Abstract
Filtered Orthogonal Frequency Division Multiplexing (FOFDM) is one of the key candidates for harvesting the features that millimetre wave cellular communication is to offer for the 5G and beyond cellular generation. Its chief robust features are increased throughput, efficient spectral utilisation, and interference immunity. Some strict methodologies must be considered to fully utilise these inherent features, especially those allowing guard tones, accurate CSI management and other filter design parameters. This paper proposes a technique to exploit the added redundancies in pilot-based estimators. Here, the authors present an extension idea of an adaptive weight estimator with dual adaptive weights instead of one unique weight factor. We show that these redundancies are insignificant when considering the massive amount of data to be processed and thus can be eliminated to expedite the estimation process efficiently. We implement a weight-based estimator with the necessary accuracy threshold but faster converging to achieve this goal. Simulations follow to demonstrate the advantages of the proposed method.