TY - BOOK

T1 - Spectrum sharing systems for improving spectral efficiency in cognitive cellular network

AU - G C, Deepak

PY - 2017

Y1 - 2017

N2 - Since spectrum is the invisible infrastructure that powers the wireless communication, the demand has been exceptionally increasing in recent years after the implementation of 4G and immense data requirements of 5G due to the applications, such as Internet-of-Things (IoT). Therefore, the effective optimization of the use of spectrum is immediately needed than ever before. The spectrum sensing is the prerequisite for optimal resource allocation in cognitive radio networks (CRN). Therefore, the spectrum sensing in wireless system with lower latency requirements is proposed first. In such systems with high spatial density of the base stations and users/objects, spectrum sharing enables spectrum reuse across very small regions. The proposed method in this Thesis is a multi-channel cooperative spectrum sensing technique, in which an independent network of sensors, namely, spectrum monitoring network, detects the spectrum availability. The locally aggregated decision in each zone associated with the zone aggregator (ZA) location is then passed to a decision fusion centre (DFC). The secondary base station (SBS) accordingly allocates the available channels to secondary users to maximize the spectral efficiency. The function of the DFC is formulated as an optimization problem with the objective of maximizing the spectral efficiency. The optimal detection threshold is obtained for different cases with various spatial densities of ZAs and SBSs. It is further shown that the proposed method reduces the spectrum sensing latency and results in a higher spectrum efficiency. Furthermore, a novel power allocation scheme for multicell CRN is proposed where the subchannel power allocation is performed by incorporating network-wide primary system communication activity. A collaborative subchannel monitoring scheme is proposed to evaluate the aggregated subchannel activity index (ASAI) to indicate the activity levels of primary users. Two utility functions are then defined to characterize the spectral efficiency (SE) and energy efficiency (EE) as a function of ASAI to formulate a utility maximization problem. The optimal transmit power allocation is then obtained with the objective of maximizing the total utility at the SBS, subject to maximum SBS transmit power and collision probability constraint at the primary receivers. Since optimal EE and SE are two contradicting objectives to obtain the transmit power allocation, the design approach to handle both EE and SE as a function of common network parameter, i.e., ASAI, is provided which ultimately proves the quantitative insights on efficient system design. Extensive simulation results confirm the analytical results and indicate a significant improvement in sensing latency and accuracy and a significant gain against the benchmark models on the rate performance, despite the proposed methods perform with lower signalling overhead.

AB - Since spectrum is the invisible infrastructure that powers the wireless communication, the demand has been exceptionally increasing in recent years after the implementation of 4G and immense data requirements of 5G due to the applications, such as Internet-of-Things (IoT). Therefore, the effective optimization of the use of spectrum is immediately needed than ever before. The spectrum sensing is the prerequisite for optimal resource allocation in cognitive radio networks (CRN). Therefore, the spectrum sensing in wireless system with lower latency requirements is proposed first. In such systems with high spatial density of the base stations and users/objects, spectrum sharing enables spectrum reuse across very small regions. The proposed method in this Thesis is a multi-channel cooperative spectrum sensing technique, in which an independent network of sensors, namely, spectrum monitoring network, detects the spectrum availability. The locally aggregated decision in each zone associated with the zone aggregator (ZA) location is then passed to a decision fusion centre (DFC). The secondary base station (SBS) accordingly allocates the available channels to secondary users to maximize the spectral efficiency. The function of the DFC is formulated as an optimization problem with the objective of maximizing the spectral efficiency. The optimal detection threshold is obtained for different cases with various spatial densities of ZAs and SBSs. It is further shown that the proposed method reduces the spectrum sensing latency and results in a higher spectrum efficiency. Furthermore, a novel power allocation scheme for multicell CRN is proposed where the subchannel power allocation is performed by incorporating network-wide primary system communication activity. A collaborative subchannel monitoring scheme is proposed to evaluate the aggregated subchannel activity index (ASAI) to indicate the activity levels of primary users. Two utility functions are then defined to characterize the spectral efficiency (SE) and energy efficiency (EE) as a function of ASAI to formulate a utility maximization problem. The optimal transmit power allocation is then obtained with the objective of maximizing the total utility at the SBS, subject to maximum SBS transmit power and collision probability constraint at the primary receivers. Since optimal EE and SE are two contradicting objectives to obtain the transmit power allocation, the design approach to handle both EE and SE as a function of common network parameter, i.e., ASAI, is provided which ultimately proves the quantitative insights on efficient system design. Extensive simulation results confirm the analytical results and indicate a significant improvement in sensing latency and accuracy and a significant gain against the benchmark models on the rate performance, despite the proposed methods perform with lower signalling overhead.

KW - spectrum sensing

KW - COGNITIVE RADIO NETWORKS

KW - ENERGY EFFICIENCY

KW - Spectral energy

U2 - 10.17635/lancaster/thesis/400

DO - 10.17635/lancaster/thesis/400

M3 - Doctoral Thesis

PB - Lancaster University

ER -