Ship speed estimation using Wireless Sensor Networks: Three and five sensors formulation

Abstract

Intrusion detection on the sea is an important surveillance problem for harbor protection, border security, and commercial facilities such as oil platforms, fisheries facilities and other marine wealth. Widely used methods for ship detection are using radar or satellite which is very expensive. Besides the high cost, the satellite image is easy affected by the cloud. And it is difficult to detect small boats or ships on the sea with marine radar due to the noise or clutters generated by the uneven sea surface. In this paper, we propose ship speed estimation by taking advantage of ship-generated wave's characteristics with Wireless Sensor Network (WSN). We use a grid fashion for sensor node deployment that can be clustered into three and five sensors. We propose the ship speed formulation for each type of claster. We use three sensors, we may expect to improve energy efficiency by involving small number of sensor for detection. We use five sensors, we may expect to improve accuracy of detection. We also propose an algorithm for detection by incorporating individual sensor detection. The individual sensor detection produces a time stamp that records the ship-generated waves intruding the sensors.