On reliability in the performance analysis of cognitive radio networks

Satisfying the quality of service (QoS) requirements of users in the form of channel availability and service retainability within the resource limited environment has been a major problem in cognitive radio networks. In this connection, several research studies have been carried out in the literature to improve the QoS of users by proposing dynamic channel reservation algorithms. However, the studies have a number of limitations in their conceptual and mathematical modeling of channel availability and service retainability, which render their performance evaluation unreliable. In this paper, we address these limitations for leading to more realistic, reliable, and practically valid modeling. For conceptual modeling, we use connection availability in place of channel availability, motivated by the fact that the latter does not necessarily lead to a successful establishment of connection and, thus, is not a suitable performance indicator. For instance, obtaining a channel for transmission is of no avail if the intended receiver is inaccessible. Similarly, we consider service retainability with accessibility/inaccessibility of the intended receiver incorporated. For mathematical modeling, we use CTMC and the resultant closed from expressions to include all the required states of channel availability yet unsuccessful connection establishment. Additionally, we derive closed form equations for channel availability and service retainability that are in exact conformance with the CTMC model. Results confirm that considering the impact of receiver’s accessibility leads to performance difference for the channel availability and service retainability presented in the sate-of-the-art.