Most university campuses employ security systems to help keep students, faculty, and staff safe. But because of their sheer size – the University of Illinois campus totals 6,370 acres, for example – it’s expensive to provide adequate coverage using traditional systems.

Illinois and ADSC researchers teamed up with startup Inspirit IoT to develop a promising low-cost solution: an audio-based security system that can alert a central command center when trouble breaks out. The system, which measures about 10 centimeters in length and can be built for around $20, recently won first place at the IEEE/ACM Design Automation Conference (DAC) 2017 International Hardware Design Contest this summer.

Winning DAC-IoT system “We’re taking a big step forward in terms of the quality of audio event analysis on a very inexpensive platform,” said ADSC researcher Deming Chen, a professor in the Coordinated Science Laboratory and of Illinois’ Electrical and Computer Engineering Department. “We’re no longer going to be limited to providing security systems for only the most high-value locations where it’s worth spending a lot of money. If we can do it cheap, we can put microphones all over the place for a variety of applications.

This system includes four microphones and a field-programmable gate array (FPGA), a type of integrated circuit, which has been configured to detect and classify audio sounds. If the system recognizes a suspicious sound, such as gunfire or screaming, it could alert a human operator at a central command station and direct police to a general location using beamforming.

The technology would employ “smart” sensors that only report suspicious activity, which both protects privacy and preserves computational power, says ECE Lecturer Zuofu Cheng, who is also a part-time employee of Inspirit IoT and helped develop the winning design.

“Our device is designed in real-time to classify events as opposed to recording audio and streaming it somewhere else, which means no one is spying on everything people say,” Cheng said. “The system might detect certain key words and trigger notification for those key words, which means we’re processing a lot less data but more targeted data.”

Similar devices that detect gunshots are already deployed in large cities. But the group’s design also identifies human screaming, which is a much difficult sound to detect, says Di He, a PhD student in Chen’s lab.

“When detecting these trigger sounds is paired with event direction recognition, it resulted in a unique system that serves more practical use,” says He, one of four students who contributed to the work. “We also published a research paper in ‘interspeech’ introducing an audio feature to improve screaming detection performance at low computational cost.”

Deming Chen The technology could work as an audio-only system or it could be combined with a network of security cameras, which are effective in providing even more comprehensive coverage with reduced false alarms.

The startup company, Inspirit IoT, was co-founded by Chen and two research scientists Kyle Rupnow and Swathi Gurumani of the Advanced Digital Sciences Center, an Illinois research center in Singapore. The company is designing an IoT (internet of things) platform that leverages machine learning to help simplify deployment of computer vision and smart sound applications in automotive, sports and entertainment, consumer, robotics and machine vision, medical, and security/surveillance domains.

“This concept behind this audio system can be used for many IoT applications,” said Kyle Rupnow, chief technology officer of Inspirit IoT. “We’re working to provide end-to-end solutions in different market segments.”

CSL Director Klara Nahrstedt believes the system could play a significant role in broader smart community initiatives at the University.

“Sensing devices are critical as we work to build smarter, safer communities,” said Nahrstedt, the Ralph M. and Catherine V. Fisher Professor of Computer Science. “This product provides a low-cost, commercially viable solution to enable safety for smart communities in diverse environments while protecting privacy and promoting the greater good.”

ADSC researchers received the best paper award at the 2017 International Conference on Information Processing in Sensor Networks (IPSN) for their work on securing the power grid using natural timestamps in the power grid’s frequency fluctuations sensed from the powerline electromagnetic radiation (EMR).

ADSC team at 2017 IPSN conference “We are quite concerned with the clocks in a power grid system because if the clocks aren’t right, the operator can’t record the data from the system correctly, which could have a tragic impact on the system,” said Rui Tan, an ADSC adjunct senior research scientist and assistant professor of computer science and engineering at Nanyang Technological University. “We’re looking at how a cyber attacker could target the clocks and maliciously change them.”

According to Tan, many systems still use legacy protocols regarding their clocks that were developed 30 to 40 years ago, leaving them especially susceptible to attacks.

The researchers designed an algorithm that evaluates the time accuracy of natural EMR timestamps on two different IoT platforms: a high-end single board computer with a customized EMR antenna and a low-end mote with a normal conductor wire acting as an EMR antenna. Both platforms can achieve lower sub-second accuracy.

“Our technique is unique because we use properties of the physical system to help the computing system,” Tan said. “Researchers usually focus on the other way, using computers to help control the physical system.”

As energy runs through a system, the power grid frequency is continuously fluctuating. The fluctuation is random, but identical across a large geographic area, and gives researchers a fingerprint of the grid.

“If we can capture these fluctuations, we can compare one device’s fluctuations to the fluctuation of a data trace collected somewhere else,” Tan said. “This helps us know the time the nodes captured the fluctuations.” They call the time inferred from the fluctuations as powerline forensic time.

There are many existing approaches that help keep the clocks synchronized, including GPS, FM radio or Wi-Fi beacons, but these methods each have intrinsic problems, such as being power hungry, too expensive, susceptible to network delays or outages, or an inability to consistently achieve a common notion of time.

“Our advantage is that we don’t suffer from the delays in transmitting the packets used for time synchronization and we can see global time,” said ADSC postdoctoral researcher Yang Li. “Our approach is very easy to use; we even don’t have to rely on a network, it’s accurate and it’s affordable.”

During their conference presentation, Yang demonstrated their technique using sensors they have installed at their homes, at NTU, and at ADSC. The researchers stream the sensors’ fingerprints to a server at ADSC and assess the accuracy of the fingerprints in representing time.“We have an interface to display the clock synchronization accuracy and at the conference, we showed this in real time,” Yang said.

Since their presentation, they have installed additional nodes at several locations across Singapore, including commercial buildings, campuses, and apartments, with up to 24 kilometers between the nodes. Preliminary results validate that their approach could be deployed city-wide.

The researchers were encouraged by the response they got at IPSN, which is a top conference in sensor networks, and are planning to develop a product based on their technique that could be commercialized.

“In addition to commercializing, we’re looking to make hardware improvements, as it’s a bit clunky right now,” said ADSC senior software engineer Sreejaya Viswanathan. “We’re hoping it could be the size of a credit card and easily plugged in. We’re developing a new demo, as well.”

Prior to this IPSN paper, the research group also published a paper in the 2016 IEEE Real-Time Systems Symposium (RTSS) on obtaining the natural timestamps directly from wall power sockets. These natural timestamps are more accurate, achieving sub-millisecond accuracy.

The research was funded under the Energy Innovation Research Programme (EIRP, Award No. NRF2014EWTEIRP002-026), administrated by the Energy Market Authority (EMA). The EIRP is a competitive grant call initiative driven by the Energy Innovation Programme, and funded by the National Research Foundation (NRF).

ADSC researchers have developed a tool, Cyber Security Argument Graph Evaluation (CyberSAGE), that allows security analysts to make practical and informed decisions about the security of their critical infrastructure system, such as a smart power grid. The tool allows users to map their system’s workflow and model failure scenarios to see how their system holds up against potential attacks.

For example, on a power grid, there are a variety of potential cyber failures, such as a loss of automatic control or wireless network complications, but system operators don’t always know how the failures will affect their specific system until it happens. By modeling failure scenarios, such as the NESCOR scenarios from the Electric Power Research Institute (EPRI) in the United States, and linking them with the concrete system models and attacker models, CyberSAGE gives the practitioners a holistic view of how a cyber-attack-induced failure can happen and where are the key places to put more security controls.

“We need to have a way to determine whether a particular system design is more secure or resilient than another design,” said William Sanders, head of the University of Illinois’ electrical and computer engineering department, ADSC faculty member and project co-PI. “There are several other cyber security assessment tools that provide overall assessments, but what CyberSAGE does is take input from a variety of sources, such as NESCOR and other models, to provide a more complete assessment.”

The project, which is funded through A*STAR, is led by University of Illinois and ADSC faculty members, including Sanders, Zbigniew Kalbarczyk and David Nicol, as well as ADSC’s David Yau. More recently, the framework and software developed under that project are also being applied in the SecUTS project, funded by Singapore’s National Research Foundation, which takes a cyber-physical approach to securing urban transportation systems.

The team began creating the framework for CyberSAGE in 2013, and there have been three major rounds of design iterations for the software since the beginning. The main developers for the latest version of the software include ADSC Senior Software Engineer Prageeth Gunathilaka, who is the architect of the latest version of CyberSAGE, ADSC Software Engineer Li Yuan, and ADSC’s former Senior Software Engineer Sumeet Jauhar. The research team is currently working on a fourth that further enhances the software’s usability and scalability, and they hope to add a real-time component in the future to allow the assessment result to change as system conditions evolve. The tool was developed for use on the power grid, but has since been applied to study metro systems. The team has also received interest from companies working with maritime and offshore systems and manufacturing and automation systems as well.

CyberSAGE is free for academic use and has six academic users from the U.S. and Europe. Recently, two companies are also testing the software on a trial basis, and the researchers recently signed a commercial licensing agreement with a multinational electronics corporation for use of the software.

“The key to the development of any of these tools is to have test cases from real practitioners,” Sanders said. “By working together with these companies, we can not only help them make their products more resilient and secure, but to make our tool better as well.”

The current tool combines a lot of information from different silos, including knowledge from security consultants who excel at knowing the threats and attacker’s behavior, along with system administrators or IT operators who know the IT systems and how devices connect, and specific domain knowledge regarding traditional operational scenarios or workflows in that specific field.“Combining this information was a challenge because we wanted to provide a better way to look at the security of a whole system, but often information in these different domains isn’t shared between the silos,” said ADSC Senior Research Scientist Binbin Chen, who is the local lead of the CyberSAGE team in Singapore. “We had to spend time finding out how to do that in an intuitive and scalable manner.”

After users map their system into the offline tool, they can run a few dozens of failure scenarios to see how their system holds up to different attacks. As security results are difficult to assign a value to, the researchers encourage users to interpret the results in a comparative manner, rather than an absolute manner. While quantifying security is a controversial academic topic, the team advises users to walk in the middle, by not reading too much into the values, but rather comparing the scores, probability or improvement values and making informed decisions on those evaluations.

“There are a lot of papers that propose a risk assessment process. It is easy to sit and write about a hypothetical process, but it’s harder to get other researchers or practitioners to care about the process, use it and understand it,” said ADSC Senior Research Engineer William Temple, who leads an effort to apply the CyberSAGE tool in the security assessment of urban transportation systems. “Building a tool that implements our process was the key to making it easy for industry people to have interest in it and want to use it.”

William H Sanders, an award-winning educator and researcher who heads nationally ranked engineering programs at the University of Illinois at Urbana-Champaign, has been named interim director of the Discovery Partners Institute (DPI), a new world-class innovation center led by the U of I System, President Tim Killeen announced today.

Sanders, who also has directed two campus-wide research centers during his more than 20 years in Urbana-Champaign, began his new role Aug. 16 in a designate capacity, pending approval by the Board of Trustees at its Sept. 27 meeting. Killeen said Sanders will be in charge of all aspects of launching and operating DPI, a Chicago-based institute that will be the cornerstone of the Illinois Innovation Network (IIN).

The initiatives were created to accelerate job creation and economic growth through groundbreaking research and innovation. Earlier this year, the state approved $500 million in capital funding to develop DPI, an innovation center that will be home to leading-edge research, and IIN, a virtually connected network of regional hubs that will spread the institute’s impact across the state.

“Bill is an acclaimed educator, innovator and administrator,” Killeen said. “He shares our vision of the power that DPI holds to foster the pioneering innovation that creates new businesses, new jobs and new waves of progress for the people of Illinois and beyond.”

Sanders, also a CSL professor, called the chance to build upon and make real the vision for DPI a “once-in-a-lifetime opportunity.” He said he is committed to fulfilling DPI’s promise of becoming a new model for higher education that combines the strengths of industry, academia and government in a new way that drives both academic discovery and economic growth.

“DPI will harness the unique strengths and scale of the U of I System universities – together with academic and industry partners from Chicago, the state, and the world – to create an unstoppable engine for innovation,” said Sanders, who moves to DPI after serving as head of Urbana’s Department of Electrical and Computer Engineering (ECE).

As interim director, Sanders will lead all facets of planning and operations, including planning to build the new downtown Chicago institute, faculty and staff hiring and engagement, development of research and educational programs, and forging agreements with corporate and academic partners. DPI will operate as an executive office of the president within the U of I System and a hub within the IIN. As such, Sanders will work as a member of the leadership team in system offices and work closely with the chancellors, provosts, and deans of all three system universities.

U of I Vice President for Economic Development and Innovation Edward Seidel will continue to help oversee development of DPI and IIN, Killeen said. Seidel has led the planning team that designed the approach for the new research and innovation enterprise, and brought the projects to the point where a full-time team is needed.

“These initiatives are the product of Ed and his team’s vision and passion,” Killeen said. “Bringing Bill on board will add to the team’s talent, creativity and bring day-to-day attention that is needed now to get DPI up, running and driving progress for our state.”

Seidel added: “I’m looking forward to working with Bill to bring the DPI vision to reality.”

Sanders, 57, has been head of the electrical and computer engineering department in Urbana-Champaign since 2014. ECE ranks third in the nation for graduate computer engineering, fifth in the nation for undergraduate computer engineering, fourth in the nation for graduate electrical engineering, and sixth for undergraduate electrical engineering in the latest ratings by U.S. News & World Report.

He joined the university’s faculty in 1994, and has a long history of academic leadership in Urbana-Champaign. Along with his term as head of ECE, he served as founding director of the Information Trust Institute (ITI) and director of the Coordinated Science Laboratory (CSL).

At ITI, he built the institute from the ground up, growing it to about 100 faculty members from 28 departments, enabling ITI to create interdisciplinary research teams to address important societal-scale problems related to cybersecurity and trust. While at CSL, he and other faculty members created the Advanced Digital Sciences Center, a bricks-and-mortar research laboratory in Singapore that has received more than $90 million in funding from the Singaporean government to achieve its research vision.

While ECE head, the department created two new ECE graduate degrees programs – a professional master of engineering degree that combines graduate-level technical work with professional training and a combined bachelor of science/master of engineering degree program. He also created two new ECE engineering undergraduate degree programs at Zhejiang University’s new international campus in China.

He also led an aggressive faculty hiring program in ECE that has hired 35 new tenure-track faculty, eight teaching faculty, and five research faculty since January 2014. During this time, he more than doubled the number of women faculty members in ECE, from seven to 17, and helped provide millions of dollars of department funds in scholarships to admitted first-year women and underrepresented minority students – resulting in the largest undergraduate enrollment of women ever in ECE.

A Donald Biggar Willet Professor of Engineering, Sanders has won awards for his teaching and research, which focuses on computer system security and dependability. He is a fellow of the American Association for the Advancement of Science, the Institute of Electrical and Electronics Engineers and the Association for Computing Machinery. Sanders holds a PhD in computer science and engineering from the University of Michigan, where he also earned his masters and undergraduate degrees.

Plans for DPI and IIN were announced in October. Work is currently underway on an implementation plan that will establish a timetable for opening and other details of the enterprise, where world-class researchers will work side-by-side with students and businesses to foster next-generation innovation and workforce development.

DPI will be developed on a donated site along the Chicago River, and will bring together top faculty in agriculture, healthcare, computing and other critical fields from the U of I System and partner universities that already include the University of Chicago, Northwestern University, and Tel Aviv University. Nearly 100 new researchers also will be added and together they will connect with hundreds of businesses and thousands of students over time, as well as with entrepreneurs and venture capital firms.

Their research and educational collaborations will address real-world challenges, promoting the kind of breakthrough discoveries that create new products and companies. Those innovations will fuel economic growth, while also providing hands-on experiences for students and nurturing a skilled workforce for the city and state.

The institute will be the centerpiece of IIN, a virtually connected statewide enterprise allowing DPI staff to work with university and business partners in other regions across the state on research and education initiatives that help launch new companies and lift communities.

IIN sites will include the U of I System’s universities in Chicago, Springfield and Urbana-Champaign, and could ultimately grow to other sites in Illinois and beyond.

ECE ILLINOIS Department Head and ADSC researcher William H. Sanders, Donald Biggar Willett Professor of Engineering, testified before the 115th United States Congress on Tuesday, October 3, 2017. He was one of four witnesses to speak as part of a hearing titled “Resiliency: The Electric Grid’s Only Hope” held by the Committee on Science, Space, and Technology.

Sanders, co-PI of the Cyber Resilient Energy Delivery Consortium (CREDC) at Illinois, is one of the authors of “Enhancing the Resilience of the Nation’s Electricity System.” This congressionally mandated report by the National Academies of Science, Engineering, and Medicine prescribes specific actions to improve the reliability of individual grid components and an increased integrated perspective among the U.S. Department of Energy (DOE), U.S. Department of Homeland Security (DHS), state groups, and private utility stakeholders.

The purpose of the hearing was to help define resiliency, discuss recommendations from the report, and highlight the importance of research focused on grid resiliency, infrastructure, and operational strategy. As an expert on secure and dependable computing, Sanders provided unique insight and recommendations concerning the impairments to and approach for providing resiliency in the electric power grid.

“The joint and collaborative involvement of government, industry, and academia in implementing these recommendations [in the report] is key to their success,” Sanders told Congress in his statement.

Representative Darin LaHood (R-IL 18th District) asked Sanders about the other entities involved in this kind of research. Sanders spoke about the many partnerships that have formed to address these issues. He explained that in 2015, the Cyber Resilient Energy Delivery Consortium (CREDC) was formed.

It’s “now ten universities, two national labs…banded together to look at resiliency issues in the grid,” Sanders said. “In fact, my colleague, David M. Nicol, who is the principal investigator of the CREDC effort, is in Texas today talking with people from the oil and gas industry about how we can transition our technologies to the real world.”

Representative Jim Banks (R-IN 3rd District) asked how often there is a cyber-attack or attempt on our national grid. Sanders acknowledged the difficulty of answering the question, explaining that information about attacks and attempts are held in many hands and at varying levels of classification.

“We need to build systems that, rather than protect against very specific cyberattacks, protect against whole classes of effects those cyberattacks may bring on the grid,” said Sanders. “By thinking about the effects, and through resiliency, we can begin to protect against zero-day attacks that we haven’t seen before.”

Sanders closed his testimony by stressing the need for taking precautionary measures.

“Unlike some, I don’t believe the sky is falling or that we are on the brink of a major disaster,” he said. “However, the threat to grid resiliency is real, and the time to act is now, so we don’t reach that brink.”

The three other witnesses for the hearing were Carl Imhoff, manager of Electricity Market Sector, Pacific Northwest National Laboratory, Gavin Dillingham, program director of Clean Energy Policy, Houston Advanced Research Center, and Walt Baum, executive director of Texas Public Power Association.

A full video of the testimony is available. Sanders’ testimony was also highlighted in the Electrical and Computer Engineering Department Heads Association October letter from the editor.

Everyone knows that trips to Disneyland, Six Flags, or Sentosa are filled with laughs, rides, sugary food, and, of course, long lines. In hopes of reducing the frustration of lines, ADSC researchers are partnering with Sentosa, an amusement resort off Singapore’s southern coast, to dramatically decrease the time guests spend waiting for shuttle busses.

The researchers have developed an app that predicts the wait times for each shuttle based on data across the island. Using data from the island, such as WiFi signals and video analysis outputs at bus stops, tram ticket sales, car parking entrance records and weather, the researchers can estimate shuttle bus wait times and the number of waiting passengers.
“The guests will be able to look at the app and see the status of each bus station and the trends in wait times and the number of passengers,” ADSC Senior Research Scientist Zhenjie Zhang said.

The app is based on previous work that examined traffic patterns in Singapore and Shanghai.

In 2014, Zhang and ADSC Research Scientist Tom Fu, partnered with a Singapore startup company to create an app to predict, with 90 percent accuracy, how many people are likely to enter a Singapore metro station within the next hour. The results were based on call detail records from mobile phone users for one month. Using information from those calls, such as which cell towers were nearby, the researchers analyzed the rough trajectories of mobile phone users and the train schedules to create a prediction model.

That project gave the researchers background information on public transportation and helped them compete in the Shanghai Open Data Competition. In this competition, the Shanghai government provided all data from their transportation systems for one month, including the statuses of taxis, metros, buses, how many passengers were riding at a certain time, accident information and weather.

Former ADSC senior research engineer Victor Chen Liang and ADSC postdoctoral researcher Wang Li built the system, Mercury, to digest large amounts of data from these transportations systems and then created a demo app that allows users to get time predictions for various public transportation options.

“If you want to take the metro at 5 p.m., you can put that information in and you’ll get updates if the time prediction changes,” Zhang said. “For taxi availability prediction and bus traveling time predictions, you can know how many available taxis there are in the area in the next hour.”

In the most recent project with Sentosa, the researchers have modified the app to add Sentosa’s attraction spots and make it real-time.

“The scale is very different, as Shanghai is a very big and national city, while Sentosa is a very small island with only a few thousand guests every day,” Zheng said. “The first two projects were more like experiments, so this is the first time we’ve been able to test the system and the models in a real-world application. Data from Sentosa is a little sparse. We are still trying to improve the accuracy of the model because of the sparseness.”

The researchers also imagine connecting the app with local food and beverage services, so if the wait time is long, the app may encourage tourists to walk to their location in exchange for a discount to a local restaurant or simply suggest nearby, less-crowded transportation or restaurant options.

The project is funded by the Government Technology of Singapore (formerly the Infocomm Development Authority of Singapore) and the researchers hope Sentosa will be able to implement their research into their everyday operations soon.

“We believe this is just the beginning,” Zhang said. “Singapore is expected to become a smart nation and traffic is one of the most important aspects of a smart nation, so we foresee more projects coming from other public transportation players.”