Improved allocation algorithm schedules radar tasks parallel

Traditional single-computing is inefficient to meet the instantaneity requirement with the increasing of radar tasks´ complexity; cluster computing is widely envisioned as a powerful paradigm for scheduling radar tasks parallel. An improved allocation algorithm is proposed for radar task allocation in distributed heterogeneous system. The continuous detection of task information of processors is put forward. Furthermore, task accumulation threshold is defined to estimate the processor is overloaded or not. In particular, the task will be allocated to the processors idle or light loading with the workload according to threshold. Simulation results show that executed time is reduced greatly and load balancing is improved immensely.