Highly scalable implementation of a robust MMSE channel estimator for OFDM multi-standard environment

Abstract

In this paper a VLSI implementation of a highly scalable MMSE (Minimum Mean Square Estimator) is presented with the ultimate goal of demonstrating the potential of MMSE as enabler for multi-standard channel estimation. By selecting an appropriate implementation, a complexity reduction of 98% is achieved when compared to Time-Domain Maximum Likelihood Estimation (TDMLE), whereas low power consumption is accomplished by implementing a low-power-mode. The architecture is capable of performing Least Square (LS) estimation and MMSE compliant with 3GPP LTE (Long Term Evolution), IEEE 802.11n (WLAN), and DVB-H (Digital Video Broadcast for Handheld Devices), The estimator is synthesized using a 65nm low-leakage high-threshold standard-cell CMOS library. The design occupies an area of 0.169 mm2, is capable of running upto 250 MHz, providing a throughput of 78M estimates/second. Simulations under a typical LTE reception show that the implementation dissipates 4.9μW per sample.