Cognitive Radio Networks With Probabilistic Relaying: Stable Throughput and Delay Tradeoffs

Abstract

This paper studies fundamental throughput and delay tradeoffs in cognitive radio systems with cooperative secondary users. We focus on randomized cooperative policies, whereby the secondary user (SU) serves either its own queue or the primary users (PU) relayed packets queue with certain service probability. The proposed policy opens room for trading the PU delay for enhanced SU delay, and vice versa, depending on the application QoS requirements. Towards this objective, the systems stable throughput region is characterized. Furthermore, the moment generating function approach is employed and generalized for our system to derive closed-form expressions for the average packet delay for both users. The accuracy of these expressions is validated through simulations. Analytical and simulation results reveal that the service probability can steer the system into prioritizing PUs traffic at the expense of SUs QoS, or vice versa, independently from the admission probability. Alternatively, the ability of the admission probability to control the throughput and delay at the PU or the SU depends on the selected value for the service probability as well as the channel conditions. Finally, it is shown how the service and admission probabilities could be used to achieve the desired QoS level to both PU and SU.Index Terms-Cognitive relaying, moment generating function, stable throughput region, average delay simulations. 0090-6778 (c)

On optimizing cooperative cognitive user performance under primary QoS constraints

“…1. We compare the proposed policy to a baseline cooperation policy (BL), coined "unconstrained partial cooperation policy" [9]. In BL, the SU probabilistically cooperates with the PU in delivering its packets in Q sp with no constraint on the delay encountered by the PU packets.…”

“…Similar to [7], [9], [10], the SU is assumed to perfectly know the state of the PU of whether it is backlogged or idle and, hence, there is no interference in our system. A possible approach to accomplish this objective is via sensing the communication channel by the SU in order to detect the time slots at which the PU is idle.…”

Stability and Dynamic Control of Underlay Mobile Edge Networks

“…In addition to above work, our results in this paper are also related to those on resource allocation and opportunistic scheduling for underlay cognitive radio networks (CRNs) [25], [26], [27], [28], [29], [30], [31], [32], [33], [34]. The authors in [25] investigated the optimal power control and the resulting throughput scaling laws for underlay CRNs under average interference power constraints at the primary users.…”

On Stable Throughput of Cognitive Radio Networks With Cooperating Secondary Users

“…9, we compare average delay performance of ARC with cooperative relaying (CR) protocol in [18] under perfect sensing case for L = 1. Average PU packet delay in CR has been derived in [18,Eq. (14), (16), (17)].…”