Online evolution of image filters based on dynamic partial reconfiguration of FPGA

Evolvable Hardware (EHW), the integration of programmable logic device and evolutionary algorithm (EA), can change its own structure and behavior according to the changing environment, so as to obtain capabilities of self-adaption and fault-tolerance. To solve the problems of huge area overhead and circuit delay in EHW based on the virtual reconfigurable circuit, the technology of dynamic partial reconfiguration of modern SRAM-based Field Programmable Gate Arrays (FPGAs) is applied to EHW to implement online evolution in this paper. Based on the technology of early access to partial reconfiguration provided by Xilinx, an evolvable IP core is customized. With EA implemented on the on-chip soft processor, a self-evolving system based on system on a programmable chip is designed for the online evolution of digital image filters. The Xilinx FPGA development tool, ISE12.4, is used to complete the design, and the Xilinx Virtex-5 FPGA development board, ML507, is used to verify the feasibility of the system structure and the self-evolving mechanism.