A 4-Stage Pipelined Architecture for Point Multiplication of Binary Huff Curves

Abstract

This work has proposed a 4-stage pipelined architecture to achieve an optimized throughput over area ratio for point multiplication (PM) computation in binary huff curves (BHC) cryptography. The original mathematical formulation of BHC is revisited with an objective to reduce the required area. Consequently, a simplified formulation of BHC is obtained with 43% reduction in the hardware resources. As far as the throughput is concerned, it is improved first by reducing the critical path and second by minimizing the number of clock cycles (CCs) required to compute one PM. The critical path is reduced through the placement of pipeline registers, whereas the number of required CCs are minimized through an efficient scheduling of computations. These two factors i.e., the area reduction and throughput optimizations, have resulted in maximizing the throughput over area ratio. The proposed pipelined architecture is implemented over [Formula: see text] field, using standard NIST curve parameters. The architecture is modeled in Verilog and synthesized using Xilinx (ISE 14.7) design tool on Virtex 7 FPGA. The implementation results show that 17% improvement in clock frequency, 13% reduction in the time required to compute one PM and 2.6% improvement in throughput/area are achieved when compared with the most recent state of the art solutions.