A Completely Blind Channel Estimation Technique for OFDM Using Constellation Splitting

TL;DR

This paper introduces a novel blind channel estimation method for OFDM that uses constellation splitting and linear precoding to resolve scalar ambiguity without pilots, achieving performance comparable to semi-blind methods.

Contribution

It presents a new blind estimation algorithm that eliminates the need for pilots by employing constellation splitting and frequency-domain precoding.

Findings

Performs as well as semi-blind methods for M-ary PAM.

Effectively resolves scalar ambiguity without pilot symbols.

Validated through MATLAB simulations.

Abstract

The problem of second-order statistics (SOS)-based blind channel estimation in OFDM systems is addressed in this paper. Almost all SOS-based methods proposed so far suffer from a complex-scalar estimation ambiguity, which is resolved by using pilots or reference symbols. We propose an algorithm to resolve this ambiguity in blind manner using frequency-domain linear non-redundant precoding and constellation-splitting among the alternate subcarriers. The performance of the proposed scheme is evaluated via numerical simulations in MATLAB environment. Simulation results show that the proposed approach performs as good as its semi-blind counterpart for M-ary PAM systems.