Besides enabling an enhanced mobile broadband access, fifth-generation (5G) wireless mobile networks are envisioned to support the connectivity of massive, heterogeneous Internet of Things (IoT) devices. Connecting these devices through 5G systems and providing them with their needed data rates require huge amounts of spectrum and power resources, thus calling for the development and design of innovative, dynamic resource identification, access and sharing methods that make effective use of these limited resources. This thesis focuses specifically on wideband spectrum sensing, and presents innovative techniques that enable efficient identification and recovery of unused spectrum opportunities in wideband dynamic spectrum access. Recent research efforts have focused on leveraging compressive sampling (CS) theory to enable wideband spectrum sensing recovery at sub-Nyquist rates. However, these approaches suffer from the following shortcomings. First, they consider homogenous wideband spectrum, where all
bands are assumed to have similar primary users (PU)s traffic characteristics whereas in practice, the wideband spectrum occupancy is heterogeneous. Second, the number of measurements that receiver hardware designs are able to perform is practically way smaller than the number of measurements required by the CS-based sensing approaches. Third, the number of measurements required by the CS-based sensing approaches depends on the number of occupied bands (i.e., sparsity level), which is often unknown
in advance and changes over time. Forth, current wideband spectrum databases suffer from scalability issues in that they incur lots of sensing overhead. This thesis proposes a set of new, complementary techniques that overcome these aforementioned challenges. More specifically, in this thesis,
1. We design efficient spectrum occupancy information recovery techniques for heterogeneous wideband spectrum access. Our proposed techniques exploit the block-like structure of spectrum occupancy behavior observed in wideband spectrum access networks to enable the development of compressed spectrum sensing algorithms. Our proposed spectrum sensing algorithms achieve more stable spectrum information
recovery than that achieved by existing approaches.
2. We develop distributed CS-based spectrum sensing techniques for cooperative wideband spectrum access that require lesser measurements while overcoming time-variability of spectrum occupancy and addressing hidden terminal challenges. Also, we propose non-uniform sensing matrices design that exploits the heterogeneity in the wideband spectrum access to further improve the spectrum sensing recovery
3. We develop scalable spectrum occupancy information recovery techniques for database-driven wideband spectrum access networks. The novelty of our developed techniques lies in combining the merit of compressive sampling theory with that of low-rank matrix theory to enable scalable and accurate wideband spectrum occupancy recovery at low sensing overhead.
4. We propose joint data and energy transfer optimization frameworks for powering mobile cellular devices through RF energy harvesting. Our proposed framework accounts for both the consumed power at the base station and the battery power available at the end users to ensure that end users achieve their required data rates with as little battery power consumption as possible. We also analytically derive closed-form expressions of the optimal power allocations required for meeting the data rate requirements of the downlink and uplink communications between the base station and its mobile users.