NeTS: Small: Enhanced Interference Alignment for Networks using Reconfigurable Antennas

Drexel University, Philadelphia PA

Investigators

Abstract

Interference between wireless devices will become a serious bottleneck for future wireless networks as they become more pervasive in our everyday lives. The traditional approach of dealing with interference is to share limited resources, such as spectrum, among many users such that a given resource is used by only one user at a time. The recent notion of interference alignment (IA) proposes a new paradigm where several users can simultaneously send information by sharing the same resource. This new approach results in increased network capacity where every user gets half the interference-free throughput. In this research, the project team will leverage reconfigurable antenna systems to design and enhance interference alignment techniques for multi-user wireless networks. Reconfigurable antennas are capable of electronically switching between different radiation patterns (or 'states') in response to the needs of the overlying communication link and network. In this regard, reconfigurable antenna systems represent a significant new degree of freedom for developing interference management techniques. The project team proposes new ways of using the antennas to enable more practical solutions for future multi-user networks and also provide a software defined radio platform to enable further research within the community. While the vast majority of existing research in this area considers an antenna to be a monolithic 'black box', the project team will develop and demonstrate new compact form-factor electronically reconfigurable antenna technologies that are capable of providing pattern agility. This additional degree of freedom allows for the dynamic selection of radiation patterns to seamlessly integrate with, and extend interference alignment techniques. The first objective is to demonstrate how electrically reconfigurable antennas can greatly enhance the sum capacity achieved via distributed interference alignment, especially at low SNR, by improving the subspace design. The second objective is to demonstrate how blind interference alignment can be practically achieved by combining antenna state selection algorithms with multiple physical layer transmission schemes. The additional channel diversity provided by reconfigurable antennas comes with an overhead to acquire information about the state of all the channels, revealing a need for a strategy to select the optimal state and most importantly an ability to learn the changes in the channel state in order to adapt. With these goals in mind, the project team will utilize online learning based on multi-armed bandit theory to design algorithms to control and adapt the state of a reconfigurable antenna system. For the multi-user network, the team will analyze the cost of learning under an unknown statistical model of the channel and compare it with the oracle with full prior knowledge. Finally, the team will focus on implementing the developed interference alignment algorithms on Drexel's Software Defined Communications (SDC) testbed. The SDC testbed is based upon a Scalable OFDM physical layer, which operates close to many different standards. The project team will develop new blocks for precoding and decoding, cross-transmitter symbol-timing synchronization, as well as implement more efficient synchronization techniques that require less data overhead. The proposed combination of the Drexel SDC testbed with IA algorithms will result in a product that has high relevance to the industrial and academic research communities. Special effort will be made to not only disseminate research results through relevant conference and journal publication, but also through testbed technology demonstrations. In order to extend the outreach of the proposed research beyond the research community, high school students and teachers, specifically under-represented populations, will be invited to participate in the PIs research laboratories through pre-existing educational programs. The developed testbed will be used to develop demonstrations and laboratory-based course modules.