Abstract
Techniques are presented herein that support examining the channel state information (CSI) matrix that arises from a channel sounding and using the results of that examination to train a machine learning (ML)-based model that considers all of the available wireless parameters at each given measurement. Under the presented techniques, after sufficient training has been completed a CSI matrix may be predicted over short intervals (using, for example, a long short-term memory (LSTM) network) thus allowing a Wi-Fi access point (AP) to reduce the sounding frequency and, as a result, improve overall wireless performance.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Barton, Robert; Henry, Jerome; Gandhi, Indermeet; Lovison, Federico; Jung, Doosan; and Shao, Qihong, "PREDICTIVE AI-BASED CHANNEL SOUNDING MECHANISM", Technical Disclosure Commons, (January 29, 2024)
https://www.tdcommons.org/dpubs_series/6632