1350 IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, VOL. 63, NO. 3, MARCH 2014 Delay Performance of a Broadcast Spectrum Sharing Network in Nakagami-m Fading Fahd Ahmed Khan, Student Member, IEEE, Kamel Tourki, Member, IEEE, Mohamed-Slim Alouini, Fellow, IEEE, and Khalid A. Qaraqe, Senior Member, IEEE Abstract—In this paper, we analyze the delay performance of a point-to-multipoint secondary network (P2M-SN), which is concurrently sharing the spectrum with a point-to-multipoint pri- mary network (P2M-PN). The channel is assumed to be indepen- dent but not identically distributed (i.n.i.d.) and has Nakagami-m fading. A constraint on the peak transmit power of the secondary-user transmitter (SU-Tx) is considered, in addition to the peak interference power constraint. The SU-Tx is assumed to be equipped with a buffer and is modeled using the M/G/1 queueing model. The performance of this system is analyzed for two scenarios: 1) P2M-SN does not experience interference from the primary network (denoted by P2M-SN-NI), and 2) P2M-SN does experience interference from the primary network (denoted by P2M-SN-WI). The performance of both P2M-SN-NI and P2M-SN-WI is analyzed in terms of the packet transmission time, and the closed-form cumulative density function (cdf) of the packet transmission time is derived for both scenarios. Furthermore, by utilizing the concept of timeout, an exact closed-form expression for the outage probability of the P2M-SN-NI is obtained. In addi- tion, an accurate approximation for the outage probability of the P2M-SN-WI is also derived. Furthermore, for the P2M-SN-NI, the analytic expressions for the total average waiting time (TAW-time) of packets and the average number of packets waiting in the buffer of the SU-Tx are also derived. Numerical simulations are also performed to validate the derived analytical results. Index Terms—Cognitive radio, delay performance analysis, multi-user, outage performance, queuing theory, underlay spec- trum sharing. I. I NTRODUCTION T HE DEMAND for high-data-rate wireless communication services is growing. Achieving a high data rate requires more wireless spectra. However, the wireless spectrum has now become scarce as most of it has already been allocated for various services. Recent measurement studies have shown that Manuscript received July 17, 2012; accepted July 3, 2013. Date of publica- tion July 17, 2013; date of current version March 14, 2014. This work was supported in part by King Abdullah University of Science and Technology and in part by the Qatar National Research Fund through National Priorities Research Program under Grant NPRP 5-250-2-087. This paper was presented in part at the IEEE Symposium on New Frontiers in Dynamic Spectrum Access Networks (DySPAN 2012), Bellevue, WA, USA, October 16–19, 2012. The review of this paper was coordinated by Dr. D. Zhao. F. A. Khan and M.-S. Alouini are with the Computer, Electrical, and Mathematical Sciences and Engineering Division, King Abdullah Univer- sity of Science and Technology, Thuwal 23955-6900, Saudi Arabia (e-mail: fahd.khan@kaust.edu.sa; slim.alouini@kaust.edu.sa). K. Tourki and K. A. Qaraqe are with the Electrical and Computer Engi- neering Program, Texas A&M University at Qatar, 23874 Doha, Qatar (e-mail: kamel.tourki@qatar.tamu.edu; khalid.qaraqe@qatar.tamu.edu). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TVT.2013.2273621 the wireless spectrum is greatly underutilized [2]. By utilizing the spectrum efficiently, it is possible to fulfill the demand for the increasing number of wireless services. As a conse- quence, cognitive radio has been proposed to improve the utilization of the spectrum [3], [4]. In cognitive radio, the spectrum utilization is improved through spectrum sharing in which the primary networks share its spectrum with a secondary network [4]. Various protocols have been proposed for spectrum sharing. One such protocol is where the secondary network uses the spectrum only when the primary network is not using the spectrum. This approach requires robust spectrum sensing algorithms that sense the spectrum perfectly (for details, see [5] and references therein). However, ideal spectrum sensing cannot be achieved in prac- tice, and in the case of missed detection, the primary network experiences severe interference. Another approach of spec- trum sharing is the underlay approach in which the secondary network is allowed to transmit concurrently with a primary network using the spectrum of the primary network if it does not cause harmful interference to the primary receiver. Thus, by limiting the interference from the secondary network, an acceptable level of performance of the primary network can be guaranteed, and the secondary network can also communicate and improve the utilization of the spectrum. In addition, it is essential that the secondary network satisfies a certain quality-of-service (QoS) requirement. The demand for delay-sensitive wireless services, such as voice over Internet Protocol services and video streaming services, is increasing [6]. These services are required to satisfy a delay QoS. How- ever, due to the mobility of the user and the various impairments caused by the wireless channel, such as multipath fading and shadowing, this delay QoS is not always satisfied, and in the cognitive setting, satisfying the delay QoS becomes even more challenging due to the spectrum sharing constraints. Therefore, it is essential to characterize the delay QoS performance of a cognitive network and use this characterization to improve the network performance. Recently, much research is being done on cross-layer design and optimization of delay-sensitive cognitive networks. The concept of effective capacity that was proposed in 2003 has been utilized to characterize and optimize the performance of the cognitive network. Effective capacity can be interpreted as the maximum constant arrival rate that can be provided by the channel while the delay QoS requirement is satisfied [7]. An optimal power-and-rate-allocation scheme to maximize the effective capacity of the underlay cognitive network un- der average interference constraint was proposed in [8] and 0018-9545 © 2013 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.