 |
 |
 |
 |
| |
|||||
|
|
|
|
|
||||
An Eye on the Sky: Detecting and Identifying Airborne Threats with Netted Sensors By Weiqun Shi, Ronald Fante, John Yoder, and Gregory Crawford "Look to the skies!" This warning has long been popular in both science fiction movies and military doctrine. However, technological advances now allow for the development of unmanned aerial systems, fixed-wing aircraft, and cruise missiles that are small enough and fly low enough to elude conventional radar detection. Such aircraft could carry out chemical, biological, or nuclear attacks, or they could be employed to smuggle drugs or illegal immigrants across the border. There currently is no effective, reliable, mobile, and low-cost solution for countering this threat. However, the Netted Sensors Initiative is working on just such a solution: the Multi-Modal Netted Sensor Fence. This easily deployable system will provide a low-cost, low-power method for protecting civilian and military targets from airborne threats around the clock. MITRE testing has already demonstrated the Sensor Fence's ability to detect small targets flying at both high and low altitudes; classify them as "threat," "potential threat," or "no threat" when possible; estimate the targets' speed and heading; and quickly communicate this information to a central processing station. Triple Protection The Sensor Fence comprises three sensor modalities: a radar fence, an acoustic microphone array, and infrared and visible-light cameras. Working together, these modes detect, track, and discriminate airborne targets. The radar fence's first task is to detect approaching targets. The radar fence consists of multiple, low-power radar units whose ranges are separated by approximately four kilometers. Each radar operates with a different carrier frequency to avoid crosstalk between radars, and each has a beam width that is broad in both azimuth (so that the number of radars can be kept small) and elevation (to detect both high and low-flying targets). The radars measure target range and report these values to a central processing station. When the presence of a target is confirmed, the station cues the acoustic and infrared sensors to identify the target. The Sensor Fence contains several equally spaced, 8-element, diagonally arranged microphone arrays between the radar sensors. The microphone arrays comprise acoustic sensors that are non-line-of-sight, passive, low-cost, portable, and can be deployed over wide areas. The task of the acoustic arrays is to detect broadband emissions from approaching targets. From these emissions, the arrays can determine the target's direction of arrival, supply a target identification and classification, and mitigate false alarms. The third sensor modality in the fence is an optical system. Once the radar and acoustic arrays have detected and confirmed a target, the central processing station fuses the radar and acoustic data to form a target track. The optical system is then automatically aimed to photograph and identify the target. The optical system contains an infrared detector array sensitive to the 8-12 µm waveband, which allows it to operate during both day and night at the cost of only a few watts of power. A visible camera is also used for improved target resolution during the daytime. Once all three sensor modes have reported back their findings, the central processing station transmits the target track and identity to rear-area weapons systems or potential interceptors so that defensive action can be taken.
Sizing Up the Sensors To put this Sensor Fence concept to the test, MITRE has developed an experimental prototype consisting of three remote sensor nodes and a central processing node. The remote nodes contain radar and acoustic arrays. Each node also contains a computer that performs target range, azimuth, and elevation detection. The remote sensor nodes make an initial target classification based on the target's radar and acoustic signature and report their findings to the central node. The central processing node contains a camera and a camera-steering device, as well as a central computer equipped with a reporting and target classification results display that performs data fusion and final target classification. System connectivity is provided through a simple point-to-point wireless communications network. The compact and modular design of the Sensor Fence system allows for quick and inexpensive production of nodes and for rapid deployment of the system. To test the prototype, we are using light weight civilian aircraft, such as Cessna, Beechcraft, and crop dusters, whose radar, acoustic, and visual signatures are well studied. In the field studies, all three modes in the Sensor Fence prototype have exhibited reasonable ranges in detecting the test planes. A multi-modal kinematic tracker is employed to predict the location, speed, and heading of the target. The tracker's test results reveal a reasonable ability to discriminate between the different test aircraft during the field experiment. Just as the Multi-Modal Netted Sensor Fence will tirelessly look to
the skies, MITRE will continue looking to the far reaches of technology
in search of better ways to protect our nation.
|
|
||||
| For more information, please contact Weiqun Shi, Ronald Fante, John Yoder or Gregory Crawford using the employee directory. Page last updated: April 28, 2006 | Top of page |
|||||
Solutions That Make a Difference.® |
|
|
Download this issue [4MB]