Collaborative relaying in downlink cellular systems

Abstract

The deployment of relays in cellular system has recently been standardized in the WiMAX, IEEE 802.16j standard and is a topic of discussion in the advanced specifications of 3GPP-Long Term Evolution (LTE) [2]. Although commercial relay deployments in cellular systems are not prominent at present, future wireless cellular systems will involve operation with dedicated relays for improving coverage, increasing cell-edge throughput, delivering high data rates and assisting group mobility. The proposed architecture is such that relays would be placed at certain locations (planned or unplanned) in the cell to help in forwarding the message from the base station to the user in the downlink, and from the user to the base station in the uplink. Relays will be more sophisticated than simple repeaters and could perform some digital base band processing to help the destination terminal get better reception. These relays will rely on air interfaces, and hence avoid the considerable back haul costs involving data aggregation and infrastructure costs associated with backbone connectivity. However, there are a lot of open issues that require research to answer. We present some of these issues in the sequel: Throughput gains due to relay deploymentsIn cellular networks that are coverage limited, deploying relays can help in multihop transmission and provide power gains due to reduction of distance attenuation [3]. These power gains, in turn, translate to throughput improvements for the edge users. However, in interference limited settings, as is common in cellular systems, uncoordinated transmission by relays lead to increase of the overall interference levels in the cell and could be counter-productive by reducing the signal-to-interference plus noise (SINR) levels of users in the system. Coordination of transmissions in the system would require centralized control and incur high costs and overhead, especially in the uplink. Thus, there is a need for a thorough evaluation of throughput improvements in a cellular system. In the cellular systems literature, there have been simulation studies to evaluate throughput gains in cellular systems, e.g., [7,8,9]. Even though the studies were conducted under different sets of (idealized) assumptions, throughput improvements in interference limited cellular systems are