A Starting-Point-Congestion Method for Over-Loaded Radar Task Scheduling

This paper proposes a novel scheduling method, the Starting-Point-Congestion (SPC) algorithm, designed to address the radar task scheduling problem in over-loaded scenarios. The SPC algorithm determines the mobility and urgency of tasks based on their priorities and crowdedness. All tasks are initially positioned at the start of the time window, awaiting reordering via the Random-Shifted-Start-Time (RSST) technique. The RSST process is constrained by the previously determined task mobility, meaning tasks with higher urgency (and thus lower mobility) have a greater likelihood of remaining within the time window and securing execution positions. Conversely, tasks with lower urgency have higher mobility, increasing the chance of them being moved out of the time window. Numerical simulations show that the schedule produced by the SPC algorithm is up to 13.1 times less costly than that generated by the Earliest-Start-Time (EST) algorithm, and up to 2.8 times better than using the RSST algorithm alone under certain loading conditions. The computational time of SPC is maintained below 8 milliseconds, thus it is considered both efficient and practical for real-time radar missions.