Abstract:
Cognitive radio (CR) is considered to enhance spectrum utilization efficiency by dy- namically sharing the spectrum between licensed/ primary users (PUs) and unli- censed/secondary users (SUs). This is achieved by permitting SUs opportunistic access to the white spaces within PUs spectrum while controlling the interference to PUs. This optimizes the use of available radio-frequency (RF) spectrum while min- imizing interference to other users. Orthogonal Frequency Division Multiple Access (OFDMA) is widely recognized as an ideal air interface for the CR system due to its flexibility in allocating radio resource among the SUs, which is the prerequisite for the CR system to acquire high throughput. Which means, the technology is recognized as an attractive modulation technique for CR due to its spectrum shaping flexibil- ity, adaptivity in allocating vacant radio resources, and capability of analyzing the spectral activities of PUs. In cognitive OFDMA systems, multiple cognitive radios are considered to compete for multiple subcarriers/subchannels. Most of the prior research efforts on cognitive OFDMA systems are on sensing subchannel, channel allocation, subchannel and transmit power allocation, transmit power allocation etc. Most of the approaches on subchannel and transmit power allocation are for downlink systems and based on maximizing capacity criterion which results on full utilization of power budget and hence, incur huge energy consumption. Recently, resource alloca- tion optimization frameworks based on minimizing energy efficiency (unit joule/bits) or maximizing energy efficiency (unit bits/joule) have got attention. However, energy efficiency based approaches are nonconvex and hence, solutions are near optimal. In this thesis, unlike to prior works, Specific Quality of Service (QoS) constrained power
xxi
xxii
minimization criterion based resource allocation convex optimization framework with transmit power as decision variable for cognitive OFDMA systems is proposed and studied in details.
At first, an uplink cognitive OFDMA system of multiple secondary users and multiple free channels is considered where, users employ orthogonal multiple access scheme. Two resource allocation optimization frameworks namely framework-I and II are pro- posed. Framework-I aims at designing a resource allocation convex optimization framework to determine the optimal transmit power with a goal of minimizing to- tal transmit power and constrained by signal to noise ratio (SNR) thresholds as a measure of QoS. Whereas, framework-II aims at designing a resource allocation con- vex optimization framework to determine the optimal transmit power with a goal of minimizing total transmit power and constrained by minimum rate requirements as a measure of QoS. Later a downlink cognitive OFDMA system is considered. Similar frameworks (framework-I and II) are developed to determine optimal allocation of transmit power. Numerical results of both uplink and downlink cognitive OFDMA systems reveal that, both of the proposed frameworks are very much successful in terms of utilization of power budget of users and Energy Efficiency compared to conventional capacity maximization based resource allocation approaches.
