Stochastic geometry-based analysis of relay-assisted spectrum sharing future generation wireless networks

Many technologies are set to revolutionize the efficiency of humans and devices communication. Three of the trending technologies envisaged to provide huge prospects towards the realization of the fifth-generation cellular systems are machine-to-machine (M2M) communication, device-to-device (D2D) communication and cognitive radio networks (CRNs). M2M facilitates the autonomous communication of smart devices while D2D facilitates direct connectivity between devices in proximity. Cognitive radio aids effective utilization of wireless spectrum as cognitive devices could opportunistically access the spectrum of licensed users. Similarly, relays improve the transmission coverage between nodes which in turn reduces the outage probability (OP) and increases the transmission capacity (TC) of the spectrum sharing network. These metrics can be effectively studied using stochastic geometry (SG): a mathematical tool for deriving insights into the performance of wireless networks of different spatial configurations. Motivated by these, this thesis is aimed at addressing three of the research gaps related to spectrum sharing systems (D2D and CRNs) assisted by relays using SG. The coexistence of a massive number of machine-type devices (MTDs) with D2D and cellular users is set to heighten the interference levels within the cellular architecture. On the other hand, D2D devices would require relays whenever they are farther apart to improve the outage performance. However, limited battery or the non-altruistic nature of certain users may deter them from helping other users to relay data. Motivated by the recent specifications of MTD devices, the first contribution in this thesis conceptualizes that MTDs can relay data for D2D devices that are not in proximity. In this context, a probabilistic model is introduced for the availability of M2M devices. Thorough investigations are made on the TC, TC gains and trade-offs involved for both underlay and D2D-overlay modes. Using SG, the successful transmission probabilities for all associated links are derived to determine the TCs in these scenarios and present computable expressions for the TC gains achieved. Furthermore, an exposition is provided on how the density and transmit power of MTDs in the network affect the D2D TC performance. Results show that the deployment of MTD devices as relays improves the TC as compared to when only traditional RNs are used. Similarly, higher peak TCs are achieved at 23dB MTD transmit power. Overall, a lower transmit power (15dB) yields better performance. Thus, high MTD density can be leveraged to improve the D2D TC when so many D2D transmissions occur within the system and when these devices are farther apart. The literature on the performance analysis of energy harvesting cognitive radio networks focused on a dual and multi-hop secondary architecture. Also, the available literature on a multi-hop primary architecture was not studied in the context of radio frequency energy harvesting which makes the impact of a multi-hop primary network on the outage performance of a dual-hop energy harvesting CRN largely unknown. Thus, the second contribution in this thesis exploits SG and the advancements in wireless energy harvesting to develop a framework for the outage probability analysis of energy harvesting underlay CRNs. In this model, #-hop primary users are equipped with constant energy source while secondary users harvest energy from the transmissions of primary devices. The transmit power of secondary users is regulated to ensure it does not violate the target end-to-end OP constraint of the #-hop primary network. Potential relays that have harvested sufficient energy are eligible to relay data for other secondary users within the network. This model reveals the impact of the number of primary hops on the relay selection region and harvested energy (which is generally unknown). Also, an expression for the total outage probability which encapsulates the impact of N-primary hops is derived. The impacts of other relevant parameters on the outage probability are shown in detail. Results show that the multi-hop primary network reduces the secondary outage probability by regulating the number of transmitting energy harvesting relays within the network. Interference cancellation has long been known as an effective approach to reducing the impact of interference in wireless networks. However, the interplay between interference cancellation and energy harvesting and how both can be used within the same architecture to improve the outage performance in cognitive radio networks is unknown. This motivates the third contribution where interference cancellation is incorporated in the outage analysis of a Poisson distributed wireless energy harvesting cognitive relay network. Based on a predefined interference threshold, devices within the primary network are assumed to be able to cancel a fraction of the strongest interferers in the entire network. To achieve this, the coefficient of cancellation is adapted into the SG analysis to reduce the level of interference. The rationale is to further help the secondary network to meet up with the primary outage constraint by reducing some of the interference experienced by primary receivers. Analytical results show that this significantly reduces the secondary OP which in turn improves the network performance based on a set cancellation threshold and residual interference power. However, this is at the cost of reducing the energy harvesting success probability of the relays within the secondary network which depends on the primary density as interference from such devices would be cancelled.

Text FSKTM 2021 13 - IR.pdf Download (1MB)

Actions (login required)

View Item