Abstract-It has been recently shown that base station cooperation may yield great capacity improvement in downlink multiple antenna cellular networks. However, the proposed solutions assume that there is a central processing unit which coordinates the information exchange and determines the optimal resource allocation of the overall cellular network. Whilst the benefits of base station cooperation are large, computational burden of the central unit can be significant. Thus distributed solutions are desirable. This paper suggests a distributed solution for base station cooperation via block-diagonalization and dual-decomposition to maximize the weighted sum network capacity under per-antenna power constraint. The block-diagonalization pre-coding matrix is determined separately at each base station. It enables the full potential of base station cooperation by determining a trade-off between inter-cell interference mitigation, spatial multiplexing and macro diversity. The power allocation problem is formulated as a network utility maximization (NUM) problem. By looking at its Lagrangian dual problem, the decomposable structure of the optimization problem is revealed. This leads to a distributed and iterative algorithm that converges to the global optimum. The advantage of macro diversity in addition to inter-cell interference mitigation and spatial multiplexing in base station cooperation context is studied and shows superior performance in terms of a higher capacity increase with lower variance.

Abstract-Multi-cellular radio systems are often limited due to the presence of cochannel interference. Proposed physical layer concepts, e.g. coordinated joint transmission and interference rejection combining, try to strengthen the signal while combating the interference. It is well known that base station cooperation yields great capacity improvement for downlink multi-antenna cellular networks. However, the proposed solutions assume a central processing unit, coordinating the information exchange and thus determining the optimal resource allocation of the overall cellular network. Recently multi-user eigenmode transmission was proposed to relax the constraint of channel state information at the transmitter. It requires the terminals to feed back their dominant eigenmodes only. To reduce the computational costs and the information exchange further, we consider limited localized base station cooperation. We demonstrate potential capacity gains in a cellular orthogonal frequency division multiplexing system using real 3D measured antenna patterns and the scaling with the number of cooperating antenna arrays. Additionally, we use minimum mean square error equalization at the terminal side to combat residual cochannel interference.

No abstract

Abstract-Cooperative joint transmission and detection algorithms have a high potential to increase the capacity of cellular radio systems. This paper investigates efficient network layer protocols to realize such cooperation over a heterogeneous, bandwidth-limited backhaul. Target systems are MIMO-OFDM cellular communication systems with flat hierarchy (no central unit), and we take 3GPP LTE as example. The proposed architecture uses IP multicast for stations receiving the same information and is scaleable by dynamically cooperating only on frequency subbands. SINR and mobility thresholds are used to first cooperate for critical cell-edge users with low mobility, where most interference reduction gain is to be expected.

Abstract-Base station cooperation is considered as a promising approach to increase the quality of service (QoS) in uplink. Base stations are connected via high-capacity backhaul links, which makes it possible to perform joint detection through information sharing among several base stations. Joint detection broadens the feasible signal to interference ratio (SIR) region, in which any SIR assignment for all users can be concurrently supported. Thus it provides a higher system throughput. We characterized the Pareto-optimal boundary of the feasible SIR region of power control with base station cooperation by utilizing the property of concave interference function. We suggest a distributed power control algorithm that exploits the broader feasible SIR region and optimizes the SIR assignment. Depending on the feasibility of the network to support cooperation, we analyze 3 degrees of cooperation: no cooperation, limited cooperation, and full cooperation.