MMSE-based iterative channel estimation for HSUPA

This paper proposes a channel estimation algorithm for the High Speed Uplink Packet Access (HSUPA) system supporting the 64-QAM modulation and operating in a high Signal-to-Interference and Noise Ratio (SINR) regime. The algorithm is aimed to overcome limitations of the conventional approach which is based on simple correlation of the received signal with the pilot sequence and has insufficient performance to achieve the required SINRs because of being self-interference limited. The proposed algorithm exploits the Minimum Mean Square Error (MMSE) estimation principles and additionally utilizes an idea of iterative usage of the channel estimate from the previous channel estimation algorithm stage as a priori information to its next iteration the same time not involving receiver decisions. The paper provides efficiency analysis of the proposed algorithm by means of Link Level Simulations (LLS). The algorithm complexity is compared to the correlation-based channel estimation approach. The results of the study demonstrate considerable accuracy improvement for the channel signal-to-noise ratios (SNRs) range of 15-30 dB with the throughput performance practically reaching the limit of ideal channel knowledge. The required increase of the channel estimation algorithm complexity is 25 to 100 times for 8 and 13 estimated channel taps respectively. The proposed algorithm is suitable for HSUPA base stations supporting 3GPP Release 11 and beyond.