Improving ITCN Channel Estimation using AI-based Techniques

• Published in: 2023 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB) pp. 1 - 6
• Main Authors: Bilbao, I., Gairaud, J., Iradier, E., Ferre, G., Montalban, J., Angueira, P.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 14-06-2023
• Subjects: Channel estimation; deep learning; full-duplex; IBFD; IDL; Interpolation; ITCN; machine learning; Proposals; Receivers; signal isolation; Spectral efficiency; Superresolution; Time-frequency analysis
• ISSN: 2155-5052
• DOI: 10.1109/BMSB58369.2023.10211203

Inter-Tower Communication Networks (ITCN) is expected to be the key enabler of many new applications inside ATSC 3.0 due to the high spectral efficiency obtained by the full-duplex essence of ITCN. Nevertheless, the implementation of ITCN in real environments is still in the development stage because of its technical challenges. The main difficulty comes from the portion of the transmitted signal coupled to the receiver (i.e., loopback signal) and the double channel estimation that has to carry out, one for the loopback channel and another one for the forward signal. In this sense, multiple processes affect the recovered forward signal and deteriorate the signal quality. In particular, this work focuses on improving the received forward signal quality by enhancing the channel estimation. In this case, an Artificial Intelligence (AI) solution is used based on the Super Resolution (SR) technique to overcome the error produced by the interpolation process followed in the pilot-based channel estimations. The results indicate that the channel estimation can be improved by more than one order of magnitude depending on the configuration parameters.