• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site

Development of Hardware Implementation of Neural Networks on FPGA

Student: Americanov Aleksandr

Supervisor: Aleksandr Romanov

Faculty: HSE Tikhonov Moscow Institute of Electronics and Mathematics (MIEM HSE)

Educational Programme: Computer Systems and Networks (Master)

Final Grade: 10

Year of Graduation: 2017

Annotation Report 58 p., 7 sections, 12 pic., 20 sources. Keywords: Neural networks, FPGAs, hardware implementation, neuron, ART. Object of this work is the development of an artificial neural network based on FPGA. The main goal of this work is to develop of mathematical model of a neurons and a neural network, and its implementation based on the Intel FPGA (Altera) on the Verilog in Quartus II programmer software. During the implementation of this work analysis existing various types on neural networks, consider its implementation based on the FPGA. The neural network of the adaptive resonance theory has been chosen and adapted. Model of neurons and architecture neural network have been developed. Based on developed models there were created modules on the Verilog, implementing the logic of the work of neurons based on the on this modules neural network has been designed. The developed neural network was trained for the classification of input vectors on 10 types (digits from 0 to 9), tested and its showed good results of classification of input data with their small noises. Testing was showed a 100% quality classification of input data when they are interference (up to 12%). Developed decision-making and navigating system can be used on the different types of wheeled platforms for development of the various robotic devices, form robot–guides, to robot–cleaner. The implementation of neural networks based on FPGAs gives an opportunity to accelerate their execution, through to better parallelization of calculations than software implementation allows.

Full text (added May 15, 2017)

Student Theses at HSE must be completed in accordance with the University Rules and regulations specified by each educational programme.

Summaries of all theses must be published and made freely available on the HSE website.

The full text of a thesis can be published in open access on the HSE website only if the authoring student (copyright holder) agrees, or, if the thesis was written by a team of students, if all the co-authors (copyright holders) agree. After a thesis is published on the HSE website, it obtains the status of an online publication.

Student theses are objects of copyright and their use is subject to limitations in accordance with the Russian Federation’s law on intellectual property.

In the event that a thesis is quoted or otherwise used, reference to the author’s name and the source of quotation is required.

Search all student theses