| Abstract— Research and discussion into more efficient utilization of frequency spectrum has led to a paradigm shift in the allocation of spectral resources. Secondary usage of licensed spectrum is being considered as a viable solution to the problem. In such a system, secondary users share spectral resources with licensed users. In the model, we assume the operation of licensed/primary system is completely unaffected by the introduction of a secondary system. Licensed users have priority in transmission over secondary users. The primary system controller will assign any frequency slot unoccupied by other primary users to the newly arrived primary user. The secondary users share the spectral resources with licensed primary users and need to sense the spectrum to look for an unused frequency slot. During its transmission, if it detects the presence of a primary user intending to use its frequency for transmission, the connection will be dropped. Using a three-dimension Markov Chain model, this paper presents an analytical analysis of the potential outage performance of secondary users for a given primary traffic. Two important parameters that are indicative of the performance of the secondary system, namely, secondary call blocking probability and secondary call dropping probability (due to onset of licensed users) are computed using the model presented. Further, we show that by reserving some channels for primary user will help improve the throughput of the secondary system in fulfillment of certain call blocking and dropping requirements.
In this paper, we first present a Markov chain for modeling spectrum sharing with priority given to primary access. The model is a modification of that in [16] to enable analytical analysis. We then present the results of the analysis and show that the throughput of the secondary system is limited by either the call blocking or call dropping probability for certain Grade of Service (GoS) requirements. Therefore if we incorporate some admission control by reserving some channels for the primary system, by decreasing the call dropping probability, the throughput of the secondary system is actually improved. We then study their impact on call blocking and dropping probabilities.
|