1. Minority Leaders Program
Sensors Directorate Fall Program Review
Presented by
Yenumula B Reddy
Grambling State University
Application of Nanotechnology
Principles to Cognitive Radio
24 September 2013
3. 3
Project Members
• AF Program Manager: Mr. Chris Bozada AFRL/RYD
• Project Lead: Yenumula B Reddy
• Students
Sanford Banks – Computer Science (UG) – Graduating in Fall 2013
Alicia Shaw – Computer Science (UG) – Graduating in Fall 2013
Christopher Small – Computer science (UG) – New Entry from Fall 2013
Orlando Elias – Computer Science (UG) – Graduated Spring 2013;
In Graduate School (California State University, Long Beach)
Prior Students:
Patrice Brown, Brandy Guillory, Jessica Harris, Shelton Mathews, Leon Sanders,
Heather Smith,, Loni Taylor, Michael Terrell, Brandon Wright, Shatay Holmes,
Brandon Howard, Nikima Smith, Yves Courtois
4. 4
Problem Statement
2012-13 Project
Conduct Research in application of
nanotechnology to cognitive radio
• Review the literature to identify the current state of nanotechnology
for application to cognitive radio technology
• Learn Neural network concepts and understand the MATLAB Neural
Networks Package
5. 5
Project Objectives
• Identify the potential role of nanotechnology in
addressing the requirements for spectrum sensing for
cognitive radio
• Investigate Role of nanotechnology in complicated
distributed processing
• Study the role of GPU technology in nanocomputing for
real-time applications
6. 6
Approach
• Literature survey
• Integration of Nanotechnology in Wireless Networks
– Tunable Radio Components
– High Frequency Electronics
– Wireless Sensors
• Relation of Nanotechnology, GPU and Cognitive Radio Networks
– Cognitive radio application requires intensive calculations
– GPU computing speeds up the calculation and provide real-time response
Related publications that requires intensive calculations
• SRIS DELIVERS REAL-TIME GEOSPATIAL IMAGE PROCESSING RESULTS,
Innovative GPU-Based Architecture 72x Faster, 12x Cheaper than CPUs
• S. Ihnatsenka, “Computation of electron quantum transport in graphene nanoribbons using GPU”, Comput. Phys.
Commun. 183 (2012) 543—546.
• A. Hafeez, W. Asghar, M.M. Rafique, S.M. Iqbal, and A.R. Butt, "GPU-based real-time detection and analysis of biological
targets using solid-state nanopores", ;presented at Med. Biol. Engineering and Computing, 2012, pp.605-615.
NanoTech-Ref
• Will Soutter, Nanotechnology in wireless Devices, http://www.azonano.com/article.aspx?ArticleID=3183, Jan 29, 2013
7. 7
Approach
GPU Implementation
• GSU Research projects Lab has 12 Dell Precision T5500
Computers. Each computer is configured with two Quadro
4000 cards.
• Each GPU card has 2048 Mbytes of global memory, 256 CUDA cores.
• The GPU clock rate is 950 Mhz, memory clock rate is 1404Mhz, and memory bus width is
256 bit.
• The amount of constant memory is 65,536 bytes and shared memory 49,152 bytes.
• The total number of registers available per block is 32,768, warp size 32, threads per
multiprocessor 1,536, and threads per block are 1,024.
• We have tested many examples including matrix addition and multiplication on windows 7
with Visual studio 10 with NVIDIA tools CUDA 4.5 and CUDA 5.0 on one GPU card.
• We downloaded GPUmat (GPU using MATLB) on windows 7 with visual studio 10 and
tested the matrix multiplication example.
• We downloaded CUDA 5.5 tool kit and planning to test on Visual Studio 12 before
December 2013.
• Planning to test two GPU cards in the Spring 2014.
10. 10
Approach
Review work completed by Students
Nanotechnology and Wireless communications
• Nano Computing and Nanocomputer
• Barriers to Nanotechnology and nanocomputing
• Hardware and Software Barriers
Computational Nanotechnology
• The Vision
• Development in Nanotechnology
• Nanofabrication, NanoComputers, NanoRobots, NanoMadicine
• Potential benefits and Threats
Nanotube Radio
• Benefits of nanotube radios
• Radio components in a Carbon Nanotube
• Nanotube radio structure and Difference in operation
• Functionality of nanotube radio, Tuning resonance frequency
• Results of nanotube radio
11. 11
Approach
Neural Networks - Overview
• NN are constructed and implemented to model the human brain.
• Performs various tasks such as pattern-matching, classification, optimization
function, approximation, vector quantization and data clustering.
• These tasks are difficult for traditional computers
• Artificial Neural networks (ANN) possess a large number of processing
elements called nodes/neurons which operate in parallel.
• Neurons are connected with others by connection link.
• Each link is associated with weights which contain information about the
input signal.
• Each neuron has an internal state of its own which is a function of the inputs
that neuron receives- Activation level
13. 13
Approach
Learning and Training
Learning
• It’s a process by which a NN adapts itself to a stimulus by making proper parameter
adjustments, resulting in the production of desired response
• Two kinds of learning
• Parameter learning:- connection weights are updated
• Structure Learning:- change in network structure
Training
• The process of modifying the weights in the connections between network layers with the
objective of achieving the expected output is called training a network.
• This is achieved through
• Supervised learning
• Unsupervised learning
• Reinforcement learning
Classification of learning
• Supervised learning
• Unsupervised learning
• Reinforcement learning
19. 19
Summary
Accomplishments/Successes
• Students Completed the Survey on nanotechnology for wireless
communications
• During the review of Literature students studied the research papers and
presented in the group; This helps the students to develop power point slides
and confidence in presentation of work.
• Students understands the current status of nanotechnology in various fields
and their functionality
• Students understood the limitations of current host memory and future need of
multi-core systems using nanotechnology for real time applications
• Learned Neural Networks Concepts and how to use MATLAB tool-kit
• Future study requires:
– Nano-computing based Machine learning algorithm for channel classifier
– GPU based channel classifier model
20. Current Research
Papers Completed in Spring
• “Security Issues and Threats in Cognitive Radio Networks” AICT 2013, June 24-28,
2013, Rome Italy.
• “Modeling Cognitive Radio Networks for Efficient Data Transfer Using Cloud Link”,
ITNG 2013, April 16-8, Las Vegas
Book
• Book: one of the editor, “Cognitive Radio Technology Applications for Wireless and
Mobile Ad hoc Networks”, June 2013
Book Chapters
• “Nanocomputing in Cognitive Radio Networks to Improve the Performance”, in
Cognitive Radio Technology Applications for Wireless and Mobile Ad hoc Networks,
June 2013
• “Application of Game Models for Cognitive Radio Networks”, in Cognitive Radio
Technology Applications for Wireless and Mobile Ad hoc Networks, June 2013
Research Papers in progress
• Security in Hadoop Distributed File Systems
• Security in Cloud Computing
20
21. 2121
Status Summary
Project Schedule
CTC#: FA8650-05-D-1912
Project Name: Nanotechnology Research for
C4ISRand EW
Dec-12
Jan-13
Feb-13
Mar-13
Apr-13
May-13
June-13
July-13
Aug-13
Sept-13
Oct-13
Nov-13
Dec-13
Jan-14
Feb-14
Mar-14
Overall Project: Nanocomputing Solution for CRN
Project Step1
Literature Review
Project Step2
Understanding Neural Networks (Fundamentals and
MATLB tool).
Project Step 3
Q1 Q2 Q3 Q4 Q5
Today