Better ground radar to find land mines

One of the more poignant legacies of war are the land mines the warring armies leave behind. The conflict may have been settled, the armies demobilized and sent home — but innocent farmers tilling the land or children playing a game of soccer keep losing limbs and life to mines. The reason is that long after a conflict, land mines remain buried underground unless someone take action to locate and remove or detonate them. According to the UN, there are more than 100 million land mines buried in 68 countries around the world. The UN estimates that more than 2,000 people are killed or injured by land mine explosions each month.

Dominic Ho, a University of Missouri engineer, has decided to do something about it, and he is working to enhance the accuracy of a land mine radar system while minimizing the number of false alarms it produces. In a land mine radar system, ground-penetrating radar scans the surface for underground objects. Besides sensing land mines, the radar also has undesirable responses from clutter objects, such as scrap metal debris, plant roots, and rocks. Ho, the Dowell Professor of Electrical & Computer Engineering in the MU College of Engineering, is working with Army employees and private defense contractors to enhance the system, and distinguish between true positive signals that are from land mines and false positive signals that are from clutter objects and can be ignored safely.

“The fewer false positives there are, the faster we can clear a mine-infected area,” Ho said. “Each time there is a false positive, the military wastes time and money to investigate the positive signal. Our goal is to keep the rate of detection high, but reduce the number of false triggers.”

Ho studies and compares how signals are reflected from land mines and clutter objects that produce false signals. He differentiates the signals and then develops radar signature patterns that can identify an underground object. Ho’s goal is to create consistent radar signature patterns for land mines so the detector will be able to discriminate between land mines and clutter objects. However, land mines can vary in shape and material making it a challenge to create consistent signature patterns. For example, land mines can be circular or square and made of plastic or metal. Plastic land mines are more of a challenge because they are light and do not create a strong signal, Ho said.

Ho has researched land mine detection for nearly ten years. His recent grant will fund research of a detector that is mounted on a vehicle-based platform. Previously, Ho researched and helped enhance a hand-held detection device that is currently being used in Iraq and Afghanistan. His research, along with advancements in radar technology, has contributed to a reduction of false alarms from a rate of 100 per 5,000 square meters five years ago to the current rate of less than four per 5,000 square meters.

“We are chasing a moving target,” Ho said. “People are still creating land mines. It’s very important that we continue to refine our techniques and improve our detection methods.”