Pros And Cons Of The Radar Detection System

Most security experts agree that there is a growing need to be able to detect an incoming drone and then disarm or disable one that is judged to be harmful. Engineers are working on uploadable software, as well as GPS jamming and spoofing techniques, in order to disable small UAVs, but the initial problem is that of detection and identification. The small drones cited above went undetected. The radar detection system installed at the White House failed to detect either of the two small UAVs that flew into its airspace (Naboulsi, 2015). Some security analysts, like those at the RAND corporation, recommend prevention rather than detection/disruption, simply because the problem of detection is so difficult (Jackson, 2008). Both the United States

Directional microphones are installed encircling the area of interest and ambient noise levels are measured (DroneShield, 2015). This research will use a combination of 16 omnidirectional and 16 parabolic dish sensors (each placed side by side with each pair equally spaced around the area of interest. This microphone setup would then gain the advantages of both increased field of view and increased detection range (Birch, 2015). Once fielded, the DroneShield then takes acoustic samples and compares the sampled noise with database values for normal ambient noises (which would include aircraft flying nearby, or birds flying past) and database noise signatures of typical small unmanned rotor

Phase 1 will involve quantitative data collection with a follow on qualitative phase 2. This method is appropriate since it will allow the researcher to collect and analyze numerical data, then use a qualitative approach to better understand the statistical results and to make conclusive recommendations.

In phase 1, the DroneShield system will be set up, initial conditions uploaded and checked, and then testing will involve three locations types: a rural flat environment, rural hilly, and a dense urban tall building environment. Next, four types of small UAVs will be flown into the detection zone at various altitudes and from various directions. This research will use the DJI Phantom 2, the Parrot, and the AEE F100, and a homebuilt UAV (one with a unique noise signature). These drones were chosen due to their current payload carrying capacities which would allow them to either carry high fidelity sensors or a small explosives package. Numerical data will be collected to indicate whether the UAV was successfully detected, how far it was when detected, and how different the noise levels were to database values. In the follow-up phase 2, a qualitative look at the data for each DroneShield misidentification will be completed in order to ascertain how the software algorithms or database could be improved to enhance the accuracy rate of the equipment. Drone Shield technicians will be surveyed to generate