A Framework For Adaptive Weather Sensing With Multi-mission Phased Array Radar

Multi-mission phased-array radars (MPAR) can execute a number of tasks that are traditionally carried out by several independent radars; thus, these multiple tasks typically compete for the limited resources that are available on a single radar. As such, it is critical to develop automated algorithms, which can adapt to the continuously evolving and complex environment to manage and allocate available resources in order to optimize radar performance. Previously, a framework for adaptive weather sensing was developed to address the problem of performing automatic tracking of multiple storm cells with fast and independent update times using a single radar. This modular framework that consists of four major functions (storm identification, storm tracking, task management, and scheduling) is a closed-loop system that adaptively defines scanning strategies based on the current radar scene and the desired tracking performance. In this context, the mean improvement factor of revisits was introduced as the weighted sum of individual revisit improvements of each storm in the radar scene, where the weights can be chosen to reflect the priority or importance of each storm. This quantitative measure provides a means to quantify the performance gain afforded by adaptive weather sensing.

In this work, we investigate the selection of optimal update times for various storms, where the mean improvement factor is maximized without exceeding a user-defined acquisition time for the entire radar scene. Specifically, optimal times are obtained by solving a constrained minimization problem. This framework of adaptive weather sensing and the approach for optimal-update time selection are demonstrated and verified using simulations. Preliminary results will be reported, and the limitations of this approach plus plans for future work will be discussed.