Multiple Cell MIMO

Multicell MIMO Communication

Multicell cooperation

Upcoming cellular standards like the 3GPP LTE Advanced are targeting universal frequency reuse in a bid to increase peak data rates. This could, however, lead to high levels of co-channel interference (CCI) arising due to simultaneous transmissions on the same frequency by neighboring base stations. CCI can significantly reduce data rates and cause outages in cellular systems, especially at the cell-edges. Multicell cooperation is one solution to manage CCI in future commercial wireless standards. It is also known as coordinated multipoint transmission (CoMP) in 3GPP LTE Advanced.

Base station cooperation entails sharing control signals, transmit data, user propagation channel state information (CSI) and/or precoders via high-capacity wired backhaul links to coordinate transmissions. In practice, however, backhaul will be bandwidth-limited due to the prohibitive costs involved in establishing high-capacity links. This restricts the amount of information that can be exchanged among base stations, which in turn determines the level of cooperation and the performance gains obtained. Multicell base station cooperation can be broadly divided into three different levels of cooperation.


1. Control-level cooperation. These cooperative strategies exchange only control-level information among base stations, leading to small load on the backhaul link. They usually involve some form of joint allocation of available resources to orthogonalize user transmissions in adjacent cells, by assigning different frequency bands of operation and/or timing cycles. While these techniques yield higher sum-rates than static transmission algorithms, they do not utilize all the available frequency and time resources and hence, do not realize the performance gains that can be potentially obtained using base station cooperation.

2. Full cooperation. Known as joint transmission in 3GPP LTE Advanced, full cooperation leads to the highest sum-rates at the cost of increased overhead due to global CSI requirements and the exchange of a greater amount of information among base stations, including CSI, transmit and precoding data. Full cooperation is typically high complexity and imposes a large load on backhaul links. Examples of these techniques include multicell dirty paper coding, multicell zero-forcing and minimum-mean squared error precoding, etc.

3. Partial cooperation. Partial cooperative strategies, where base stations exchange only the CSI of active users, offer a fair balance between ensuring a reasonable load on the backhaul and attaining the performance gains using cooperation. The shared CSI can be used by base stations to design individual precoding matrices (or beamforming vectors, for single-stream transmission) on site to transmit exclusively to users within their own cell. This is known as coordinated beamforming in 3GPP LTE Advanced.

In my work with my research group WSIL and collaborators, we have studied the impact of out-of-cell interference on uncoordinated systems and have studied cooperative MIMO systems. We have developed new cooperative strategies to reduce the high complexity involved with cooperation by proposing simple linear and non-linear cooperative algorithms. We studied the clustering of coordinating cells to decrease the overheads associated with cooperation. We also investigated different aspects of practically implementing cooperation, like multicell limited feedback.

Select Publications

C. B. Chae, S. Kim, and R. W. Heath, Jr., “Network Coordinated Beamforming for Cell-boundary Users: Linear and Non-linear ApproachesIEEE Journal on Sel. Topics in Sig. Proc., vol. 3, no. 6, pp. 1094-1105, Dec. 2009.

This paper proposes low complexity linear and non-linear network coordinated beamforming algorithms for the multicell downlink channel. A three-cell scenario with one user/cell is considered. It is shown that the proposed linear algorithms approach the sum capacity realized by multi-cell dirty paper coding. Low complexity non-linear algorithms that avoid power enhancement and maximize the effective channel gain for enhancing the bit error rates are also presented in the paper for a three-cell system.

J. Zhang, J. G. Andrews, A. Ghosh,  and  R. W. Heath, Jr., “Networked MIMO with Clustered Linear Precoding,” IEEE Trans. on Wireless vol. 8, no. 4, pp. 1910-1921, April 2009

This paper proposes a clustered base transceiver station coordination for a large cellular MIMO network, which includes full intra-cluster coordination to enhance the sum rate, and limited inter-cluster coordination to reduce interference for the cluster edge users. It is shown that a small cluster size (about 7 cells) is enough to get most of the sum rate benefits from clustered coordination while greatly relieving channel feedback requirements.

R. W. Heath, Jr., T. Wu,  and  A. C. K. Soong, “MIMO Spatial Mode Adaptation at the Cell Edge Using Interferer Spatial Correlation,” (invited) Proc. of theIEEE Vehicular Tech. Conf. , Barcelona, Spain April 26 – 29, 2009, pp. 1-6.

In this paper, MIMO spatial multiplexing systems are analyzed in uncoordinated cellular systems. Without coordination, the spatial transmission mode, e.g. spatial multiplexing or transmit diversity is chosen independently in each cell. Unfortunately, the optimum transmission mode for users that are interference limited depends to a large extent on the transmission mode used by interfering base stations. This paper proposes interference-aware link adaptation where base stations exchange their transmission plans with neighboring base stations and broadcast this information to the active users.

Seijoon Shim, Jin Sam Kwak, R. W. Heath, Jr., J. G. Andrews, “Block Diagonalization for Multi-User MIMO with Other-Cell Interference,” IEEE Trans. on Wireless, vol. 7, no. 7, pp. 2671-2681, July 2008.

This paper examines the impact of interference on an uncoordinated system that employs block diagonalization. It is shown that CCI substantially reduces performance. A two-state transceiver solution is proposed that uses a whitening phase to reduce some of the effects of CCI and improve the performance of block diagonalizaiton.

R. Chen, J. G. Andrews, R. W. Heath, Jr.,  and  A. Ghosh, “Uplink Power Control in Multi-Cell Spatial Multiplexing Wireless Systems,” IEEE Trans. on Wireless, vol. 6, no. 7, pp. 2700-2711, July 2007.

This paper looks at the uplink of spatial multiplexing cellular systems without base station coordination. It proposes a power control algorithm that improves performance in the presence of CCI.


The research in the area of multicell base station cooperation was funded through NSF grants, particularly, NSF-CCF-0514194 and NSF-CCF-0830615, and a gift from Huawei.

Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the sponsors.

Comments are closed.