Thesis Keywords: Spectrum Sharing, Spectrum Management, 5G Networks, Full-Duplex, Device-to-Device Communications, Heterogenous Networks, Interference Management, Stochastic Geometry. Thesis Abstract: Opportunistic Spectrum Access has recently been the most desirable solution to efficiently optimize system performance of telecommunication systems since it plays a pivotal role in allowing multiple access technologies to use the available spectrum based on share-it or use-it basis. This has proven to be a viable solution to cope with the challenging problem of spectrum scarcity and also widely explored in 5G networks for the coexistence of multiple random access technologies with incumbents. In 5G networks, such secondary technology candidates like Device-to-Device (D2D) communications, and Licensed-Assisted Access are envisioned to opportunistically exploit spectrum opportunities and coexist with primary technologies like LTE, or WiFi. Moreover, Full Duplex (FD) technology is envisioned to play a significant role in 5G networks by allowing a user to transmit and receiver on the same frequency band at the cost of increased interference. However, recent advancements in the transceiver design and self-interference cancellation paved the way to benefit from the FD gains. Ultra-dense and random network models as envisioned in future networks especially in the urban scenario which requires average system performance over various deployment scenarios. In this context, the stochastic geometry is a powerful mathematical tool to capture and analyze the key properties of random points, such randomness is also depicted in the wireless networks. In this thesis, we present a stochastic geometry based model of a heterogeneous system where a cellular network allows FD-Enabled D2D network to opportunistically use its spectrum while ensuring protection for its transmission/reception through guard zone. Each user experiences interference from both (cellular and D2D) users within a cell. The main contributions and emphasis of this work are to explore the spectrum opportunities for secondary users by deriving their probability of successful transmissions, deciding the feasible mode of operation (half-duplex/full-duplex or silent), and incorporating the protection zone for primary users. We assess the overall system performance, analyze the impact of different access mechanisms and propose efficient mode selection for secondary users. We also analyze different coexistence scenarios and spectrum sharing frameworks in 5G networks. Moreover, the impact of key network configuration parameters is evaluated for different network abstractions to assess the feasibility of integrating key enabling technologies and also for different coexistence mechanisms. This thesis proposes an innovative FD enabled D2D cognitive setup and carefully studies the improvement in system performance and also the cost of these gains in 5G networks.