Guest Editorial Multiuser Detection for Advanced Communication Systems and Networks

detection of multiuser signals, coupled with associated challenges in achieving this potential at affordable receiver complexity, has motivated significant amount of research to be carried out in the area of multiuser detection (MUD) in the past two decades. Much of the early research in this important area has been centered around systems employing code division multiple access (CDMA) promising capacity improvement in terms of the number of simultaneous users supported in the system. The optimum MUD complexity, which is exponential in the number of users, has inspired a considerable effort toward the development of low-complexity, suboptimal alternatives capable of resolving the detrimental effects of multiple access interference. Interference cancellation strategies have received particular attention, due to their competitive performance at low complexity and simple modular structure. Their performances, however, are still far from the optimum maximum-likelihood (ML) performance. Iterative methods based on soft-decision cancellation have been shown to achieve near-ML performance. Since most practical communication systems use coding, iterative multiuser decoding of coded CDMA signals has received considerable research attention , and so has the topic of joint multiuser channel estimation and decoding. More recent approaches to low-complexity MUD involve application of techniques from, e.g., belief propagation, Bayesian inference, neural networks, Markov-Chain Monte-Carlo methods, and probabilistic data association, just to name a few. This trend is reflected in several of the papers in this special issue. Several detectors based on the above techniques have been shown to achieve an average per-bit complexity that is linear in the number of users, while achieving near-optimal performance in a large system setting. We see an emerging trend in MUD research: achieving near-optimal performance in linear complexity and the establishment of the fundamental connections between the various approaches which achieve this complexity. Another interesting feature of current multiuser detection research is its immediate relevance/applicability to low-complexity multiple-input multiple-output (MIMO) detection. Multiuser systems and MIMO systems are both described by a vector-matrix channel model with the same structural format. In case of a multiuser system the channel matrix is defined by the normalized cross-correlations between the signature sequences of the active users, whereas the channel matrix in a MIMO system is defined by the spatial signatures between the receive and transmit antennas. Thus, it is no surprise that this special issue has several papers that propose and investigate detection methods for CDMA and MIMO systems without and with transmit pre-processing. Another new multiuser research avenue …