Enabling Practical Interference Alignment in Real-time Hardware
Abstract
Major Goals: The major goal of this effort is: To enable interference alignment in concert with interference cancellation in a manner that is implemented in practice. Interference Alignment, a strategy for linearly improving the throughput of wireless networks, is yet to be realized in practice. Similarly, interference cancellation, a strategy for enabling much higher throughput for a particular bandwidth, is studied heavily theoretically but yet to be fully enabled in practice. In this project, we use the funding to enable practical interference management. Interference alignment, together with other coding and decoding mechanisms in communication systems, including systems such as beam forming, beam steering etc., can massively improve the throughput and performance of wireless communications systems. In particular, the combination of such schemes with machine learning, especially deep learning, can be hugely beneficial to reduce such complex algorithms to practice in the real world. Accomplishments: This grant was instrumental in helping realize the following results: 1. Interference Alignment Using Deep Learning: In this line of work, we develop a framework for an autoencoder based transmission strategy for achieving distributed interference alignment and optimal power allocation in a multiuser interference channel. The users in the interference channel have access to the local channel state information only. We compare the explicit schemes, such as MaxSINR, against the autoencoder schemes. We find that the MaxSINR schemes outperform the autoencoder networks which are either jointly or distributively trained. However, we find that the autoencoders which are pre-trained with the beamforming vectors and the power allocation obtained from the explicit schemes outperform the explicit schemes when the interference gets stronger. The explicit schemes perform well as they are effective in choosing the set of users which are to be suppressed.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 10, 2020
- Accession Number
- AD1229525
Entities
People
- Sriram Vishwanath
Organizations
- University of Texas at Austin