[Back]

M. Guillaud:
"Recent Results on Interference Alignment in Partially Connected Networks and Sum-Rate Optimization";
Talk: UT Austin Wireless Networking & Communications Group Workshop, Austin, TX, USA (invited); 12-02-2011.

In this talk, I will present two facets of our recent research on the interference channel.

In the first part, I will focus on interference alignment in partially connected networks with constant MIMO channels, and present a simple sufficient condition for the network to be proper. Interestingly, this condition is independent on the total number of users in the network, indicating that IA can be possible in networks of arbitrary size while keeping the signaling dimension finite (no channel extension). We further consider the approximation of a 2D cellular network by a partially connected network in which we apply IA, and present performance results obtained for this scheme.

In the second part, I will discuss the problem of sum-rate maximization over the interference channel. Several optimization algorithms based on the optimization of the position of the transmission subspace (optimization on the Grassmann manifold) will be considered, and their respective performances and convergence properties compared.

interference alignment