A scalable architecture of high-performance Montgomery multiplier for design reuse

This paper describes a new scalable architecture of Montgomery modular multiplier (MMM) using our improved FIOS algorithm. This algorithm and architecture can manipulate operands of any precision (bit length) in dual field (prime field and binary field) and has the advantage of using relatively smaller latency (clock cycles) to complete one modular multiplication. Our architecture is reusable and has high performance with respect to latency, timing, area, etc.