Study On The Regular Expression Matching Technology Of Collaborative Design Between Software And Hardware

Regular expression matching technology is one of the key technologies in network security system. With popularization and rapid development of the Internet, high speed growth of network bandwidth and emergence of Internet new business have put forward higher requirements to the regular expression matching technology. In order to improve the processing abilities of regular expression, this paper proposed improved method aimed at shortcomings of existing algorithms and architectures.As for algorithm, this paper proposed an improved DFA matching algorithm completely based on FPGA logic circuit to solve the problem that DFA algorithm occupies too many memory resources, through selection of reasonable hardware algorithm and optimization design based on platform specific algorithm. This algorithm has been used in the project practice to verify its practicability. In terms of architecture, this paper select the architecture of "general processor and coprocessor", and build a collaborative processor platform between software and hardware in the latest Zynq7000 series embedded chip. The heartbeat testing function of processor platform can detect whether the board is normal; IP address configuration function based on platform can modify the IP address of the board; the platform can be used to achieve the efficient operation of rules compilation and update.Based on the collaborative processor platform between software and hardware, ARM processor and Linux are responsible for scheduling and communication, coprocessor(FPGA) implement complex matching function, thus realizing the combination between high performance of hardware and flexibility programming of software. Finally, this paper use the corresponding test cases according to GTEST framework to verify the performance of the system.