Optimized Polynomial Multiplier Architectures for Post-Quantum KEM Saber

Saber is one of the four finalists in the ongoing NIST post-quantum cryptography standardization project. A significant portion of Saber's computation time is spent on computing polynomial multiplications in polynomial rings with powers-of-two moduli. We propose several optimization strategies for improving the performance of polynomial multiplier architectures for Saber, targeting different hardware platforms and diverse application goals. We propose two high-speed architectures that exploit the smallness of operand polynomials in Saber and can achieve great performance with a moderate area consumption. We also propose a lightweight multiplier that consumes only 541 LUTs and 301 FFs on a small Artix-7 FPGA.