We consider an interference network with independent links, whose multi-antenna transmitters have access to an imperfect analog estimate of their local channels. Assuming that the receivers treat the interference as noise, we define the outage rate region as the set of rate-tuples that are achievable with a given probability and we characterize the boundary of the region for transmit beamforming. Our study shows that the Pareto-optimal beamforming vectors judiciously balance the desired signal power and the interference power based on the quality of the estimated channel state. Our analysis further reveals that, in contrast to the well-known results by Jorswieck, for the perfect channel side-information case, transmission at full power is not necessarily Pareto-optimal.

We study a multinode network, where a multi-antenna transmitter Tx communicates with its desired receiver Rx, whereas a cluster P ≡̂ {Px,n, n = 1, . .. ,N} of unintended nodes is disturbed by the Tx-Rx (TR) communication. To prevent severe performance degradation, we impose a constraint on the total interference that is inflicted at the nodes of P. The TR link contains a line-of-sight component, whereas the propagation environment for each Tx-Px,n link is shadowed. The Tx node is preprocessing the information sequence by means of a precoding matrix that is optimized to achieve the ergodic capacity under a constraint on the maximum admissible ergodic interference power, arriving on P. In this paper, we show that the optimum precoding strategy involves the transmission of a single stream over the precoding direction, i.e., the eigenvector of the precoding matrix, which corresponds to beamforming along the instantaneous direction of the TR-link channel. The solution of the remaining power allocation problem yields the optimal precoding matrix. For this setup, we provide an efficient stochastic characterization of the network, which allows us to obtain an analytical expression for the TR-link ergodic capacity; this problem has been previously open, even for the case of a single-antenna node Tx and a single-element set P. We complement the analysis by deriving the TR-link signal-to-noise ratio and the average bit error rate, which are associated with our transmission scheme. Numerical results corroborate the theoretical analysis and reveal an interplay between the network parameters and their impact on the TR-link performance.

A cognitive radio network with a multiple-input single-output secondary link and a multi-antenna primary receiver is considered. The secondary transmitter steers its transmission into the direction of its intended destination in order to maximize the received signal-to-noise ratio. Under this beam-forming strategy, the power allocation is optimized to achieve the ergodic capacity under the constraint that the long-term interference, caused at the primary receiver, will not exceed a predefined threshold. Assuming line-of-sight communication (Rician fading) for the secondary link and Rayleigh fading for the secondary-to-primary link channel, we derive the exact closed-form expression for the ergodic capacity. Numerical results corroborate theoretical findings and illustrate the manifold impact of the system parameters into its performance.

We consider a pair of secondary users that coexist, in a cognitive network, with multiple primary user pairs. The secondary link is supplied with partial network side information (NSI), which comprises message side information and partial channel side information (CSI), available in different levels at both the transmitter and the receiver of the cognitive link. The cognitive transceiver design has to obey predefined quality-of-service (QoS) criteria, that need to be maintained at the primary receivers, and at the same time properly handle the incoming interference from each primary transmitter in order to establish reliable communication. In this framework, we investigate the design and performance of the combined beamforming and orthogonal space-time block coding (BOSTBC) strategy, whose merits are well-documented, as a candidate transmission scheme for the secondary link. We study both aspects, QoS and interference, of the composite problem and characterize how they affect the beamformer design and the cognitive link performance, in the presence of partial NSI. Further, we propose a CSI quality-dependent model for the QoS criteria which yields an interesting trade-off between the cognitive link design and the primary QoS. Numerical results illustrate the system performance in this framework.

A multi-antenna cognitive radio network, with a single pair of primary users and a secondary broadcast channel, is considered. Under perfect channel state information (CSI), the rate-optimal strategy for the primary link is waterfilling, resulting in possibly unused dimensions. The secondary base station, supplied with perfect and global CSI, employs block diagonalization for interference-free message precoding and opportunistic interference alignment to avoid disturbing primary communication. In the absence of sufficient resources to serve all secondary users, a low complexity adaptive antenna selection strategy is proposed. This scheme constructs, in each time-slot, the set of active antennas by opportunistically reusing information and resources from the preceding selection round. Numerical results indicate a complexity reduction at a small rateloss penalty, in comparison with existing selection methods.

We consider a pair of secondary nodes (SU) coupled, in Z-topology, with multiple pairs of primary nodes (PU). The secondary (cognitive) transmitter is combining beamforming with orthogonal space-time block coding (BOSTBC) and operates under Quality-of-Service (QoS) constraints that must be guaranteed for the primary receivers (PURx). The cognitive link is designed assuming imperfect channel state information (CSI) for all links, available at the SU transmitter (SUTx). Under this premise we characterize the optimal design in terms of CSI quality and interference and evaluate their impact on the performance of BOSTBC transmission in underlay cognitive networks.