RTL to GDS II Simulation of SAD algorithm with Kogge stone adder using Cadence tools

The effective method used for image recognition or motion estimation is the sum of the absolute difference algorithm. The obtained absolute difference value from the two blocks is added. The adder plays a crucial role in determining speed of SAD algorithm. Previous research has utilized various adder techniques such as adder tree and adder compressor. Initially, an adder tree was proposed to simplify the addition process of the intermediate SAD algorithm value, which is obtained by calculating the absolute difference between the video blocks. However, it was later suggested to use a high-speed adder compressor instead of the adder tree, as it provides better performance. This paper presents the SAD algorithm using the efficient and high-speed Kogge Stone adder. The design is implemented using Verilog hardware description language. The RTL design to the GDS II simulation of the SAD algorithm is implemented using the cadence tool.