MIT researchers report dramatic improvement in Wi-Fi performance with MegaMIMO 2.0


While vendors have made claims about making breakthroughs in distributed Multiple Input Multiple Output (D-MIMO), researchers at MIT’s Computer Science and Artificial Intelligence Lab (CSAIL) are demonstrating a system called MegaMIMO 2.0 that can transfer wireless data more than three times faster than existing systems while also doubling the range of the signal.

Led by Professor Dina Katabi, the research team released a new paper describing the design and implementation of MegaMIMO 2.0, which they said represents the first real-time fully distributed 802.11 MIMO system. MegaMIMO 2.0 delivers a full-fledged 802.11 PHY while meeting the tight timing constraints of the 802.11 protocol. It also supports dynamic environments and mobile clients.

The researchers acknowledged there is a large body of theoretical work that analyzed the performance gains provided by distributed MIMO and showed that it can scale wireless throughput with the size of the network. Significant efforts have been made in recent years in moving from theory to practice. But they said that these systems focus only on the problem of synchronizing the transmitters in time, phase and frequency, and don’t address power control and the overhead of learning and tracking the channels.

Existing distributed MIMO systems are all based on cooperative multipoint (CoMP), which assumes a shared clock, distributed either via GPS or a wire, and a dedicated high throughput fiber backhaul infrastructure to deliver signals to all antennas with a very high throughput and carefully controlled latencies. Examples cited were a system by PCell and a demonstration by Ericsson.

“In contrast, MegaMIMO 2.0 operates with fully distributed independent radios and does not need a single clock,” their paper stated. “Further, it introduces a new technique for extending reciprocity to distributed MIMO systems and presents detailed evaluation results.”

“In today’s wireless world, you can’t solve spectrum crunch by throwing more transmitters at the problem, because they will all still be interfering with one another,” said Ezzeldin Hamed, a PhD student who is lead author on the paper, in a CSAIL News post. “The answer is to have all those access points work with each other simultaneously to efficiently use the available spectrum.”

The researchers set up a mock conference room with a set of four laptops that each roamed the space on top of Roomba robots. The experiments found that the system could increase the devices’ data-transfer speed 330 percent.

MegaMIMO 2.0’s hardware is the size of a standard router and consists of a processor, a real-time baseband processing system and a transceiver board.

Katabi and Hamed co-wrote the paper with Hariharan Rahul, as well as visiting student Mohammed A. Albdelghany. Rahul said that the group’s technology can also be applied to cellular networks. The team plans to expand MegaMIMO 2.0 to be able to coordinate dozens of routers at once, which would allow for even faster data-transfer speeds.

It’s soon to be commercialized, according to MIT CSAIL, so it could end up dramatically improving the speed and strength of wireless networks during high-usage events like concerts, football games and conventions.

For more:
- download the research paper
- see this Forbes article
- see this MIT CSAIL post

Related articles:
Ericsson teams with MIT on autonomous cars
ZTE claims milestone with first pre-commercial test of distributed MIMO
MIT lasers in on wireless broadband for the cosmos