Neal Patwari - US grants

Integrative, Hybrid and Complex Systems
University of Utah
Neal K. Patwari
CAREER: RF-Sensing Networks for Radio Tomographic Environmental Imaging


Intellectual Merit: This research focuses on the development of new technologies to ?see? through walls into buildings to show interior structures and the motion of people within the structure. Rather than relying on a single self-contained short-range radar, this method uses a large-scale network of low-cost sensors as multi-static radio frequency (RF) radars whose pair-wise and spectral measurements can be used to image the environment. This research lies at the intersection of statistical signal processing and radio propagation and addresses the necessary key advances related to dense networks of RF sensors and accurate statistical channel models. The proposed research (1) uses extensive measurements to develop valid statistical channel models that depend on the attenuation field, (2) develops and tests estimation algorithms for radio tomographic imaging, and (3) analyzes their estimation performance.

Broader Impact: If successful in leading to new tomographic environmental imaging systems, the proposal has the potential to significantly benefit fire fighters, other first responders, and building occupants in emergency situations. In addition, the research has the potential to benefit other types of communication networks by advancing cooperative spectrum sensing in dynamic spectrum access radio networks and improving channel simulation in multi-hop networks. The project is a crucial part of the principal investigator?s goal of integrating research and education in signal processing and wireless networks. This project will lead to a new wireless communication system laboratory, a key part of a departmental curricular initiative to provide students with integrative lab experiences. Further the project will develop and disperse new interactive modules to be used with students in grades 10 through 12, in particular in programs targeted towards students from under-represented groups, both part of department goals to increase the diversity and the total enrollment of students in electrical and computer engineering. Undergraduate student research will also be integrated into this project.

The objective of this research is to build new measurement-based methods for secure secret key establishment between two wireless devices, without ever communicating the secret key, using diverse physical characteristics of the wireless medium, notably an innovative measurement called temporal link signatures, and using device mobility.

The intellectual merits of this research include (i) collection of extensive measurements and characterization of spatial and temporal behavior of wireless link signatures in a variety of indoor and outdoor settings, (ii) determination of the shared secret space in different settings through the use of analytical methods and the measurement data, (iii) novel methods for enhancing the length of the shared secrets to make them robust against brute force attacks, (iv) novel methods for dealing with asymmetric measurements of link signatures, and (v) a complete, implementable methodology, with specific configuration recommendations, for secure secret key establishment in real scenarios.

This research will have a significant broad impacts. It generates a large amount of measurement data useful for various future wireless communications and security research; this data is and will continue to be available to the research community through the NSF-funded CRAWDAD repository. The results of this research potentially make mobile applications safer to use in ways that require far less complex management and far less computation than the existing cryptographic methods; these results benefit both security and goals of using mobility in green applications. Finally in broader impacts, this research will be used in conjunction with the PI's existing NSF STEP project for creating learning modules for high school students to demonstrate the problems and limitations of public key cryptosystems and the need for new ideas for secret key establishment.

The objective of this project is to acquire and set up a wireless measurement infrastructure and use it to research new wireless link signatures and their applications. The signature of a wireless link between a wireless transmitter and a receiver represents the wireless link's unique physical characteristics. Wireless link signature applications include secret key establishment between a wireless transmitter and a receiver without ever communicating the secret key, and location distinction which is the ability to detect at one or more receivers when a transmitter changes its location.

The intellectual merit of this research includes (i) extensive measurements of wireless link characteristics under (a) heterogeneous indoor and outdoor settings, (b) a variety of wireless standards with different types of transmitters, and (c) different frequency bands with the help of highly capable spectrum analyzers, (ii) development of novel methodologies for different wireless link signature applications including location distinction, and secret key establishment, using these measurements, and (iii) evaluation of the methodologies through implementation.

This research impacts the development and deployment of wireless link signature-based applications. It also contributes a vast amount of measurement data that is useful for understanding unique wireless link characteristics, and also traditional wireless link performance modeling and evaluation. This research activity is integrated in the education curriculum through new measurement and inference projects in the networking and security classes. Furthermore, this research is used in conjunction with an existing NSF STEP project for high school outreach.

The objective of this research is to enable cyberphysical systems (CPS) to be context-aware of people in the environment and to use data from real-world probabilistic sensors. The approach is (1) to use radio tomography (RT) and RFID to provide awareness (location and potential identification) of every person in a building or area, and (2) to develop new middleware tools to enable context-aware computing systems to use probabilistic data, thus allowing new applications to exploit sometimes unreliable estimates of the environment.The intellectual merit of the proposal is in the development of new algorithms and models for building-scale RT with low radio densities and across multiple frequencies; the development of efficient multichannel access protocols for rapid and adaptive peer-to-peer measurements; the development of space-time and probabilistic data representations for use in stream-based context awareness systems and for merging ID and non-ID data; (4) and the development of a human context-aware software development toolkit that interfaces between probabilistic data and context-aware applications.

The proposal impacts broadly the area of Cyberphysical systems that reason about human presence and rely on noisy and potentially ambiguous (practical) sensors. The research has additional dramatic impact in: (1) smart facilities which automatically enforce safety, privacy, and security procedures, increasing the ability to respond in emergency situations and prevent accidents and sabotage; (2) elder care, to monitor for physical or social decline so that effective intervention can be implemented, extending the period elders can live in their own home, without pervasive video surveillance.

This grant will defray (partial) travel costs of PhD students attending the IEEE Communications Society Conference on Sensor, Mesh, and Ad Hoc Communications and Networks (SECON) 2010. SECON has established itself as one of the premier conferences in the areas of sensor, ad hoc, and mesh networking, serving as a meeting point of researchers from academia and industry, as well as practitioners in diverse fields of wireless networking. SECON 2010 will be held in Boston, Massachusetts, June 21-25, 2010. Participation in such technical forums is critical to a PhD student?s maturity and development. To facilitate such experiences, this proposal requests funding to defray partial travel support to students who might otherwise be unable to attend the conference. Today?s graduate students will play a central role in shaping the future of wireless technology. Attendance and participation in the top conference in one's field is a key part of the socialization of graduate students to the academic and research community, which will broaden their perspectives and have cascading impacts. We outline specific plans to use the travel grant to encourage the broadening of the demographics of students in attendance at SECON.

Temporal fading is the change in the radio channel between a transmitter and receiver, for example, a fluctuating "number of bars" or signal strength between a laptop and access point, even when neither are moving. Past research in temporal fading treats it only as a problem that degrades wireless communication. Emerging research has shown that temporal fading can be exploited to locate, automatically recognize the activity or gesture, and monitor the health of people in the vicinity of a wireless network. These localization, recognition, and monitoring systems are called RF-based environment monitoring (REM) systems. Improvements in REM technologies could aid in the design of police and search-and-rescue systems that locate breathing people in dangerous or collapsed buildings. As another example, REM technologies deployed in a home could detect falls and detect signs of cognitive or physical decline as part of an aging-in-place sensor system. REM technologies could allow people to diagnose disordered sleeping via wireless devices (e.g., cell phones) left on their bedside. Finally, REM systems could revolutionize indoor and outdoor security systems, helping to protect areas and buildings which are difficult to monitor with existing technologies. To date, no fundamental research in temporal fading mechanisms has been performed to support REM applications. Research in this project considers temporal fading and seeks to establish how it is affected by the movements of people in the environment so that it can be exploited for environmental monitoring.

This project will develop, verify, and exploit new models for the temporal changes that occur to received power and channel response. The scope includes both large-scale motion, i.e., a person walking across a room, and small-scale motion, i.e., a person breathing, or moving an arm. It is known that the effects of motion vary as a function of other multipath fading characteristics of the channel and as a function of the person's position with respect to the transmitter and receiver, but it is not known how. This project will explain these relationships with statistical models that describe temporal fading changes due to human and object motion, to breathing and other periodic motion, as a function of the position of the person or object and a function of other multipath channel characteristics. All models will be verified using real-world measurements, using controlled measurements in an anechoic chamber, and using numerical electromagnetic simulation. From them, the project will develop improved estimators, identify improved features for learning methods, and reduce the manual training required to deploy RF-based environment monitoring systems. A wide variety of device-free localization (DFL) and activity recognition algorithms, as well as gesture recognition and breathing monitoring algorithms, can be improved both in terms of accuracy and efficiency. For example, fingerprint-based DFL methods can be enabled to work as well with fewer training measurements, statistical inversion DFL methods can achieve improved accuracy, and RF-based breathing monitoring systems can be made more robust.

A patient taking a prescription opioid pain reliever, while sleeping, can have ones' breathing rate slow to a dangerous level and then stop. The drug overdose problem has become an epidemic with 40,000 U.S. deaths annually, five times more deadly than twenty years ago, and now is the leading cause of accidental death in the U.S. This project proposes to monitor people at risk of prescription drug overdose in their home using the PlusOne monitor, a wireless network of sensors deployed in a home that collect and process radio channel data in order to continuously estimate breathing rate. The sensors are noncontact and always on, and work wherever the person is in his or her home. If a person is "still" and the breathing rate is dangerously low, the system alerts a caretaker by text or call, or automatically calls 911. Although prescription drug overdose deaths happen at a higher rate than motor vehicle deaths in the US, no safety systems have been developed to reduce the overdose death rate. If successful, this commercial system could dramatically reduce the death rate. Beyond the prescription drug safety market, this I-Corps team believes a noncontact breathing monitoring system has application in personal health monitoring, baby monitoring, and search and rescue.

This team hypothesizes that drug rehabilitation centers will value the improved safety that the PlusOne monitor provides and will pay to deploy it in the homes of their patients when they are released. While other products exist to monitor breathing rate, they require the person to be physically connected to, or within a short distance of the device. The PlusOne system is a network that monitors the radio wave propagation channel between each pair of deployed sensors, forming a mesh that fully covers a home. No other product has the capability to be always on and cover an entire home. In this project the team will perform customer segment validation to determine how to successfully bring this safety product to market; (2) further develop a prototype for customer and partner demonstrations; and (3)
perform further validation in human subjects studies. If successful, the team will position the technology for investment, future regulatory approval, and success in the market, and in so doing, be able to provide a safety system for people who are in greatest danger of a prescription opioid overdose.

Asthma is the most common pediatric chronic illness in the US and associated with significant health care burden. Despite advances in treatment, asthma control in children remains generally poor. Poor control is associated with physiological and behavioral factors as well as with environmental exposures. While research efforts have been devoted to studying the impact of physiological and behavioral factors on asthma, research on the effects of environmental exposures in children with asthma is limited. Recent advances in sensor technologies offer new ways to measure environmental exposures and improve asthma research. However, use of sensor data in research remains uncommon and multiple barriers exist. Sensor data are complex, being variable in structure from sensor to sensor, high volume and acquired at a high frequency (e.g. every minute or more), and multi-dimensional, with sensor data linked to geo-spatial location, time, and description of exposures. Further, sensors lack standards for data collection and data exchange of environmental exposures, lack common data elements, have inadequate interfaces for families, and lack of secure communication modalities for data transmission. Moreover, little effort has been made to integrate sensor data with clinical data, making it difficult to study the actual effects of the environment on asthma. The proposed PRISMS ? Informatics Federation Architecture Center provides standard-based, open access architecture and standardized processes for acquisition, integration, and management of sensor data, along with clinical data from longitudinal assessments of asthma symptoms, quality of life, health care usage, and other asthma related outcome metrics. Specifically, we propose to develop three synergistic and tightly integrated projects that make up an innovative sensor monitoring system, including: Project 1: Develop infrastructure and software to facilitate data acquisition and information exchange. We will develop a platform for mobile apps to support data extraction from the child?s home and environment sensors, and data processing and user-friendly interface for data presentation to the participants. Project 2: Develop a high-resolution data integration platform that will provide the common core, allowing sensor data to be integrated with clinical data, and mechanisms for securely transmitting data to the PRISMS data coordinating center. This project will leverage existing OpenFurther architecture, open source frameworks that promote code reusability and interoperability. Project 3: Develop a platform for researchers with flexible user interface to configure a variety of experimental designs. We will use user-centered design to develop both user and researcher interfaces. These projects are tightly integrated, linked by the architecture in project 2. Overall, these projects will standardize data collection from environmental various sensors, data integration with clinical data and presentation to enhance research to improve our knowledge of the impact of environmental exposures on children with asthma.

Software defined radio (SDR) is emerging as a key technology to satisfy rapidly increasing data rate demands on the nation's mobile wireless networks while ensuring coexistence with other spectrum users. When SDRs are in the hands and pockets of average people, it will be easy for a selfish user to alter his device to transmit and receive data on unauthorized spectrum, or ignore priority rules, making the network less reliable for many other users. Further, malware could cause an SDR to exhibit illegal spectrum use without the user's awareness. The FCC has an enforcement bureau which detects interference via complaints and extensive manual investigation. The mechanisms used currently for locating spectrum offenders are time consuming, human-intensive, and expensive. A violator's illegal spectrum use can be too temporary or too mobile to be detected and located using existing processes. This project envisions a future where a crowdsourced and networked fleet of spectrum sensors deployed in homes, community and office buildings, on vehicles, and in cell phones will detect, identify, and locate illegal use of the spectrum across a wide areas and frequency bands. This project will investigate and test new privacy-preserving crowdsourcing methods to detect and locate spectrum offenders. New tools to quickly find offenders will discourage users from illegal SDR activity, and enable recovery from spectrum-offending malware. In short, these tools will ensure the efficient, reliable, and fair use of the spectrum for network operators, government and scientific purposes, and wireless users. New course materials and demonstrations for use in public outreach will be developed on the topics of wireless communications, dynamic spectrum access, data mining, network security, and crowdsourcing.

There are several challenges the project will address in the development of methods and tools to find spectrum offenders. First, the project will enable localization of offenders via crowdsourced spectrum measurements that do not decode the transmitted data and thus preserve users' data and identity privacy. Second, the crowd-sourced sensing strategy will implicitly adapt to the density of traffic and explicitly adapt to focus on suspicious activity. Next, the sensing strategy will stay within an energy budget, and have incentive models to encourage participation, yet have sufficient spatial and temporal coverage to provide high statistical confidence in detecting illegal activity. Finally, the developed methods will be evaluated using both simulation and extensive experiments, to quantify performance and provide a rich public data set for other researchers.

?DESCRIPTION (provided by applicant): Deterioration of vision is inevitable to all humans. By the age of 45, the biological lens in our eyes starts to lose its elasticity thus producing refractive vision errors, and the eye cannot clearly focus the images. The result is blurred vision, which is sometimes so severe that it causes visual impairment. Three very common refractive errors are myopia (nearsightedness), hyperopia (farsightedness) and presbyopia (difficulty in reading at arm's length). In particular, presbyopia is an inevitable, universal age-related condition where the accommodative ability of the eye is lost permanently. Accommodation refers to the ability of the eye to increase its refractive power of the crystalline lens in order to focus near objects on the retina. In comparison to normal range of 7-10 diopters, it typically decreases to ~0.50 diopters by the age of 50. In 2005 over one billion people worldwide were estimated to suffer from presbyopia alone. Despite its ubiquity, the exact mechanism behind presbyopia remains unknown. Such refractive errors cannot be prevented, but they are treated with corrective glasses, contact lenses or refractive surgery. Eyeglasses are the most common inexpensive tools for correction of refractive vision errors. However, conventional corrective eyeglasses have a number of drawbacks; most importantly inability to fully restore the vision accommodation range of a normal eye. The principal goal of this work is the development of integrated sensing, actuating, control and data collection Smart Eyeglasses for adaptive correction of blurred vision caused by refractive errors. This cyber-physical opto-electro-mechanical system uses a combination of large-aperture fluidic lenses, ultralight actuators (aim 1) object distance and eye direction sensors (aim 2), and embedded control, communications and computing electronics to continuously produce full-field sharply focused images at any object range (aim 3). Furthermore, these glasses can record the observer's behavior, which can be utilized to predict prescription drift corrections and better eye medical treatment for patients and improvements in the observer environment. The testing of the smart eyeglasses will be performed with an instrumented phantom simulating visually impaired human subjects.

Traumatic brain injury is a leading cause of death and disability in the United States. Over 1.7
million people sustain a brain injury each year and make up one-third of all injuries seen in the
emergency room. Developing rehabilitation and treatment strategies to manage this disease are
important, but preventing the occurrence of brain trauma is also critical component to the solution.
The goal of this proposal is to reduce the risk of traumatic brain injury through smart technology
that collects sensory data to predict and characterize head impact in real-time, optimizes protective
mechanisms based on those impact characteristics, and sends impact trauma attributes to a clinical
database for further analysis and injury risk prediction. This technology will substantially improve
traumatic brain injury prevention and diagnosis in motor vehicle crashes, sports, and industrial
accidents. To accomplish this goal, fundamental research efforts include (1) real-time situational
monitoring to predict when and how dangerous impacts are about to occur and (2) active
prevention mechanisms to reduce the risk of brain injury impact. Initial evaluation of the
technology is in a sports setting, but the system components can be widely adaptive for
implementation in motor vehicles, industrial safety helmets, and living environments for the
elderly.
The research goals of this proposal are to (1) reduce the risk of traumatic brain injury through
advanced situational monitoring, musculoskeletal activation, and impact-specific force reduction;
and (2) to improve potential identification of head injury risk based on multiscale brain
deformation modeling. These goals are accomplished by integrating four fundamental research
efforts. First, tracking and collision detection algorithms are developed based on radio frequency
(RF) sensing, processing, and flexible antenna design. When used in conjunction with triaxial
accelerometers, gyroscopes, and magnetometers, these algorithms provide the sensing capabilities
required to detect objects, capture directional velocity data of surrounding objects, and process
data in real-time to determine probabilities and characteristics of impending collision. Second,
musculoskeletal clenching following auditory warning is investigated as a means of minimizing
head angular acceleration following head or body impact. The development of auditory warning
cues and muscle clench strategies utilizes kinematic musculoskeletal modeling and human subject
studies to identify required auditory cues and response times as well as muscle activation
parameters that best mitigate head angular acceleration during a collision. Third, active force
reduction specific to impending impact characteristics are implemented using a unique controllable
air-filled bladder. Optimal pressure and deflection characteristics of the bladder are based on
impact velocity and direction, and evaluated with a novel three-dimensional multiscale finite
element model of the human head. This model incorporates anatomical variability in the
microstructures at the brain-skull interface, a region that is critical to predictions of head injury.
The fourth fundamental research area uses the multiscale model to investigate the relationship of
head impact force and acceleration to regional deformation of brain tissue upon impact. These
studies will be used to improve predictions of TBI risk from impact kinematics.

The POWDER-RDZ project investigates ways to share the electromagnetic (radio-frequency) spectrum between experimental or test systems and existing spectrum users, and between multiple experimental systems. This research team will deploy and evaluate a prototype automatic spectrum sharing management system for the POWDER testbed in Salt Lake City, Utah (part of the NSF’s “Platforms for Advanced Wireless Research” program). Spectrum access challenges currently create significant constraints on experimentation and testing at wireless testbeds. Automatic spectrum sharing for safe access to additional frequencies – beyond the frequencies reserved exclusively for testing – will relax these constraints and thus increase the nation’s capacity to conduct wireless research and development. Increasing this capacity will help accelerate growth and global leadership of the US communications industry, strengthen academic research into wireless systems, and benefit other spectrum-dependent sectors such as radar, public safety, and national defense. As a pathfinder for the National Radio Dynamic Zone concept, the project will help future federal/non-federal spectrum sharing arrangements assure that spectrum sharing does not negatively impact government users. <br/>The POWDER-RDZ team will design, develop, and prototype an end-to-end radio dynamic zone (RDZ) for advanced wireless communication. They will validate its functionality by performing spectrum sharing experiments and field studies on the resulting artifact. The project uses the existing POWDER mobile and wireless testbed as the physical infrastructure of the RDZ. POWDER’s existing radios and other equipment supports the spectrum sharing experiments and provides part of the RF sensing functionality needed by the RDZ. The project will design and develop a modular zone management system (ZMS) to manage, control and monitor all aspects of the RDZ. The project plans to conduct experiments on spectrum sharing with users outside of POWDER. Experiments potentially include RDZ shared access to federal, non-federal, and commercial spectrum, such as coarse- and fine-grained spectrum sharing with a commercial mobile operator and spectrum sharing with a weather radar.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Spectrum in the US is incredibly valuable to science, education, commerce, transportation, and contemporary life. Wireless bandwidth needs are rapidly growing, but allocation of spectrum to wireless users should not come at the expense of scientific observation in astronomy and earth science which are already under-allocated and subject to interference when operating opportunistically outside the narrow protected bands. This project is developing new secure and accountable sharing protocols that not only enable more efficient sharing of the spectrum between terrestrial commercial wireless systems and passive receivers, but also empowers passive systems to force a particular interfering transmitter to switch band. Improved coexistence allows more reliable operation of radio astronomy receivers. This project is creating open source software and data, and working with the ITU-R to ensure that results impact the research community and future spectrum sharing policy. Developments are impacting undergraduate and graduate education through course material and research experiences, and the project is engaged in K12 outreach.<br/><br/>This project is developing RF watermarks that embed random pseudonyms into transmitted wireless communication signals so that passive receivers can demodulate the pseudonym of any interferer. The proposed system then allows passive receivers to indirectly inform the interfering device to change band. To prevent any other device from inferring private information, the proposed system leverages differential privacy to quantitatively limit privacy leakage. Further, the project is adapting software attestation to develop proof of correct execution of spectrum decision on user equipment, and complementing the watermark-based detection system with spectrum policy enforcement. Protocols under study are being implemented and tested on PAWR testbeds as an open source project. Extensive experimentation, including at the Owens Valley Radio Observatory, is validating the technical contributions, and quantifying its performance and robustness to attacks.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.