Cognitive radio (CR) technology has been considered one of the most promising solutions to make
the best potential use of the scarce spectrum resources in next generation wireless communication.
Because of the stringent limitations on the utilization of the radio resources in a cognitive radio network
(CRN), the available resources are usually scarce and dynamic. Developing appropriate mechanism to
construct allocation protocols for CRN that can assign the resources in a fair and efficient manner to
diverse users with different features is essential to achieve the promise of the CR technology in reality
and worth studying. “The mechanism for resource allocation protocol design in CRN” describes the
topic of this research work and the contents of this thesis are organized around it.
Before investigating how to design appropriate mechanism for CR resource allocation (RA) protocols,
the model of the radio resource and the model of how a CR system utilizes the resource have to
be clarified. In this thesis, by a comprehensive study of the relevant literatures, the modeling of
critical components of a CR system and the theory behind it are identified and investigated. The
existing protocols are also analyzed to address the limitations and gaps in the design of allocation
protocols. An important feature, if not the most, that the allocation protocol for the CRN should possess is the
ability to adjust the strategy of RA when the circumstances that affect the data transmission of the CR
system varies, such as the activity of the license users or the variety of CR users’ demands. According
to such considerations, the mechanism, namely distribution probability matrix (DPM), is designed
to guide and describe the channel allocation protocol quantitatively. To testify how this mechanism
works, the protocols based on the DPM, together with reference protocols in a multichannel CRN, are
analyzed by a queuing model. The numerical results shows the flexibility and adaptability that the
protocols based on the DPM can achieve.
Another important function of the DPM mechanism is the potential to describe complex protocols with
specific principles and to help design protocols aimed at specific objectives. To demonstrate this, a
maximum throughput (MT) protocol that is aimed at maximizing the overall throughput of the CR
system in a multi-user multi-channel scenario is structured using the DPM mechanism. The existing
reference protocols, including maximum rate and random allocation, are described under the DPM
mechanism as well as the MT protocol. All the protocols are implemented in a multiuser multichannel
CRN and analyzed by a queueing analytical framework proposed by the author that is capable of
acquiring the performance metrics of each CR user independently with efficiency. Numerical results
show that the MT protocol is able to outperform the existing protocol in terms of the overall throughput
of the CRN.
The last problem identified and addressed in this thesis is to investigate the potential of specific
performance metrics that a CR system can achieve under certain conditions with the help of the DPM
mechanism. An optimization problem that is aimed at maximizing the overall weighted throughput of
a multi-user multi-channel CRN under the delay constraint is formulated and solved using the DPM
mechanism. The key factors that affect the optimal weighted throughput and the optimal allocations
are investigated through the numerical results obtained by the proposed queueing analytical framework.
The advantages of the DPM mechanism in optimizing the performance of CRN is also revealed by the
formulation and solution of the optimization problem.
The RA protocol is the essential kernel to address the RA problem in the CRN context. This thesis, by
providing an objective and effective mechanism that works as a universal tool to describe and evaluate
the RA protocols, succeeds in making effective contribution to addressing the RA problem in the CRN
and further, helping the CR technology realize its meaningful promises.