Enhanced Interference Cancellation of Cell-Specific Reference Signals for LTE-A

Abstract

The performance of the physical downlink shared channel (PDSCH) of Long-Term Evolution Advanced (LTE-A) carrying downlink data traffic to the users served in a cell is typically limited by co channel interference from neighboring cells. One of the most significant interference sources in LTE/LTE-A is cell-specific reference signals (CRS), which are transmitted by neighboring cells across the entire frequency band even when downlink data traffic is not present. As a consequence, in scenarios without PDSCH from neighboring cells, CRS interference becomes the limiting factor in achieving higher downlink throughput. Therefore, interference cancellation from CRS becomes one of the important signal processing problems in LTE/LTE A systems. In conventional receivers, CRS interference is canceled by using channel estimates corresponding to neighboring cells, which are obtained from the received signal. However, the presence of the useful signal transmitted by the serving cell in the received signal makes channel estimation for neighboring cell CRS less accurate. As a result, the performance of such CRS interference cancellation (CRS IC) receivers is typically poor in medium and high signal to interference ratio (SIR) scenarios. In this paper, we propose a novel decoding aided CRS-IC receiver for LTE A systems. The key idea of the proposed scheme is to generate a soft estimate of the useful signal using the turbo decoding scheme and cancel this from the total received signal prior to the estimation of the channels corresponding to interfering cells. Compared to conventional schemes considered in 3GPP until now, the proposed approach provides performance improvements in a wider range of SIRs, while the performance approaches the one of an ideal system using perfect cancellation of CRS interference from neighboring cells.