Oscillator-Based Computing



Sep 2013 ~ Present

This large collaborative project aims at helping blind people or those who have visual impairment by developing AI systems with technologies of computer vision and machine learning. This wearable intelligent system will process the video data from the camera and perform image analysis, feature extraction, pattern recognition like human visual cortex. It will interact with users and provide supportive valuble information to help their daily activities.



The partners and details of this project can be found on its official website. Our group is focused on the hardware design of the image processing pipeline, especially for pattern matching. We are developing circuits, architectures, and algorithms around nano-oscillators, such as spin torque oscillators (STO), FinFET Thyristor oscillators. When we couple these oscillators together, their synchonization and desynchronization can perform pattern matching.

Furthermore, we can build associative memory or neural networks based on oscillators, since the basic neuron model can be abstracted and viewed as a relaxation oscillator (see Neural Oscillation). We are also looking at the implementation of a spiking neural network model.

Meanwhile, our colleagues from UCSD and Stanford University are designing neuromorphic systems of spiking neural network with emerging nano-device technologies. For more information, see the websites of

Dr. Gert Cauwenburghs

Dr. Philip Wong



Sensing and Computing with Oscillating Chemical Reactions
Sep 2013 ~ Present


Anna C. Balazs (PI); Steven P. Levitan (Co-PI) (Univ. of Pittsburgh), NSF DMR-Award # 1344178

Our goal is to develop materials that compute by using non-linear, oscillating chemical reactions. We focus on polymer gels undergoing the oscillatory Belousov-Zhabotinsky (BZ) reaction. The novelty of our approach is in employing hybrid gel – piezoelectric MEMS to couple local chemo-mechanical oscillations over long distances through electrical connections. Our modeling revealed that: (1) the interaction between two such units is sufficiently strong to yield synchronization of the gels' oscillations; (2) the mode of synchronization is determined by the polarity of the connection; (3) each mode has a distinctive pattern of oscillations and generated voltage. The results indicate the feasibility of using the hybrid gel-piezoelectric micro-electro-mechanical systems (MEMS) for oscillator-based unconventional computing.


Ultra Low Power Non-Boolean Systems
Aug 2011 ~ Aug 2013
This project was a feasibility study of building a non-Boolean computing system with current emerging nano-device technology, funded by Intel Labs University Research Office. The motive of this project came from a series of reports on emerging nano-devices by Intel Labs that try to address the scaling problem of CMOS technology. However, these novel devices failed to outperform traditional CMOS technology in the general Boolean logic computing system. Nonetheless, due to their nonlinearity and multiple stable state, we notice the potential of these new devices in an unconventional computing system without using Boolean logic, and the applications include pattern recognition, neural network, image processing and so on.

This short research project was successful as a proof of concepts and feasibility exploration, which lead to several successive research projects and grants. During this project, PhD candidate Yan Fang finished his M.S. thesis, where a tree structure hierarchical associative memory was proposed based on nano-oscillator network. From there, we started our research on oscillators.

Image Processing Example: Edge Detection Using Local Coupled Oscillator 2-D Array






Some of our work will be published by Wiley as a chapter in a new book: “Emerging Nanoelectronic Devices”. Our collaborators at HRL, Intel, and Stanford are also featured in this volume.


Papers



1. Steven Levitan, Yan Fang, Donald M. Chiarulli, "Using Analog Memory With Coupled Oscillators for Pattern Recognition Applications," 5th Non-Volatile Memory Workshop, San Diego, CA, March 9-11, 2014.

2. Vijaykrishnan Narayanan, Gert Cauwenberghs, Donald M. Chiarulli, Suman Datta, Steven P. Levitan and Philip Wong "Video Analytics Using Beyond CMOS Devices", Design Automation & Test in Europe (DATE 2014), Dresden, Germany, 24-28 March 2014.

3. Stefano P Coraluppi Craig Carthel Samuel J. Dickerson, Donald M. Chiarulli, Steven P. Levitan, "Multiple-model filtering for particle tracking and classification," (Paper 9092-13) Signal and Data Processing of Small Targets, SPIE Defense + Security, Baltimore, Maryland May 5-9, 2014.

4. Yan Fang, Matthew J. Cotter,Donald M. Chiarulli, Steven P. Levitan, "Image Segmentation Using Frequency Locking of Coupled Oscillators",51st Design Automation Conference (DAC '14), (Work in Progress Poster), San Francisco, CA, June 1-5, 2014.

Updated October 14 2014|Contact Us
website template by Free HTML5 Template