Optimizing large-scale MIMO cellular downlink: Multiplexing, diversity, or interference nulling?

Abstract

A base-station (BS) equipped with multiple antennas can use its spatial dimensions in three ways: (1) serve multiple users to achieve a multiplexing gain, (2) provide diversity to its users, and/or (3) null interference at a chosen subset of out-of-cell users. The main contribution of this paper is to answer the following question: what is the optimal balance between the three competing benefits of multiplexing, diversity and interference nulling? We answer this question in the context of the downlink of a cellular network in which each user chooses its best serving BS, and requests nearby interfering BSs for interference nulling. BSs are equipped with a large number of antennas, serve multiple single-antenna users using zero-forcing beamforming and equal power assignment, and null interference at a subset of out-of-cell users. The remaining spatial dimensions provide transmit diversity. We assume perfect channel state information at the BSs and users. Utilizing tools from stochastic geometry, we show that, surprisingly, to maximize the per-BS ergodic sum rate, at the optimal allocation of spatial resources, interference nulling does not bring tangible benefit. A closeto-optimal strategy is to use none of the spatial resources for interference nulling, while reserving 60% of spatial resources for achieving multiplexing and the rest for providing diversity.

A Stochastic Analysis of Network MIMO Systems

“…Such possibility has been considered in the literature [23], [24]. The optimal use of spatial dimensions for interference nulling is considered in our related work [25] for the largescale MIMO system.…”

Optimizing the MIMO Cellular Downlink: Multiplexing, Diversity, or Interference Nulling?

“…Indeed, the recent focus of MIMO systems has been on BSs with a large, even an asymptotically large, number of antennas (massive MIMO systems). The availability of multiple antennas provides the flexibility to achieve multiple system objectives simultaneously: (1) to serve multiple users in the same time-frequency slot to achieve a spatial multiplexing gain, (2) to provide spatial diversity for these users, and, as often stated as a major advantage of massive MIMO systems, (3) to null interference at out-ofcell users. In fact, as often pointed out in the massive MIMO literature, in the limit as the number of BS antennas goes to infinity, transmit beamformers matched to the channel completely eliminate multiuser and intercell interference, leaving pilot contamination as the only limiting factor [2].…”

“…In fact, as often pointed out in the massive MIMO literature, in the limit as the number of BS antennas goes to infinity, transmit beamformers matched to the channel completely eliminate multiuser and intercell interference, leaving pilot contamination as the only limiting factor [2]. This paper focuses on the design of large-scale MIMO (LS-MIMO) downlink systems [3], where each BS is equipped with a large but finite number of antennas 1 . Specifically, we consider a scenario in which the number of users being served is comparable to or is a significant fraction of the number of BS antennas; in this case, the overall system does not necessarily operate in the so-called massive MIMO regime.…”

“…In fact, it is near-optimum, in sum-rate sense, to use a fraction of the available spatial dimensions for multiplexing (i.e., with M antennas at each BS, serve K < M users) and to use the remaining dimensions for spatial diversity, thereby leaving none of the spatial dimensions for intercell interference nulling. This is despite the fact that (a) the BSs 1 While we do not make any assumptions as to the number of BS antennas, our question is most relevant to a system with a sufficiently large number of antennas such that the three uses, multiplexing, diversity and interference nulling, are possible simultaneously. are densely deployed and the overall network is interference limited, and (b) the analysis does not account for the cost of channel estimation.…”

Performance analysis of partial interference cancellation in multi-antenna UDNs

“…Ultra-dense networks (UDNs), i.e., dense and massive deployment of small cells that have a wired/wireless backhaul connection, represent a core element of emerging 5th generation (5G) wireless systems, which is expected to cope with the proliferation of wireless devices and the ever-growing demand for high data rates. Due to such massive and dense deployments, modern cellular networks are becoming interferencelimited, thus motivating the use of interference management techniques [1], [2] such as successive interference cancellation (SIC) at both base stations (BSs) and user terminals. SIC receivers attempt to decode and subtract/cancel interfering signals in the order of decreasing interference power level and are characterized by a considerable complexity (see, e.g., [3], [4]) compared to linear processing.…”