Published in IET Communications Received on 14th July 2010 Revised on 24th December 2010 doi: 10.1049/iet-com.2010.0642 In Special Issue: Cognitive Communications ISSN 1751-8628 Secondary spectrum access and cell-edge coverage in cognitive cellular networks A. Zarrebini-Esfahani M.R. Nakhai Centre for Telecommunications Research, King’s College London, Strand WC2R 2LS, UK E-mail: azar.zarrebini-esfahani@kcl.ac.uk Abstract: This study focuses on the problem of cell-edge user coverage in the context of cognitive radio networks operating within the vicinity of primary cell borders. Two strategies are introduced such that the primary cell-edge users get assisted by the cognitive base station (BS) to receive a consistent quality of service (QoS) because of their long distance from the primary BS. In return, the cognitive BS is rewarded by using the same spectrum that has already been allocated to the primary users’ link to serve a group of cognitive users. In the first strategy that we call cooperative, the cognitive BS relays the primary cell-edge users’ data, sent by the primary BS, through spatial multiplexing and beamforming, while transmitting towards its cognitive users. In the second strategy that we call soft interference shaping, the cognitive BS serves cognitive users as well as primary cell-edge users by spatial multiplexing and beamforming, while forming controlled nulls towards the primary users located outside but within the close vicinity of the cognitive cell border. This technique is done to avoid the interference towards the primary users surrounding the cognitive cells border. 1 Introduction Cognitive radio has been proposed as a promising technology for improving the utilisation efficiency of the radio spectrum [1]. In a cognitive system, the cognitive (secondary) user seeks to overcome the spectral shortage problem by using the primary users’ bandwidth without causing any interference [2, 3]. In a spectrum-sharing scenario, the secondary user can co-exist with the primary user all the time as long as the interference power received by the primary user is less than a threshold, which is determined by the quality of service (QoS) of the primary user [4–6]. In order to fully utilise the limited spectrum, the spectrum-sharing strategy between the primary and secondary users is an important issue. Two approaches to spectrum sharing have been addressed: spectrum overlay and spectrum underlay. In spectrum overlay paradigm, a secondary spectrum user is constrained to avoid interfering with an active primary spectrum user via spectrum sensing, adaptive detection and allocation of unused spectrum portions by the primary users. Then, the detected spectrum opportunities can be shared within a network of cognitive users [7, 8]. In spectrum undelay paradigm, a secondary spectrum user is restricted in terms of transmitting power level so that its inflicted interference on a receiving point in a primary network falls below the noise floor [9]. The power allocation and beamforming problems for multi- user systems have been widely studied over the past decade. On the basis of several early papers on optimal downlink beamforming and power control [10, 11], there has been a more recent work on signal to interference and noise ratio (SINR) balancing for designing downlink beamformers [12]. In [13] optimal downlink beamforming in a communication system has been formulated to minimise the transmission power subject to QoS constraints, using rank- constrained solutions. The transmit beamformer design for cognitive radio networks at the secondary base station (BS) has been used to control the interference level, because of the secondary transmission, at the primary users. By doing so, the signal- to-noise ratio of the secondary user can be improved, yet the interference power to the primary user is limited to a certain acceptable level. In [14], spatial diversity has been exploited in downlink to improve the channel capacity between secondary users, whereas imposing constraints on the secondary user transmit power and the primary user interference power. In [15], the author has considered the co-existence between cognitive users and a primary user, where an efficient transmit beamforming technique combined with user selection is proposed to maximise the downlink throughput and satisfy the SINR constraint as well as limit interference to the primary user. This paper focuses on downlink cognitive communication within a cellular network. We consider a cluster of primary cells sharing the same bandwidth. In addition, small cognitive cells are located within the vicinity of the primary cell borders. First, we introduce a cooperative strategy where the secondary BS relays data, received from the primary BS, to a primary cell-edge user, while transmitting data to the secondary users within the cognitive cell, through spatial multiplexing. Second, we propose soft interference-shaping strategy where the secondary BS can focus its radiation pattern along the direction of the users within the cognitive IET Commun., 2012, Vol. 6, Iss. 8, pp. 845–851 845 doi: 10.1049/iet-com.2010.0642 & The Institution of Engineering and Technology 2012 www.ietdl.org