Drexel University

University of Pennsylvania

Nikhil Gulati

Kapil R Dandekar

David Gonzalez

Dwight L Jaggard

Rohit Bahl

A method for selecting the state of a reconfigurable antenna installed at either the receiver or transmitter of a communication system is provided. The proposed method uses online learning algorithm based on the theory of multi-armed bandit to perform antenna state selection. The selection technique utilizes the Post-Processing Signal-to-Noise Ratio (PPSNR) as a reward metric and maximizes the long-term average reward over time. The performance of the learning based selection technique is empirically evaluated using wireless channel data. The data is collected in an indoor environment using a 2&#xd7;2 MIMO OFDM system employing highly directional metamaterial Reconfigurable Leaky Wave Antennas. The learning based selection technique shows performance improvements in terms of average PPSNR and regret over conventional heuristic policies.

Method for selecting state of a reconfigurable antenna in a communication system via machine learning

By using reconfigurable antenna based pattern diversity, an optimal channel can be realized in order to maximize the distance between two subspaces, thereby increasing sum-rate. The inventors show the benefits of pattern reconfigurability using real-world channels, measured in a MIMO-OFDM interference network. The results are quantified with two different reconfigurable antenna architectures. An additional 47% gain in choral distance and 45% gain in sum capacity were achieved by exploiting pattern diversity with IA. Due to optimal channel selection, the performance of IA can also be improved in a low SNR regime.

Reconfigurable antennas for performance enhancement of interference networks employing interference alignment

A method for selecting the state of a reconfigurable antenna installed at either the receiver or transmitter of a communication system is provided. The proposed method uses online learning algorithm based on the theory of multi-armed bandit to perform antenna state selection. The selection technique utilizes the Post-Processing Signal-to-Noise Ratio (PPSNR) as a reward metric and maximizes the long-term average reward over time. The performance of the learning based selection technique is empirically evaluated using wireless channel data. The data is collected in an indoor environment using a 2&#xd7;2 MIMO OFDM system employing highly directional metamaterial Reconfigurable Leaky Wave Antennas. the learning based selection technique shows performance improvements in terms of average PPSNR and regret over conventional heuristic policies.

Growing community of inventors

Philadelphia, PA, United States of America

Nikhil Gulati