Research And Implementation Of Compressed File Password Recovery Algorithm Based On Multi–FPGA

With the increasing number of compressed documents on the Internet,compressed documents of encrypted involving secret information are increasing with the increasing awareness of information security.Therefore,the decryption of encrypted compressed documents is of vital importance to information security.At present,the mainstream compression software of the Internet includes WinRAR(RAR3 and RAR5),WinZip and 7-zip,and their encryption of information mainly takes SHA-1 and SHA-256 as the core of the algorithm.Algorithms such as AES-128?AES-256 and CRC32 are used as checksum encryption to improve security and enhance the difficulty of deciphering.The traditional cracking methods mainly include CPU-based general server cracking.The computing speed is limited by the CPU architecture.The computing resources are limited by system scheduling,and the instructions are serially executed,which cannot realize high-speed parallel computing.The GPU platform limits the deciphering performance due to the memory access problem,and its architecture is fixed,the flexibility is small,and the power consumption is high.Therefore,computing platforms that require higher performance find the correct password in a limited amount of time.The main work of this thesis is as follows:1.The architecture,computing resources and functional characteristics of FPGA are deeply studied.FPGA has unique reconfigurable characteristics,high processing power and perfect software development environment.Therefore,FPGA is a suitable computing component for password recovery algorithm.2.This thesis analyzes the PMC characteristics of RAR3,RAR5,WinZip and 7-zip password recovery algorithm,and proposes four decryption algorithms based on FPGA.High-density computing requirements of the algorithm are mainly concentrated in sha-1,sha-256 and data splicing,requiring repeated operations for many times.At the same time,the algorithm has high storage requirements and communication requirements.Therefore,combined with the computing characteristics and structural characteristics of the FPGA,the different parts of the algorithm can be optimized accordingly to achieve high-performance computing of the algorithm.3.Four compressed file password recovery algorithms are implemented based on FPGA design and optimized.By analyzing the SHA-1 and SHA-256 algorithms,the pipeline algorithm is used to implement the hash algorithm,and the pre-calculation and carry saving adder are used to optimize the critical path of the algorithm,thereby improving the throughput of the algorithm.The data splicing part is implemented by a loop-based queue method,which forms a complete pipeline structure with the hash algorithm and improves the performance of the decryption algorithm.At the same time,the dual-port RAM is used to store data that needs to be reused,and the FIFO is used to buffer data to isolate different clock domains,thereby improving the storage requirements of the algorithm.Finally,the overall implementation structure of the algorithm is also optimized to improve the resource utilization of the FPGA.4.The implementation effect and experimental results of the algorithm are analyzed.The throughput of the hash algorithm SHA-1 is 160.256 Gbps,and the throughput of SHA-256 is 110.080 Gbps.Due to different decryption algorithms,the crack speed is different.Crack 10-byte password,RAR3 is 20145 per second,RAR5 is 47056 per second,WinZip is 1149358 per second,7-Zip is 12640 per second,and 7-Zip crack performance is 226 times of the CPU at the same time,1.4 times of the GPU,and the energy efficiency ratio is 8 times that of the GPU,which indicates that the performance of the optimized algorithm has been greatly improved,and the power consumption is low.