Exploiting Sparsity for Interference Management in Broadband Communications

Prof. Waheed Bajwa received funding from the Qatar National Research Foundation, for the project::

"Exploiting Sparsity for Interference Management in Broadband Communications: Theory, Applications, and Testbeds"

This is a 3-year, approx. $1 million grant with UT-Dallas, and Qatar University (http://www.qnrf.org/newsroom/press_releases/detail.php?ID=3366) (pending award paperwork). The acceptance rate for this cycle of awards was 20% (137 out of 710 proposals recommended).

This proposal is expected to fund one graduate student each at Rutgers, UT-Dallas, and QU and also a joint post-doc with residence in QU.

Details of the project are as follows:

Description of the proposal: The main objectives of this proposal are to: i) develop a new unified signal processing framework for interference mitigation in broadband communication systems based on identifying and exploiting sparsity ii) extend the theoretical guarantees in the field of compressive sensing (CS) to encompass the mathematical structure arising in the proposed framework and develop auto-tunable sparse recovery algorithms based on a frame-theoretic understanding of CS theory, and iii) demonstrate on experimental testbeds the value of the proposed framework in enhancing the performance and reducing the complexity of broadband transceivers.

Our proposed approach enjoys many advantages including: i) Computational efficiency due to its focus on low-complexity solutions of the interference mitigation problem; ii) no a priori assumptions are made on the interference probability density function, second-order statistics, frequency support, or power level. Instead, we make only a mild, but highly realistic, assumption on sparsity of the interference signals; and iii) The proposed sparsity-aware approach is applicable to a wide range of interference sources (e.g. radio frequency interference, impulse noise, intersymbol interference, crosstalk, etc..) and a number of applications including digital subscriber lines, wireless local area networks, power line communications, and millimeter-wave beamforming.