Benchmarking Machine Learning Techniques for THz Channel Estimation Problems

TL;DR

This paper benchmarks various machine learning algorithms for THz channel estimation, highlighting the effectiveness of projected gradient ascent in low SNR conditions for future 6G wireless systems.

Contribution

It introduces a comprehensive benchmarking of ML techniques for THz channel estimation, emphasizing PGA's superior performance over neural networks and logistic regression.

Findings

PGA achieves NMSE of -12.8 dB at 0 dB SNR.

Neural networks and logistic regression are less effective than PGA.

Benchmark results guide future THz communication system design.

Abstract

Terahertz communication is one of the most promising wireless communication technologies for 6G generation and beyond. For THz systems to be practically adopted, channel estimation is one of the key issues. We consider the problem of channel modeling and estimation with deterministic channel propagation and the related physical characteristics of THz bands, and benchmark various machine learning algorithms to estimate THz channel, including neural networks (NN), logistic regression (LR), and projected gradient ascent (PGA). Numerical results show that PGA algorithm yields the most promising performance at SNR=0 dB with NMSE of -12.8 dB.