MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202641047835 A) filed by Hindusthan College Of Engineering And Technology on April 15, 2026, for Design And Implementation Of An Approx Log Domain Multiplier Using Neural Network System.
Inventors include Dr. N. J. R Muniraj; Dr. K. A. Pranesh; G. Harish; A. Varshini; and M. Kabilesh.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: ABSTRACT By using deep neural networks, the operation of the multiplication process is much larger and consumes more energy. In order to reduce the multiplier complexity, it is necessary to make the energy much more efficient.. This paper is focused on designing an approximate log domain multiplier. In order to reduce the complexity of the multipliers, there are no computation of the partial products This should be done when the multiplier design is made. This operation is converted to a logarithmic function, the addition hac a very low complexity. The log domain is energy efficient, with a least chip area, the time required to process the multiplication is reduced to minimum. The circuit's complexity is converted to a very low complexity, with no partial products to have a approximate result, to make the trade off.
Disclaimer: Curated by HT Syndication.
