|
|
Cyber
Physical Computing
Department of
Computer Science
The
|
|
|||||||
|
|
||||||||
For inquiries,
please see contact info.
|
Cyber Physical Computing Computer Science
has always been an application-oriented discipline. Its greatest advances
that broadly altered the quality of life, such as the Internet, databases, and
supercomputers stemmed primarily from applications brought about by needs of
individuals, businesses, and the government. At present, an expanding
frontier for computer scientists lies at the intersection of the logical and
physical realms. As computing elements become embedded more pervasively in
our environment, a new cyber-physical fabric arises in which logical
processing is very deeply intertwined with the distributed physical
environment in which it occurs. Computing becomes less obtrusive and a more
natural part of the external world. It becomes more autonomous and less
reliant on human input, intervention, and administration. Physical objects
acquire new logical properties due to embedded computation, sensing, and
actuation. New applications arise that improve the quality of life (e.g.,
smart assisted living facilities), enhance social experiences and human
communication (e.g., new cyber-physical communication media), improve
accessibility of information (e.g., wide-area data services), and help advance
fundamental knowledge in many environmental, biological, and physical
disciplines. In this new realm,
computer science must be redefined. New models and paradigms are needed for
computation. New underlying theoretical foundations are needed to support
such paradigms. New programming languages and distributed middleware tools
must be developed around the emerging abstractions of cyber-physical
computation. Networking must be redefined to integrate myriads of physical
data sources, actuators, and computing elements, as well as to develop
appropriate application-layer data services. New operating systems are needed
that are optimized for the new computing realm, as opposed to the current
machine architectures and applications. Data mining and machine learning
techniques are needed to identify data patterns, learn context, and act
autonomously without human assistance.
The Cyber-Physical
Computing Group is a multidisciplinary team that investigates the
aforementioned aspects of tomorrow’s computing systems. Selected Publications L.
Luo, T. Abdelzaher, T. He, and J. A. Stankovic, "EnviroSuite: An
Environmentally Immersive Programming Framework for Sensor Networks,"
Accepted to ACM Transactions on
Embedded Computing Systems (TECS), 2006. William
Hawkins and Tarek Abdelzaher, "Towards Feasible Region Calculus: An
End-to-end Schedulability Analysis of Real-Time Multistage Execution," IEEE Real-time Systems Symposium, Tarek
F. Abdelzaher, Shashi Prabh, Raghu Kiran, "On Real-time Capacity Limits
of Multihop Wireless Sensor Networks," IEEE Real-time Systems Symposium, |
Research Themes A versatile collection of new research
directions are started in this group. On the analytic front, new fundamental
theory and models are sought for distributed computation. Capacity limits are
investigated to quantify fundamental tradeoffs between time, space, and
energy. On the systems side,
challenges are addressed in developing the next generation middleware for
cyber-physical applications. New protocols, compilers, programming languages,
and operating systems are needed. A categorization of research directions is
listed below. |
|
Sensor
Networks: Today,
hundreds of millions of sensors are deployed in our environment. This
includes cameras on myriads of cell-phones, ubiquitous indoor temperature
sensors, garage-door motion sensors, web cams in public spaces, among others.
Hardware miniaturization and wireless communication advances suggest
increased proliferation of sensing devices and their integration into
wireless sensor networks for myriads of urban, military, social, and medical
applications. Zigbee, Bluetooth and other wireless technologies present
possible options for the interconnection of sensors. Research on sensor
networks investigates network protocols, services, resource management, and
programming paradigms tailored to the new environment. |
|
|
Future Internet
Architecture: The
Internet was born as a data communication medium, yet it is being used
increasingly for information retrieval. Such information will typically
originate from embedded sources. Human needs for bandwidth are limited, since
our capacity to source or sink information is bounded (e.g., by limits of
sensory perception). Future growth in network bandwidth demand will
eventually come from embedded sources. Data collected from such sources will
need to be organized for efficient search, retrieval, and proactive event
detection. This suggests a fundamental re-thinking of the Internet
architecture around the concepts of data retrieval, as opposed to
point-to-point or multiparty communication. |
|
|
Smart Attire: Our future wardrobe is the next
likely platform for embedded computing, after PDAs, cell phones, and
multimedia devices such as iPODs. A suite of human-centric software
applications will execute on the new platform. Research is needed into new
operating system abstractions, computing models, languages, and services
(such as support for security and privacy) in Smart Attire applications. This
project develops experimental prototypes of attire fitted with sensors,
microprocessors, and memory, using them as the vehicle to investigate the
aforementioned topics. |
|
|
Environmentally
Immersive Programming: This project develops a new distributed computing paradigm
suitable for cyber-physical computing applications marked by heavy
interactions between the logical realm and the external environment. The
paradigm exports an address space that combines logical objects and
(addressable representations of) external physical objects in the same
“cyber-physical” space. |
|
|
Feasible
Region Calculus: This project develops a fundamentally new theory for analysis of
temporal and spatial properties in open systems based on feasible regions. A calculus
is developed for composing the feasible regions of subsystems to generate
those of the overall application. Results applicable to the understanding of
fundamental capacity limits of new models of computation. |
|
|
Software
Predictability:
The increasing complexity and heterogeneity of future distributed embedded
software systems and the increasing sources of unpredictability that affect
software performance in the face of physical time and space constraints call
for new theory, architectural mechanisms, computing paradigms, and
programming abstractions to support predictable software behavior,
analyzability, and performance guarantees. In this group, we develop new
analysis tools, middleware, high-level interfaces, and underlying analytic
foundations for software predictability. Of particular interest is
predictability of temporal behavior, which is of special importance in
real-time and embedded systems. |