Event-Based Phased Array Signal Processing

Abstract

In the last few years video-based techniques for identifying the dynamic properties of structures have become of increasing interest. These techniques offer the possibility of capturing high-resolution structural dynamics information at very low cost. In recent years event-based imaging technology has also become commercially available. Event-based imagers differ from conventional imagers in that they only report data for pixels whose intensity value has changed. The result is that these imagers can capture high-speed phenomena with relatively low data rates. The data rate of the imager essentially automatically adapts to the speed at which event that it observes are occurring. As a result the eventbased imager is very efficient. In this work we begin exploring the possibility of adapting techniques from the field of phased-array signal processing to the data captured from event-based imagers. In this case we consider each pixel of the event-based imager to potentially be a sensor in the array. The goal of this work is to eventually develop high-speed, event-based image processing algorithms that can be used to help monitor structural integrity. These techniques could also be applied to emerging types of eventbased signal processing sensors such as ultrasonic sensors that could be used to enable low-cost, low-power ultrasonic phased arrays.