An Analogue Channel Estimation Method Suitable for DigiLogue Receivers

Terabits per second (Tbps) wireless links are essen-tial for realizing emerging mobile immersive experiences such as holographic telepresence. A promising approach to achieve such data rates is to exploit ultra-wide bandwidths. However, the need for ultra-fast, high-precision analogue-to-digital converters (ADCs) in digital systems to exploit such ultra-wide bandwidths becomes impractical in terms of power consumption. In addition, the increased sampling rates that challenge the speed of modern digital processors and the algorithmic complexity make performing ultra-fast digital signal processing challenging in terms of processing latency and power consumption. A promising approach to overcome these bottlenecks is DigiLogue processing, which suggests to remove ADCs and to perform signal processing directly in the analogue domain. However, practical DigiLogue receivers need to support fading channels, in which accurate channel estimation becomes crucial for reliable signal detection. In addition, systems operating in high-frequency bands (e.g., mm-Wave and THz), where ultra-wide bandwidths are available, experience a small number of channel taps, in which analogue-based channel estimation approaches can be realized with low complexity. However, existing analogue-based channel estimation approaches target hard detection/decoding and they are not able to provide the channel information required to perform “soft” processing in DigiLogue receivers. In this work, we propose an analogue-based channel estimation approach based on utilizing Golay sequences employed in recent high-speed standards, which is suitable for DigiLogue receivers. Our approach achieves a 1 dB better signal-to-noise ratio (SNR) in error performance compared to existing analogue methods and saves more than 29 x power than the digital counterpart.