Target Localization Using Radar Sensor Networks (Under Construction)


Overview

Target localization in WSN(Wireless Sensor Networks) has been an active research area for its importance and application. Most of the works so far, however, have been focused on cooperative system approach, which requires targets of helping infrastructure nodes localize. Rather, we take non-cooperative approach requiring zero per cent help from targets. For this purpose, we use Radar Sensor Network (RSN), a group of Pulsed Doppler radars interfacing with Tmotes for target tracking problem.

Platforms

  • Bumblebee: A Pulsed Doppler radar manufactured by Samkrash Company for low-power operation. This small, 5cm x 3cm, device is specially designed for interfacing with WSN devices, such as Telos or Mica platform. The maximum detection range is about 10m, and it has 60 angular degree conical pattern for 5.8 GHz signal radiation. The minimum and maximum radial velocities this device can measure are between 2.6 cm/s and 2.6 m/s, respectively.
  • IRobot Create: one of the the most challenging problem in target localization is to provide solid, error-free ground truth values for comparing with your measurements to evaluate your location and velocity estimation accuracy. Human being as a target may not be a good candidate mainly because it is hard to walk or run at the speed you are supposed to move. To overcome this challenge, we use the IRobot create equipped with an embedded Linux box on top of it, which is programmable and controllable. More specifically, the IRobot provides us a set of library for scheduling the robot's trajectories or controlling the robot's speed and orientation via keyboard. On top of this, it can also give us the useful information, such as the distance it has moved or speed the robot has traveled.

Members: