By Rachel Harrington
When it comes to wireless communications, interference between users is an obstacle to avoid. When multiple users transmit on the same frequency band, nearby receivers only see the superposition of their signals, which makes it hard to discern the individual packets of data.
Work by Assistant Professor Bobak Nazer (ECE, SE) and Professor Michael Gastpar, who holds positions at the École Polytechnique Fédérale de Lausanne in Switzerland and the University of California, Berkeley, is causing other researchers to rethink that notion.
A paper by them titled, “Compute-and-Forward: Harnessing Interference Through Structured Codes,” explores the possibility of exploiting the algebraic structure of interference to achieve higher data rates.
Their framework makes it possible for a receiver to recover linear combinations of packets from superimposed signals. Recovering the original packets is simply a matter of collecting enough equations to solve for the unknowns. Ultimately, this technique may enable wireless networks to operate at significantly higher throughputs by allowing several users to simultaneously occupy the same channel.
“Much of the prior work has focused on the statistical aspects of the interference problem,” said Nazer. “One of the main emphases of this paper is that there is a benefit to thinking about algebraic structure as well.”
“I was very happy to hear about the award and really appreciate the recognition from the communications and information theory communities,” said Nazer. “This is a project we’ve been working on for a long time. More than anything it’s nice to see that others are getting as excited about the work as we’ve been.”
In July, Nazer and Gastpar were recognized with a plaque and honorarium at the IEEE International Symposium on Information Theory in Istanbul, Turkey.