This Small Business Innovation Research Phase I project will develop a system for locating a smartphone inside a building very accurately, helping users navigate and find items of interest indoors, where current systems do not function. Building on academic research from MIT, the system will use existing Wi-Fi signals combined with motion sensors and map data to produce estimates with a 90th percentile accuracy of 1-3 m. This is significantly more accurate than the currently available systems, which are accurate to within 5-10 metres in the mean and worse in the 90th percentile. It uses existing Wi-Fi signals, which means the system does not require additional hardware in most buildings. The system incorporates three novel technologies: 1) a novel training system that uses motion sensors and map data to dramatically reduce the time it takes to calibrate the system in a new building compared to existing technologies, 2) a new location matching algorithm that fuses motion sensors and WiFi signals to improve the accuracy and reliability of location estimates from limited training data, and 3) a system to predict the accuracy using the available Wi-Fi access points, to determine if new hardware or additional training is required. All three technologies are key competitive advantages essential to bring this technology to market. This proposal will develop a Wi-Fi-based indoor location system that is not only accurate, but also easy and cheap to deploy.

The broader impact/commercial potential of this project is that smartphones, that have become indispensable tools for locating buildings and businesses in a city, can now also be used to show a user his/her location on an indoor map, and guide them to find what they are looking for inside large buildings where people get lost all the time today --- while looking for points of interest like ticket counters, products on shelves, airline gates, and stores. InFix's approach of using existing WiFi beacons, ubiquitous everywhere today, has the highest potential to achieve scale and reach, and realize the vision of "precise location everywhere". This is in contrast to previous academic, research and commercial efforts that all require installing specialized hardware. Infix will license this technology to space owners, to integrate into their smartphone applications. As the map and location database grow, the company will allow third-parties to build their own innovative applications.

Project Start
Project End
Budget Start
2012-07-01
Budget End
2012-12-31
Support Year
Fiscal Year
2012
Total Cost
Indirect Cost
Name
Infix Corporation
Department
Type
DUNS #
City
Cambridge
State
MA
Country
United States
Zip Code
02139