Research And Design A Configurable Processor For Elliptic Curve Cryptography Algorithm

With the rapid development and wide application of computer and network, information security is given more and more attention. As new public-key cryptography, elliptic curve cryptography (ECC) has many excellent attributes: short key length, fast speed, low transport bandwidth and high security. These attributes make it an ideal choice for application. This paper researches and implements a configurable processor for ECC based on transport triggered architecture. This processor can obtain a good compromise between speed and area and better meet the real needs, compared to the traditional implementation, such as application specific integrated circuit, FPGA and general-purpose processor.Firstly, the scalar multiplication algorithm, the core operations of ECC, is analyzed and optimized in this paper, including coordinate of the curve point, the data flow’s planning over operations of point addition and double point, optimization of modular multiplication algorithm and so on. Then combined with features of TTA’s hardware and software, the whole ECC processor is customized, including instruction format, pipelining, control path, data path and each functional unit (FU). As key computing components of processor, FUs directly determine the overall performance of the processor, so this paper designs different FUs for the specific ECC algorithm, such as modular multiplication precomputed unit, modular multiplication accelerate unit, modular addition and modular subtraction unit, arithmetic logic unit, load_store unit and jump unit. When processing the scalar multiplication, the speed will be increased by using those FUs.In the hardware verification of ECC processor, the paper firstly insures that the function of each module is right, then writes application program to simulate and compares it with the C module. The result shows that the processor can compute scalar multiplication correctly. Finally, the process is synthesized and it shows the processor uses 90Kgates, runs at 105MHz clock frequency and is able to complete 427 times scalar multiplication in one second. Compared to the traditional implementations, this processor has a better holistic performance.