Deep learning based channel estimation

Betreuer: Lingrui Zhu
Art der Arbeit: Projekt (MSc)
Ausgabe: 11/2022
Bearbeiter: Muhammad Umair
Status: in Arbeit
Kurzfassung:

Channel estimation, as one crucial step in a wireless communication system, is an appropriate area where ML plays an important role. In [7] and [9], deep learning is applied for channel estimation. However, the following two aspects need to be taken into consideration when using ML/AI for channel estimation. The first problem is that complexity of calculation of ML method is relatively high. For a scenario with strict requirements on latency, it will not be feasible to achieve calculation with a high complexity. The other one is, for offline training, model need to be trained before deploying. If the channel for inference phase has been changed and is different from the channel during the training phase, performance will be influenced. In our research, we are trying to find methods to solve the above problems.

Reference:

Soltani, Mehran, et al. "Deep learning-based channel estimation." IEEE Communications Letters 23.4 (2019): 652-655.

Zuletzt aktualisiert am 29.11.2023 von L. Zhu
AIT ieee GOC tzi ith Fachbereich 1
© Arbeitsbereich Nachrichtentechnik - Universität BremenImpressum / Kontakt