device-to-device (D2D) and unmanned aerial vehicle (UAV) assisted communication networks. On one hand, in D2D, the information is not routed through the base station (BS) and hence, helps in offloading cellular traffic, lowering the delay, and improving the spectral efficiency. On the other hand, UAVs offer the opportunity to realize UAV-mounted BSs and UAV relays that can dynamically reposition themselves to boost coverage, spectral efficiency, and user quality of experience. Furthermore, D2D and UAVs are also useful in areas with limited cellular connectivity. However, the user devices in D2D and UAV networks are hand-held, which necessitates the knowledge of user mobility behavior for determining and optimizing the network performance. Hence, the goal of this thesis is to optimize the D2D and UAV-assisted communication network performance based on user mobility behavior for different applications such as content caching as well as disaster-affected scenarios. Studies reveal that the user mobility has certain spatio-temporal correlations. Hence, in our work, we assess the viability of D2D and D2D multicast (D2MD) networks based on real-world mobility traces of users. Specifically, in case of D2MD networks, we propose two novel methods for exploiting spatio-temporal correlations present in the user mobility. We also demonstrate the relationship between spatio-temporal user locations and D2D/D2MD network performance. Based on the above observations, D2D/D2MD networks must carry out mobility aware network optimizations. Hence, we propose a real-world spatio-temporal user behaviour aware optimal cache selection in the D2MD network. Content caching allows certain popular multimedia content to be cached at some user devices in the network in order to reduce the content delivery time. The proposed optimization framework minimizes the number of caches while achieving a desired load on the cellular network. Next, in order to overcome the problem of limited access to cellular network, we investigate multi-hop D2D. Specifically, we utilize multi-hop D2D to establish a disaster-resilient communication network to serve users in the disaster-struck area. We solve the problem of coverage maximization, i.e., maximizing the number of end to-end connection from user to an active BS, with joint routing, scheduling and source-destination pairing which outperforms shortest path routing based solutions. Due to UAVs’ mobility and ease of deployment, we explore UAVs to facilitate a disasterresilient communication network. However, the optimal placement of UAVs over a disasteraffected area must be aware of the mobility of the ground users. Hence, we propose a ground user mobility aware multi-UAV placement strategy for disaster-resilient communication network where we maximize the number of ground users covered by the UAVs, while also taking into account the UAV flight time constraint. The above work in multi-UAV network does not consider user coverage fairness and resource allocation. It also considered disaster-specific user mobility model and users’ mobility is assumed to be synchronous. Further, the UAV-BS placement update interval is also a crucial parameter that impacts the network performance; hence, it needs to be optimized. Motivated by the above, we consider a more practical mobility model while jointly optimizing the UAV-BS placement, resource allocation, user association and update interval. To conclude, we demonstrate the impact of user mobility on the D2D and UAV-assisted communication networks. We propose various network optimization frameworks while accounting for user mobility, which will be helpful in network planning and deployment. Specifically, in our work, we target the content caching, disaster-specific and remote connectivity applications