Research on the Lifetime Estimation of Secondary Connections Using the PRP M/G/1 Queuing Network Model

Abstract

Allocated Spectrum band has always been a limited resource; being later a more challenging issue due to the growth of user demands. The past decade is marked by important changes in spectrum access through especially cognitive radio technology. Here, spectrum access issue can be studied in all the different aspects of cognitive capabilities. This paper focuses on spectrum handoff in spectrum mobility. We will then study secondary connections behaviors after multiple interruptions providing from multiple secondary connections. In the IEEE 802.22 standards [, two Spectrum Handoff Sequences have been defined to characterize Secondary Connections behaviors after each Primary Connections interruption. These sequences are known as always-leaving and always-staying sequences. A recent analysis uses the extended data delivery time metric to analyze these Spectrum Handoff Sequences. It shows the exponential value of connections service time in the first sequence (always-leaving sequence) and in the other (always-staying sequence), the lack of fairness due to the acquisition of channels low-priority queue by the ongoing secondary connection (i.e. the secondary connection actually being served in the low-priority queue). Weve also noticed that the Extended Data Delivery Time is a good but limited metric because of the lack of prevision, which is a quality needed in cognitive radio networks, especially when implementing spectrum handoff sequences. Our present study aims to analyze these limitations. In what follows, we use the PRP M/G/1 queuing model in cognitive radio network to evaluate latency-sensitive traffic of Secondary Connections through a novel performance metric that we name Lifetime of Secondary Connections.