Design OFMAC using feed forward neural networks

B.Divya

Author(s)

B.Divya Electronics and Communication Engineering, Vignana Bharathi Institute of Technology, Telangana, India.

Abstract

This research investigates the usage of feed forward neural networks,traditional Indian Vedic multiplier,carry skip adder,parallel in parallel out register and MAC. An Artificial Neural Network (ANN) operates in a parallel information processing structure, comprising processing units. The efficiency of the network is determined by the processing unit. Hence, there is a need to design a processing unit that is both efficient and capable of delivering superior performance. This processing unit encompasses a MAC unit (Multiplication and Accumulation) and an Activation unit. The main focus of high-speed processors is to minimize power consumption and processing times. The implementation of the design utilizes Verilog, a hardware description language, and the testing phase is carried out using the Modelism simulator.The study performance the 4 bit mac using feed forward network with 4 bit mac using logic gates.

Publication Details

Page(s) 1062-1068

DOI DOI not available

Published Journal: International Journal of Communication Networks and Information Security, Volume: 16, Issue: 4, Year: 2024

Request the full text from PASTIC

Keywords

MAC, Artificial Neural Network, Vedic Multiplier, carry skip adder and Verilog HDL

References

. .Energy-efficient neural network design using memristive MAC unit" by Frontiers - This paper describes a novel MAC unit based on mixed-signal multipliers using transistor-memristor cells on crossbar intersections, focusing on energy-efficient neural network design.. , : .

Howard A. G.,Zhu M.,Chen B.,Kalenichenko D.,Wang W.,Weyand T.,Adam H. .2017 .Mobilenets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:1704, : 04861.

. .Design of MAC unit in artificial neural network architecture using Verilog" by IEEE - This paper discusses the design of MAC units in ANN architectures using Verilog HDL. , : .

Dynamic MAC . .-based architecture of artificial neural networks suitable for hardware implementation on FPGAs" - This study presents a dynamic MAC based architecture for ANNs that is suitable for hardware implementation on Field-Programmable Gate Arrays. , : .