Femtocells, also called home base stations, have emerged as a low cost solution for residential and indoor coverage. They are low-power (< 100 mW), short-range base stations, deployed by end customers to provide better indoor voice and data reception. Femtocells operate on licensed spectrum and provide data and voice services to user terminals by connecting them to the core network through a broadband Internet connection such as digital subscriber line (DSL) or cable. Femtocells are deployed by customers with minimal or no radio frequency planning to reduce the network capital and operational expenditure. They are projected to share the same spectrum as the macro-cellular system.
Femtocells are currently deployed to be operated in closed access mode, restricting access to the femtocells to subscribed home users only. Unfortunately, this causes signi?cant interference between adjacent femtocells, as well as cross tier interference between the femtocells and the macrocells. Moreover, interference management in femtocell networks does not conform with traditional cellular networks, as it cannot rely on full coordination between the macrocells and the femtocells and needs to be decentralized. Considering the fundamental architectural and network design paradigm changes caused by the femtocells deployment vis-a-vis advanced antenna interference mitigation strategies, is key to reaping the most benefits out of femtocell networks.
My research group has been studying several aspects of femtocells, with an emphasis on using multiple antennas for interference mitigation and cancelation for the femtocell networks. Our area of interest mainly includes investigating transmission techniques that rely on closed loop MIMO transmission strategies, to coordinate the transmissions in heterogeneous networks between the femtocells, and mitigate cross-tier interference. For other work at UT Austin on femtocells, see Professor Jeff Andrews’ web page.
S. Akoum, M. Kountouris, and R. W. Heath, Jr.,“Limited Feedback over Temporally Correlated Channels for the downlink of a Femtocell Network,” submitted to IEEE Transactions on Wireless Communications, Jan. 2010. Preprint available at ArXiv.
When restricting access to home users, femtocells cause a substantial interference problem that cannot be mitigated through coordination with the macrocell base station. In this paper, we analyze multiple antenna communication on the downlink of a macrocell network, with femtocell overlay. We evaluate the feasibility of limited feedback beamforming given delay on the feedback channel, quantization error and uncoordinated interference from the femtocells. We model the femtocell spatial distribution as a Poisson point process and the temporal correlation of the channel according to a Gauss-Markov model. We derive the probability of outage at the macrocell users as a function of the temporal correlation, the femtocell density, and the feedback rate. We propose rate backoff to maximize the average achievable rate in the network. Simulation results show that limited feedback beamforming is a viable solution for femtocell networks despite the CSI inaccuracy and the interference. They illustrate how properly designed rate backoff improves the achievable rate of the macrocell system.
Ronghong Mo and Tony Q. S. Quek and R. W. Heath, Jr., “Robust Beamforming and Power Control for Two-Tier Femtocell Network,” Proc. of the IEEE Vehicular Tech. Conf. , Budapest, Hungary, May 15-18, 2011.
In this paper, we consider a two-tier network with a macrocell and several closed-access femtocells. The macrocell and femtocell base stations have multiple antennas. We propose an algorithm for downlink power control and beamforming. In the proposed algorithm, the base station performs maximum ratio transmission and the femtocells employ beamforming with power allocation, determined by the base station. Robust optimization is used to improve the performance of the beamforming algorithms.