Design of an energy efficient approximate BinDCT module in quantum cellular automata

Abstract

The quantum cellular automata (QCA) paradigm offers an ultra-low-power approach for realizing nanocomputing circuits at the molecular level, offering high parallelism capabilities. This study introduces a coplanar and energy-efficient implementation of the approximate binary discrete cosine transform (BinDCT) module using QCA technology. The proposed BinDCT module integrates various sequential and combinatorial submodules, including multiplexers (MUXs), demultiplexers (DeMUXs), parallel-in-parallel-out right-shift registers (PIPO-RSRs), ripple carry adders (RCAs), and ripple borrow subtractors (RBSs). Each submodule is systematically designed following the standard single-layer design principles, which are crucial for maximizing circuit performance, enhancing reliability, and minimizing power dissipation. Extensive simulations were conducted to validate the logic operation and energy dissipation of each submodule. The simulation results demonstrate a significant reduction in power dissipation- up to and an improvement in circuit area efficiency by compared to previous QCA implementations.