Where Are You, Precisely? MIT Looking At Ways To Improve Positional Accuracy

As more and more daily activities have become dependent on the positioning and navigation assistance of GPS, companies are racing to take advantage of it with new products, services, and the ability to target them to individuals in a specific place at a specific time. But reliable and accurate localization of mobile nodes, a key enabler for emerging applications in the commercial, public safety and military sectors, is challenging to achieve in harsh environments with limited infrastructures. So MIT’s Wireless Information and Network Sciences Lab has been working on an alternative and/or complement to GPS called network localization and navigation (NLN), an emerging paradigm for providing accurate positional information. During a visit to MIT on April 14th, the Marconi Society Board received a preview of the lab’s work.

“An NLN user’s localization accuracy is affected by the allocation of transmission resources (such as time, energy, and frequency) to the user’s neighbor nodes,” says Professor Moe Win. So, the lab has created a geometric framework for optimal resource allocation in NLN, developed by Ph.D. candidate Wenhan Dai and Prof. Win. (After some experimentation, they decided that the resources allocation problem can best be visualized as the intersection of a polyhedron and hyperboloids in 3-D space.) By exploiting the geometric structure, the MIT team has proven that the best possible performance can be achieved by allocating resources to only a small subset of possible neighbor nodes. The team also found that the geometric structure significantly reduces the computation complexity of optimal solutions. In addition, they translated these theoretical ideas into efficient algorithms that have been implemented as a software solution.

The resulting performance improvement was demonstrated to the Marconi board using the lab’s radio network testbed.

“The questions from the Marconi board, which focused on the applications and impact of our research, were inspiring as well as insightful,” says Professor Win. “For instance, articulating for them how the algorithms we’ve developed would be ported from the radio network testbed to commercial devices helped us imagine future applications of our work. It was also encouraging—and reassuring—to hear how much the Marconi board members value problem-driven research like ours.”