Department of Computer Science and Electrical Engineering

EE Graduate Seminar

Energy Efficient and High Performance Architectures
for DSP and Communication Applications

Tinoosh Mohsenin, PhD
Assistant Professor of Computer Engineering
CSEE Dept/UMBC

11:30am-12:45pm, 9 March 2012, ITE 231

Many emerging and future communication applications require a significant amount of high throughput data processing and operate with decreasing power budgets. This need for greater energy efficiency and improved performance of electronic devices demands co-optimization of algorithms, architectures, and implementations. This talk presents several design projects that illustrate the cross-domain optimization.

The design of System-on-Chip (SoC) blocks becomes increasingly sophisticated with emergent communication standards that have large real-time computational requirements. Two such algorithms, Low Density Parity Check (LDPC) decoding and Compressive Sensing (CS), have received significant attention. LDPC decoding is an error correction technique which has shown superior error correction performance and has been adopted by several recent communication standards. Compressive sensing is a revolutionary technique which reduces the amount of data collected during acquisition and allows sparse signals and images to be recovered from very few samples compared to the traditional Nyquist sampling. While both LDPC decoding and compressive sampling have several advantages, they require high computational intensive algorithms which typically suffer from high power consumption and low clock rates. This talk presents novel algorithms and architectures to address these challenges.

As future communication systems demand increasing flexibility and performance within a limited power budget, multi-core and many-core chip architectures have become a promising solution. The design and implementation of a many-core platform capable of performing DSP applications is presented. The low power and low area core processors are connected through a hierarchical network structure. The network protocol includes contention resolution for high data traffic between cores. The result is a platform with higher performance and lower power consumption than a traditional DSP with the ease of programmability lacking in an ASIC. Early post place and route results from a standard-cell design gives processor areas of 0.078 mm2 each using TSMCs 65 nm.

Dr. Mohsenin received the B.S. degree in electrical engineering from the Sharif University of Technology, Iran and the M.S. and PhD degrees in electrical and computer engineering from Rice University and University of California Davis in 2004 and 2010, respectively. In 2011, she joined the Department of Computer Science and Electrical Engineering at the University of Maryland Baltimore County where she is currently an Assistant Professor. Dr. Mohsenin's research interests lie in the areas of high performance and energy-efficiency in programmable and special purpose processors. She is the director of the Energy Efficient High Performance Computing (EEPC) Lab, where she leads projects in architecture, hardware, software tools, and applications for VLSI computation with an emphasis on DSP workloads. Dr. Mohsenin has been consultant to early stage technology companies and currently serves as a member of the Technical Program Committees of the IEEE Biomedical Circuits & Systems Conference (BioCAS), the Life Science Systems and Applications Workshop (LiSSA), and IEEE Women in Circuits and Systems (WiCAS).

Host: Prof. Joel M. Morris

Military personnel are facing an increasing security threat posed by their smart phones and other mobile devices, argues an article on DefenseSystems.com. While the potential of mobile devices to leak personal secrets has been a cause for concern for all, the issue has the potential to be devestating for soldiers who could unwittingly leak crucial information to the enemy through compromised networks.

In the article, our very own Richard Forno, graduate program director for cybersecurity, shares his own concerns about cloud safety:

“One key area of emerging concern is data-in-motion within a cloud — i.e., ensuring that data is protected as it transits and/or exists in multiple servers at the same time, and by extension, the issue of ‘availability’ of data in a cloud environment,” Forno said. “The more moving parts you have to deal with, the easier it is to gum up the works, inadvertently cause self-inflicted problems, or make it easier for an adversary to do the same thing.”

Check out the entire article: "Smart phones pose emerging security threat."

Congratulations to CSEE Professor Tulay Adali, who has been named a 2012 IEEE Signal Processing Society Distinguished Lecturer. Nominated by the Machine Learning for Signal Processing Technical Committee, Dr. Adali is one of only five Distinguished Lecturers appointed this year.

The position commits Dr. Adali to travel around to world to present her current research, which focuses on data-driven and complex-valued signal processing and their applications in medical image analysis.

Her lectures will revolve around the following topics:

• Data-driven Analysis and Fusion of Medical Imaging Data
• Complex-valued Adaptive Signal Processing: When and How to Take Noncircularity into Account
• ICA, ISA, and IVA: Theory, Connections, and Applications in Medical Image Analysis
• Optimization in the Complex Domain using Wirtinger Calculus: Applications to ICA
• Joint Blind Source Separation: Applications in Medical Image Analysis

“My research group, the Machine Learning for Signal Processing Lab [MLSP-Lab], has been conducting research in two of the most active areas in my field: data-driven signal processing and medical image analysis and fusion,” explains Dr. Adali. “I am looking forward to telling a wider audience than I have in the past about the exciting research results we have, as well as better introducing these important areas to new audiences.”

The appointment will last from January 1, 2012 until the end of December 2013.

Dr. Finin has been a faculty member of UMBC’s Computer Science and Electrical Engineering Department for over twenty years. A member of UMBC’s Ebiquity research group, Dr. Finin’s research is in Artificial Intelligence.

Congratulations to Dr. Tim Finin, who was just named this year’s Presidential Research Professor. The appointment, which lasts from the beginning of July 2012 through June 2015, is awarded to faculty members whose outstanding scholarship and excellent teaching have stood out at UMBC.

“I’m very honored to be selected,” says Dr. Finin of the award. He credits his research success to the collaborative research environment at UMBC and the talented students and professors that he has worked with over the years. “I feel like I’ve been lucky to be here at UMBC because having a good set of colleagues and students to work with is the reason for [this] success.”

Through his research in Artificial Intelligence, Dr. Finin is constantly searching for answers to the question: “How can we make [software] systems more intelligent?” He has applied his research to the increasingly popular areas of Mobile Computing, Social Computing and Security. Recently, Dr. Finin has been working on a project that looks at the potential of smartphones to understand a user’s context. The project– a collaborative effort with fellow CSEE professor, Dr. Joshi–is being sponsored by the National Science Foundation (NSF). More information about Dr. Finin’s research can be found in his research profile.

Finin’s appointment as Presidential Research Professor, comes with a $2,500 allowance, and a $2,500 per semester gift to the Computer Science and Electrical Engineering Department, to be used towards enhancing teaching and research. A formal award ceremony will take place this Spring to celebrate Dr. Finin and the other recipients of 2012 Presidential Faculty and Staff Awards.

ACM has recognized CSEE Professor Marie desJardins as a Distinguished Member for her contributions to the field of computing. ACM is the world's largest educational and scientific computing society. Each year it recognizes a handful of its members for significant advances in computing technology that have dramatically influenced progress on a range of human endeavors. This year, Dr. desJardins was one of just 54 computer scientists, educators, and engineers from leading academic and corporate institutions worldwide who were recognized.

Dr. desJardins is well known for her artificial intelligence research, which focuses on planning, learning, and multiagent systems. She leads the large and active MAPLE research group and also works on developing new techniques to improve computer science education.

UMBC was recognized as a “Great College to Work For” in a survey done by the Chronicle of Higher Education. The 2011 survey was based on responses from nearly 44,000 people at 310 US institutions. UMBC was as one of ten large four-year colleges (> 10K students) selected for the honor roll based on its scores on nearly 100 questions across twelve key categories.

The CSEE Department wishes to extend its congratulations to Dr. Marie desJardins for her recent promotion from associate professor to full professor. Dr. desJardins began teaching at UMBC in 2001 as an assistant professor of Computer Science and Electrical Engineering, and since then has taught courses in areas such as artificial intelligence, multi-agent systems and computer programming. As director of the Multi-agent, Planning and Learning lab (MAPLE) at UMBC, Dr. desJardins works with students to find AI solutions to real world problems. In addition to teaching, Dr. desJardins has actively been pursuing research in the areas of multi-agent systems, machine learning, and planning.

Throughout her decade-long career at UMBC, Dr. desJardins has been an active member of several university organizations and committees. From 2008 to 2010, she served as the Undergraduate Program Director for the Computer Science program. Currently, she is a member of the CSEE Executive Committee, the UMBC Faculty Affairs Committee and the CWIT Internal Advisory Board. Dr. desJardin’s numerous contributions to the CSEE Department, whether through teaching, research or committee involvement, have not been overlooked and the Department is confident that she will excel as she takes on her new role this July.

To learn more about Dr. desJardins’ research pursuits, read her research profile.

Dr. Marc Olano is the director of the Computer Science and Electrical Engineering Department's Game Development Track and has been pursuing research in computer graphics and computer hardware for more than twenty years. Currently, he is working at Firaxis Games on texture compression for the Civilization V video game and collaborating with Dr. Erle Ellis of the Geography and Environmental Systems Department on a project dubbed Ecosynth.

To read more about Dr. Olano's research pursuits, see his full research profile. 

Dr. Tim Oates, associate professor of computer science, does research in the field of machine learning and is interested in understanding the development of the human brain. Dr. Oates is also fascinated by the idea of making robots that are capable of learning and exhibiting human characteristics.  “I don’t know if we’ll ever have androids walking among us that are indistinguishable from humans,” says Dr. Oates, “but I bet we’ll get pretty darn close."

To read more about Dr. Oates' research pursuits, see his full research profile.

The UMBC CSEE Department will be publishing a series of short research profiles describing the research activities of its faculty and students. The first features Professor Marie desJardins and the work of her Multi-Agent, Planning and Learning Lab at UMBC, where she works on developing A.I. solutions to real world problems. Dr. desJardins is especially interested in collaborating with students and helping them develop their own research interests. She says that nearly ninety-five percent of her research is with students. “I like the students to learn about a problem and find something that they think is interesting,” she says.