高级检索

    可重构分组密码处理结构模型研究与设计

    Research and Design of Reconfigurable Computing Targeted at Block Cipher Processing

    • 摘要: 随着信息技术的发展和网络规模不断扩大,网络通信等应用对数据加解密处理提出了更高的要求.可重构计算是将可重构硬件处理单元和软件可编程处理器结合的计算系统.因此采用可重构计算技术来设计密码处理系统,使同一硬件能够高效灵活地支持密码应用领域内的多种算法.同时满足了密码处理对性能和灵活性的要求,提高了密码系统的安全性.论文在分析分组密码算法处理结构的基础上,结合了可重构结构的设计思想和方法,提出了一种可重构密码处理结构模型RCPA,并基于该模型实现了一款验证原型.原型在FPGA上成功进行了验证测试并在0.18μm CMOS工艺标准单元库下进行逻辑综合以及布局布线.实验结果表明,在RCPA验证原型上执行的分组密码算法都可达到较高的性能,其密码处理性能与通用高性能微处理器处理性能相比提高了10~20倍;与其他一些专用可重构密码处理结构处理性能相比提高了1.1~5.1倍.结果说明研究的RCPA模型既能保证分组密码算法应用的灵活性又能够达到较高的性能.

       

      Abstract: With the development of the information technology and network communication, the increased security demands should be satisfied in the network communication applications. Reconfigurable computing is a novel computing system which can combine the reconfigurable hardware processing unit and the software programmable processor. The design of a cipher processing system adopts reconfigurable computing technology, which can support multiple cryptographic algorithms in the cipher application. Therefore, it can achieve crypto algorithms processing with efficiency and flexibility, and it also solves the hidden trouble in the cipher processing system. Based on the analysis of processing structure characteristics of popular block cipher algorithms, the authors propose a reconfigurable cipher processing architecture (RCPA) using the design method of reconfigurable processing architecture. And a prototype is implemented based on RCPA. The prototype is realized using Altera’s FPGA. The logic synthesis, placement and routing of RCPA are accomplished using 018μm CMOS technology. The experiment results indicate that on the prototype based on RCPA, the performance of many block ciphers is 10~20 times higher than on highperformance general purpose processor and 1.1~5.1 times higher than on other specialized reconfigurable frameworks. The results prove that RCPA can guarantee high flexibility for most of the block cipher algorithms and can achieve relatively high performance.

       

    /

    返回文章
    返回