MIMO processing plays a central part towards the recent increase in spectral and energy efficiencies of wireless networks. MIMO has grown beyond the original point-to-point channel and nowadays refers to a diverse range of centralized and distributed deployments. The fundamental bottleneck towards enormous spectral and energy efficiency benefits in multiuser MIMO networks lies in a huge demand for accurate channel state information at the transmitter (CSIT). This has become increasingly difficult to satisfy due to the increasing number of antennas and access points in next generation wireless networks relying on dense heterogeneous networks and transmitters equipped with a large number of antennas. CSIT inaccuracy results in a multi-user interference problem that is the primary bottleneck of MIMO wireless networks. Looking backward, the problem has been to strive to apply techniques designed for perfect CSIT to scenarios with imperfect CSIT. In this paper, we depart from this conventional approach and introduce the readers to a promising strategy based on rate-splitting. Rate-splitting relies on the transmission of common and private messages and is shown to provide significant benefits in terms of spectral and energy efficiencies, reliability and CSI feedback overhead reduction over conventional strategies used in LTE-A and exclusively relying on private message transmissions. Open problems, impact on standard specifications and operational challenges are also discussed. IntroductionPromising approaches for 5G consist in densifying the network by adding more antennas in a distributed or co-localized manner. A distributed deployment leads to dense homogeneous/heterogeneous networks where the widely recognized bottleneck is interference. Interference management relying on multi-point cooperation have drawn a lot of attention in industry (i.e. CoMP in LTE-A [1]) and academia. Co-localized deployment leads to massive MIMO (i.e. FD-MIMO in LTE-A).Although appealing in their concept, those aforementioned MIMO techniques are hampered by several practical factors. Among these, the acquisition of accurate CSI knowledge at the transmitter (CSIT) is the major challenge. The availability of accurate CSIT is crucial for Downlink (DL) multi-user MIMO wireless networks. The beamforming and interference nulling performance heavily depends on the channel estimation accuracy. Unfortunately, pilot reuse tends to impair channel estimation in TDD and a significant feedback overhead is required to guarantee sufficient feedback accuracy in FDD due to the large number of antennas. Delay and inaccurate calibrations of the RF chains also contribute to making the CSIT inaccurate. CSIT inaccuracy results in a multi-user interference and link adaptation problem that is the primary bottleneck of MIMO wireless networks, as highlighted e.g. in [2] for for CoMP.Looking backward, the problem has been to strive to apply techniques designed for perfect CSIT to scenarios with imperfect CSIT. Following the same path will only increase the ...

To enhance the multiplexing gain of two-receiver Multiple-Input-Single-Output Broadcast Channel with imperfect channel state information at the transmitter (CSIT), a class of Rate-Splitting (RS) approaches has been proposed recently, which divides one receiver's message into a common and a private part, and superposes the common message on top of Zero-Forcing precoded private messages. In this paper, with quantized CSIT, we study the ergodic sum rate of two schemes, namely RS-S and RS-ST, where the common message(s) are transmitted via a space and space-time design, respectively. Firstly, we upperbound the sum rate loss incurred by each scheme relative to Zero-Forcing Beamforming (ZFBF) with perfect CSIT. Secondly, we show that, to maintain a constant sum rate loss, RS-S scheme enables a feedback overhead reduction over ZFBF with quantized CSIT. Such reduction scales logarithmically with the constant rate loss at high Signal-to-Noise-Ratio (SNR). We also find that, compared to RS-S scheme, RS-ST scheme offers a further feedback overhead reduction that scales with the discrepancy between the feedback overhead employed by the two receivers when there are alternating receiver-specific feedback qualities. Finally, simulation results show that both schemes offer a significant SNR gain over conventional single-user/multiuser mode switching when the feedback overhead is fixed.The authors are with the Communication and Signal Processing

We focus on a two-receiver Multiple-Input-Multiple-Output (MIMO) Broadcast Channel (BC) and Interference Channel (IC) with an arbitrary number of antennas at each node. We assume an imperfect knowledge of local Channel State Information at the Transmitters, whose error decays with the Signal-to-Noise-Ratio. With such configuration, we characterize the achievable Degrees-of-Freedom (DoF) regions in both BC and IC, by proposing a Rate-Splitting (RS) approach, which divides each receiver's message into a common part and a private part. Compared to the RS scheme designed for the symmetric MIMO case, the novelties of the proposed block lie in 1) delivering additional non-ZF-precoded private symbols to the receiver with the greater number of antennas, and 2) a Space-Time implementation. These features provide more flexibilities in balancing the common-message-decodabilities at the two receivers, and fully exploit asymmetric antenna arrays. Besides, in IC, we modify the power allocation designed for the asymmetric BC based on the signal space where the two transmitted signals interfere with each other. We also derive an outer-bound for the DoF regions and show that the proposed achievable DoF regions are optimal under some antenna configurations and CSIT qualities.

Recently, the Degrees-of-Freedom (DoF) region of multiple-input-single-output (MISO) networks with imperfect channel state information at the transmitter (CSIT) has attracted significant attentions.An achievable scheme is known as rate-splitting (RS) that integrates common-message-multicasting and private-message-unicasting. In this paper, focusing on the general K-cell MISO IC where the CSIT of each interference link has an arbitrary quality of imperfectness, we firstly identify the DoF region achieved by RS. Secondly, we introduce a novel scheme, so called Topological RS (TRS), whose novelties compared to RS lie in a multi-layer structure and transmitting multiple common messages to be decoded by groups of users rather than all users. The design of TRS is motivated by a novel interpretation of the K-cell IC with imperfect CSIT as a weighted-sum of a series of partially connected networks. We show that the DoF region achieved by TRS covers that achieved by RS. Also, we find the maximal sum DoF achieved by TRS via hypergraph fractional packing, which yields the best sum DoF so far. Lastly, for a realistic scenario where each user is connected to three dominant transmitters, we identify the sufficient condition where TRS strictly outperforms conventional schemes.