Kianoush Hosseini

Massive MIMO and small cells: How to densify heterogeneous networks

We propose a time division duplex (TDD) based network architecture where a macrocell tier with a “massive” multiple-input multiple-output (MIMO) base station (BS) is overlaid with a dense tier of small cells (SCs). In this context, the TDD protocol and the resulting channel reciprocity have two compelling advantages. First, a large number of BS antennas can be deployed without incurring a prohibitive overhead for channel training. Second, the BS can estimate the interference covariance matrix from the SC tier which can be leveraged for downlink precoding. In particular, the BS designs its precoding vectors to transmit independent data streams to its users while being orthogonal to the subspace spanned by the strongest interference directions; thereby minimizing the sum interference imposed on the SCs. In other words, the BS “sacrifices” some of its antennas for interference cancellation while the TDD protocol allows for an implicit coordination across both tiers. Simulation results suggest that, given a sufficiently large number of BS antennas, the proposed scheme can significantly improve the sum-rate of the SC tier at the price of a small macro performance loss.

Large-scale MIMO versus network MIMO for multicell interference mitigation

This paper compares two important downlink multicell interference mitigation techniques, namely, large-scale (LS) multiple-input multiple-output (MIMO) and network MIMO. We consider a cooperative wireless cellular system operating in time-division duplex (TDD) mode, wherein each cooperating cluster includes B base-stations (BSs), each equipped with multiple antennas and scheduling K single-antenna users. In an LS-MIMO system, each BS employs BM antennas not only to serve its scheduled users, but also to null out interference caused to the other users within the cooperating cluster using zero-forcing (ZF) beamforming. In a network MIMO system, each BS is equipped with only M antennas, but interference cancellation is realized by data and channel state information exchange over the backhaul links and joint downlink transmission using ZF beamforming. Both systems are able to completely eliminate intra-cluster interference and to provide the same number of spatial degrees of freedom per user. Assuming the uplink-downlink channel reciprocity provided by TDD, both systems are subject to identical channel acquisition overhead during the uplink pilot transmission stage. Further, the available sum power at each cluster is fixed and assumed to be equally distributed across the downlink beams in both systems. Building upon the channel distribution functions and using tools from stochastic ordering, this paper shows, however, that from a performance point of view, users experience better quality of service, averaged over small-scale fading, under an LS-MIMO system than a network MIMO system. Numerical simulations for a multicell network reveal that this conclusion also holds true with regularized ZF beamforming scheme. Hence, given the likely lower cost of adding excess number of antennas at each BS, LS-MIMO could be the preferred route toward interference mitigation in cellular networks .

Distributed clustering and interference management in two-tier networks

Employing centralized resource management schemes is generally infeasible in large-scale networks. The deployment of heterogeneous Femtocell Access Points (FAPs) over the cellular licensed spectrum is therefore challenging. In particular, the resulting inter-node interference inhibits the network performance. In this paper, we design a hierarchical, distributed, interference management scheme that exploits the benefits of clustering. First, in order to reduce the cross-tier interference, each FAP independently identifies vacant subbands for potential transmission. Then, by exchanging some simple messages with its immediate neighbors in an iterative fashion, coalition clusters are formed. Given the small population of each group, centralized resource management is subsequently performed to avoid intra-cluster interference. Different clusters, however, may still share a fraction of common idle channels, which degrades system performance. Therefore, this paper further considers inter-cluster interference management to determine the set of privileged FAPs that can share a subband via solving a binary power control optimization problem. While the optimal solution requires prohibitive complexity, this paper provides tight bounds on the sum rate of the binary power control problem. The simulation results show that, in a high interference regime, inter-cluster coordination provides a significant performance improvement compared to the case of no coordination.

A Stochastic Analysis of Network MIMO Systems

This paper quantifies the benefits and limitations of cooperative communications by providing a statistical analysis of the downlink in network multiple-input multiple-output (MIMO) systems. We consider an idealized model where the multiple-antenna base-stations (BSs) are distributed according to a homogeneous Poisson point process and cooperate by forming disjoint clusters. We assume that perfect channel state information is available at the cooperating BSs without any overhead. Multiple single-antenna users are served using zero-forcing beamforming with equal power allocation across the beams. For such a system, we obtain tractable, but accurate, approximations of the signal power and inter-cluster interference power distributions and derive a computationally efficient expression for the achievable per-BS ergodic sum rate using tools from stochastic geometry. This expression allows us to obtain the optimal loading factor, i.e., the ratio between the number of scheduled users and the number of BS antennas, that maximizes the per-BS ergodic sum rate. Further, it allows us to quantify the performance improvement of network MIMO systems as a function of the cooperating cluster size. We show that to perform zero-forcing across the distributed set of BSs within the cluster, the network MIMO system introduces a penalty in received signal power. Along with the inevitable out-of-cluster interference, we show that the per-BS ergodic sum rate of a network MIMO system does not approach that of an isolated cell even at unrealistically large cluster sizes. Nevertheless, network MIMO does provide significant rate improvement as compared to uncoordinated single-cell processing even at relatively modest cluster sizes.

Relay Selection and Max-Min Resource Allocation for Multi-Source OFDM-Based Mesh Networks

We consider a multi-source mesh network of static access points wherein sources use decode-and-forward to cooperate with each other. All transmissions use orthogonal frequency division multiplexing (OFDM). Our objective is to maximize the minimum achievable rate across all flows. We find a tight upper bound on the performance of the subcarrier-based cooperation and show that selecting a single relay for each subcarrier is optimal for almost all subcarriers. The solution to the related optimization problem simultaneously solves the relay, power, and subcarrier assignment problems. Second, unlike previous works, we also consider relay selection for the entire OFDM block. This addresses the fact that, in addition to the synchronization problems caused, it is likely impractical for a relay to only decode a subset of subcarriers. We propose three selection-based cooperation schemes to relay the entire OFDM block with varying complexity. Simulation results show that under the COST-231 channel model, the performance of the simplest scheme almost exactly tracks that of an exhaustive search.

Comprehensive node selection and power allocation in multi-source cooperative mesh networks

This paper considers resource allocation with relay selection in a multi-source multi-destination mesh network wherein dedicated relay nodes use the decode-and-forward (DF) protocol. The key difference from previous work is that we consider resource allocation across the source-relay, relay-destination, and source-destination channels in a multi-source network. The solution to the related optimization problem simultaneously solves for relay selection, power allocation, and the cooperation strategy (direct transmission, if optimal, is a valid solution). Since the jointly optimal solution is of exponential complexity, we introduce a set of time-sharing factors and relax the selection constraint, resulting in an upper bound to the true solution. Imposing selection leads to a feasible, but tight, lower bound on the optimal solution. Second, we propose a decentralized selection and power allocation scheme. Simulation results show that the performance of the decentralized selection scheme almost exactly tracks that of the upper bound for both the max-sum and max-min rate metrics while offerring computational benefits.

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

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 close-to-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.

Modeling and analysis of ergodic capacity in network MIMO systems

This paper studies the downlink ergodic capacity of a network multiple-input multiple-output (MIMO) system. The system model includes base-stations (BSs) randomly distributed with a fixed density, each equipped with M antennas, scheduling K single-antenna users, and forming cooperating clusters via perfect backhaul links. Intra-cluster interference is eliminated by joint transmission using zero-forcing beamforming assuming perfect channel state information (CSI), while inter-cluster interference remains. This paper shows that although coordinating a large cluster of BSs eliminates strong interferers, the coordination gain depends on the network load factor, defined as the relative ratio of M and K. In particular, we show that with M = K, increasing the coordination cluster size is only beneficial for the cluster-edge users, while degrading the ergodic capacity of the users located close to the cluster center. In contrast, when M > K, increasing the cluster size potentially improves every users ergodic capacity. In the second part of this paper, we use tools from stochastic geometry to account for random BS locations in characterizing the performance of network MIMO systems. In this setting, we model the BS locations according to a homogeneous Poisson point process with a fixed density, and propose tractable, yet accurate, distribution functions for the signal and inter-cluster interference powers. We then derive an efficiently computable expression for the user ergodic capacity as a function of the distance between a user and the cluster center.

Cooperative Strategies and Fairness-Aware Resource Allocation in Selection-Based OFDM Networks

This paper considers a multi-source orthogonal frequency division multiplexing (OFDM)-based network of access points wherein dedicated relays use the decode-and-forward relaying. We investigate the joint problem of transmission strategy selection (relaying v/s direct), relay assignment, and power allocation to maximize the minimum rate across sources in two different cooperative scenarios: subcarrier-based and block-based relaying. In the subcarrier-based scheme, each subcarrier is treated as an independent transmission; however, in addition to the synchronization problems caused, it is likely impractical for a relay to decode a subset of subcarriers. Thus, we study relay selection for the entire OFDM block. The key difference from previous work is that we consider resource allocation across source-relay, relay-destination, and source-destination channels. Second, a simple, distributed, block-based scheme is proposed. Simulation results using the COST-231 channel model reveals that the performance of this heuristic scheme tracks that of the optimum block-based scheme while it significantly decreases the required computational complexity.

Link-Level Analysis of Low Latency Operation in LTE Networks

This paper studies the physical-layer benefits of low latency operation in long-term evolution (LTE) networks. Latency reduction can be achieved by reducing the transmission time interval (TTI) from 1ms to the duration of only a few orthogonal frequency-division multiplexing symbols. The TTI shortening potentially enables faster link adaptation, thereby enhancing system performance. However, enabling low latency operation in a backward compatible manner requires a careful design and performance characterization. This paper conducts a link-level performance analysis of a low latency LTE network in both downlink and uplink with different transmission schemes and under various operating regimes. Our results reveal that a low latency LTE network can provide reasonable link-level performance improvements as compared to a legacy LTE network.