Telecom%20Research%20at%20Rutgers%20University

1
Telecom Research at Rutgers University

• Fred S. Roberts
• Director, DIMACS
• froberts_at_dimacs.rutgers.edu

Commission on Jobs Growth and Economic Development
1
2
Outline

• What is Telecom?
• The research players and their background
• A selection of research topics suggested by
  Rutgers faculty
• Short synopsis
• Food for thought aimed at stimulating discussion

Commission on Jobs Growth and Economic Development
2
3
What is Telecom?
The set of technologies and sciences at the
intersection of communications, information, and
computing
Commission on Jobs Growth and Economic Development
3
4
What is Telecom?

• 20th century transport, switching, and storage
  of narrowband voice and data
• 21st century? Reasonable goal fully integrated
  and networked broadband multimedia including
• data of all types
• text, images, audio, video
• virtual reality
• searchable, browseable multimedia documents
• shared reality tele-collaboration

Commission on Jobs Growth and Economic Development
4
5
What is Telecom?

• So telecom is all about networks
• interconnections of networks (e.g., the Internet)
• operation and maintenance of networks
• things that make up networks (routers, hubs,
  switches)
• things that get moved around networks (data,
  text, voice, images, video, )
• things that attach to networks (devices, sensors,
  monitors)
• services that run on networks

Commission on Jobs Growth and Economic Development
5
6
Outline

• What is Telecom?
• The research players and their background
• A selection of research topics suggested by
  Rutgers faculty
• Short synopsis
• Food for thought aimed at stimulating discussion

Commission on Jobs Growth and Economic Development
6
7
Major Hubs of Telecom Research at Rutgers

• CAIP (Center for Advanced Information Processing)
• DIMACS (Center for Discrete Mathematics and
  Theoretical Computer Science)
• WINLAB (Wireless Information Laboratory)
• Computer Science Dept.
• Statistics Dept.

Commission on Jobs Growth and Economic Development
7
8
CAIP

• Established by NJCST in 1985
• Mission industrial applications of advanced
  computing technologies
• Industry partners
• ATT, Avaya, Cisco, Datacube, Fujitsu, General
  Motors, IBM, InfoValue, Intel, Iscan, Kodak,
  Lucent, NEC, NIST, Oracle, OSS Nokalva,
  Panasonic, Sarnoff, Siemens, SpeechWorks, SUN,
  Telcordia, Texas Inst., CECOM, Picatinny Arsenal,
  Verizon, Xybernaut
• University partners
• UMDNJ, NJIT, Princeton, Stanford, MIT, Johns
  Hopkins, CMU, Colorado, Cal Tech, Columbia, New
  Mexico State
• 86 faculty, visiting scientists, staff, and
  students

Commission on Jobs Growth and Economic Development
8
9
Telecom-related Research at CAIP

• Multimodal interfaces (NSF)
• Image and speech pattern recognition (DOD)
• VLSI design (NJCST)
• Bio/nano mechatronics (NSF)
• Applications to homeland security (DOD, CECOM)
• SiC semiconductors (DARPA, Union Carbide)
• Collaborative networking (DOD, NSF)
• Distributed grid computing (NSF)
• Data visualization (NSF)
• Telemedicine/rehabilitation (NSF, Novartis)
• Virtual environments (NSF)
• Speech production (NIH)

Commission on Jobs Growth and Economic Development
9
10
Telecom-related Research at CAIP Future
Opportunities

• Natural communication with information systems.
• Virtual environments for collaboration
• Internet delivery of rehabilitative therapies
• Autonomic grid computing
• Systems and sensors on a chip
• Detection of radioactive materials
• Human imaging for dosimetry analysis
• Low bit-rate communication for security

Commission on Jobs Growth and Economic Development
10
11
CAIP Cumulative Impacts

• External Contract Funding 55M
• (17M in current contracts)
• Ph.D.s and MSs graduated 213
• Patents filed 80
• Startup companies assisted 20
• CAIP spinoff companies created 3
• Small business outreach, new jobs 100

Commission on Jobs Growth and Economic Development
11
12
DIMACS

• The importance of discrete math and theoretical
  CS (algorithm development) led Rutgers,
  Princeton, ATT Bell Labs, and Bellcore to
  develop strong research groups
• In 1988, they joined to form DIMACS
• Telecommunications ATT Labs, Bell Labs,
  Telcordia, Avaya
• Computing NEC Research, IBM Research, Microsoft
  Research, HP Labs (Princeton)
• 1989 prestigious NSF science and technology
  center award. 10M grant largest at Rutgers.
  NJCST played important role.

Commission on Jobs Growth and Economic Development
12
13
Telecom-related Research at DIMACS

• next generation networks technologies
• computational information theory and coding
• communication security
• simulations of communication architectures
• computer-aided verification of software
• massively parallel computing
• massive data sets
• applications of large scale discrete optimization
  to communication networks
• cryptography
• complexity of interactive computing

Commission on Jobs Growth and Economic Development
13
14
Telecom-related Research at DIMACS - II

• telecom researchers find new applications of
  their methods
• homeland security research
• epidemiology/public health
• computational biology
• DNA computing

Commission on Jobs Growth and Economic Development
14
15
Telecom Research at DIMACS

• More than 50M in external funding for research
  and education program at DIMACS since its
  inception
• NSF, ONR, NSA, NIH, DARPA, ICMIC (intelligence
  community), Sloan Foundation, Burroughs Wellcome
  Fund, NJCST, numerous companies.
• Solution of Gilbert-Pollak conjecture led to
  highly efficient heuristics for design of
  communication networks.
• Pioneer in field of computer-aided verification
  methods now used widely by Intel, Sun, Motorola,
  ATT, Lucent.
• Simulation software for the global internet
  adopted by more than 40 companies/universities.

Commission on Jobs Growth and Economic Development
15
16
Telecom Research at DIMACS

• Work on error-correcting codes led to new
  techniques for the design of efficicient encoders
  and decoders.
• A remarkably simple on-line algorithm for bin
  packing small information packets of varying
  sizes into bins of fixed capacity.
• Powerful cryptographic methods for secure
  authorized access.
• The players at DIMACS
• 230 scientists from partner universities and
  companies
• partner company scientists directly involved in
  DIMACS projects
• more than 1000 visitors a year

Commission on Jobs Growth and Economic Development
16
17
WINLAB

• Founded in 1989
• Broad experimental and theoretical expertise in
  wireless technologies
• Broad collaborative experience with industry
• about 20 industry sponsors
• major partners brought into NJ include Intel,
  Nortel, Thomson, Samsung, NTT, Sprint, Motorola,
  Mitsubishi,
• Implementing technology transfer through both
  sponsor companies and startups

Commission on Jobs Growth and Economic Development
17
18
Telecom Research at WINLAB

• Freebits -- short range ultra high speed
  communications
• NJ Center for Wireless Communication
• 4th Generation Radio Resource Management
• Adaptive Networking for 3rd Generation Cellular
• Security in Next Generation Wireless
• Dynamic Spectrum Management
• First Generation of MUSE sensor program
• Research Wireless Testbed
• Cognitive Network Management

Commission on Jobs Growth and Economic Development
18
19
Telecom Research at WINLAB

• Pioneer in hot spot wireless networking
  technology now appearing at Starbucks, McDonalds,
  etc. through its infostations program. Going
  out to startups and Army STTR tech transfer.
• 20 faculty/staff 40-50 students
• Currently over 2M a year in funding.
• This year won IEEE Marconi and William R. Bennett
  Awards

Commission on Jobs Growth and Economic Development
19
20
WINLAB Implications for the Future

• Wireless is fastest growing segment of telecom.
• Almost 500M cell phones sold/year.
• 1/3 of calls in US are already wireless, not
  wired (FCC).
• 148M US subscribers, half the population. (FCC).
• 76B Wireless revenues in 2002 30 of telecom
  (FCC).
• Wireless Data devices market expected to be 10B
  by end of 2003.
• 21M American users of Wireless Hot Spots by 2007
  (IBM).
• 6,300 global hotspots in 2001 expect 114,000 by
  2006. (IBM).
• NJ needs a world-class center of expertise in all
  major areas of wireless communications.

Commission on Jobs Growth and Economic Development
21
Economic Impact of Wireless Research in NJ

• 100s of new high tech jobs/year via startups and
  partnerships.
• Retain high tech talent in NJ.
• Train and retain the best students.
• Diversify the telecom industrial base in NJ
  through diverse wireless end-user applications,
  not just the traditional (and now stagnant) core
  infrastructure.

22
Outline

• What is Telecom?
• The research players and their background
• A selection of research topics suggested by
  Rutgers faculty
• Short synopsis
• Food for thought aimed at stimulating discussion

Commission on Jobs Growth and Economic Development
22
23
Multimodal Integrated Wireless Sensor on-Silicon
(MUSE)(WinLAB, ECE, BME, CS, UMDNJ, GaTech)

• Today, sensors are individual units as
  transistors once were.
• Temperature, pressure, light, chemicals, etc.
• Expensive controllers, readouts, and
  communications
• Usually physically large and often hand-made.
• With new technology, we should be able to link
  sensors in complex networks to gather information
  in new ways.

Commission on Jobs Growth and Economic Development
23
24
Multimodal Integrated Wireless Sensor on-Silicon
(MUSE)

• Networks of Sensors
• Applications to medicine, consumer, environment,
  security, military, etc.
• Need new wireless networking technology
• Need ultra-low cost sensors and controllers
• New sensor technology that can measure many
  properties
• Ultra low power electronics, algorithms, and
  protocols
• All on one chip, reusing as much of integrated
  circuit technology as possible

Commission on Jobs Growth and Economic Development
24
25
Multimodal Integrated Wireless Sensor on-Silicon
(MUSE)

• Low Cost, Wireless Networked Sensors
• Strongly multidisciplinary program
• Draws in all levels of technology from devices to
  networks to applications and security.
• Build on strengths of the partners and ongoing
  programs.
• Too large to tackle without cohesive program with
  a shared vision and strong core funding.

Commission on Jobs Growth and Economic Development
25
26
Multimodal Integrated Wireless Sensor on-Silicon
(MUSE)

• Economic impact for NJ
• Market for integrated sensors estimated at 3B in
  2005 and 10B in 2010
• This is before security adders or changes in
  military needs.
• Can build on existing industrial partnerships and
  experience to make the technology transfer
  happen.

Commission on Jobs Growth and Economic Development
26
27
Locating Mobile Users

• Estimating the location of wireless
  communications users attracts huge attention
• Applications include
• location-aware services
• finding the nearest vending machine or printer
• finding the nearest buyer or seller in a market
  of buyers and sellers
• in a museum setting, presenting artifact-specific
  descriptions on a handheld device
• locating a misplaced handheld device

Commission on Jobs Growth and Economic Development
27
28
Locating Mobile Users

• Other applications
• emergency location
• identifying the room location of a crime victim
• in a prison setting, locating a distressed guard
• access control
• blocking access to a Wi-Fi network from outside a
  building
• blocking access for specific users from specific
  locations

Commission on Jobs Growth and Economic Development
28
29
Locating Mobile Users

• Team from Rutgers (Statistics Dept.) and Avaya
  Inc. (wireless expertise) have developed novel
  and highly sophisticated statistical algorithms
  unlike any of the existing approaches
• Substantially more accurate location estimation
  with dramatically less training data
• Immediate application in enterprise settings

Commission on Jobs Growth and Economic Development
29
30
Locating Mobile Users

• Every major telecommunications company working on
  this problem
• Tremendous commercial potential
• Avaya team has significant experience with
  wireless technology and markets
• Rutgers has a long track record of funding and
  innovation in statistical methods
• Urgent need for seed funding for experimentation
  and software development

Commission on Jobs Growth and Economic Development
30
31
Massive Data Analysis Lab (MassDAL)

• Agenda Gather, manage and process massive data
  logs----Web, IP/wireless traffic data, location
  trajectories of objects, sensor readings of
  physical world.

Commission on Jobs Growth and Economic Development
32
Massive Data Analysis Lab (MassDAL)

• Key Challenges
• Scale Beyond the traditional human scale. Eg.,
  IP data at a single router interface for an hour
  exceeds total yearly worldwide credit card
  transactions!
• Data Collection probes/sensors with associated
  data quality and communication problems.
• Need breakthroughs in Mathematics, Algorithms,
  Systems and Engineering, to meet these
  challenges.
• Potential Major impact in Telecom,
  Transportation and Society-at-large.

Commission on Jobs Growth and Economic Development
33
State of MassDAL

• Engineering
• Consulting in analysis of wireless network logs.
  Client ATT Wireless, 3rd largest in US, 20
  Million customers. Terabytes/month. Current
  value 3M per year. 5 pers. Fully operational,
  telco-grade! Interest from Cingular wireless.
• Incorporated novel algorithms in operational IP
  network data analysis tools. Current partner
  ATT. Potential partner Lucent.
• Mathematics and Computer Science.
• Algorithms, Databases, Statistics, and Data
  Mining on novel models and algorithms.
• Supported by NSF grants. Partners Rutgers CS,
  DIMACS, MIT.

Commission on Jobs Growth and Economic Development
34
State of MassDAL (Contd)

• Science
• Developing wearable sensors for tracking location
  of objects as well as interactions between
  objects.
• Current partner Telcordia. Their initial
  investment 300K/3 months (est). Potential
  partner in works Los Alamos National Lab.
• Potential Analysis of social networks for
  Epidemiology and Homeland Security.

Commission on Jobs Growth and Economic Development
35
Future of MassDAL

• Research Need breakthrough research in
  mathematics, systems, databases, algorithms,
  sensor networking.
• Expand data domains.
• Potential partners Google, NJ auto insurance
  fraud data, USPTO patent data, AWS location
  trajectories, etc.
• Build state-of-art facility at Rutgers.
• Secure, 24x7, data hosting and analysis
  infrastructure capable of gathering and
  processing petabytes of data/month across
  domains, data sources, etc. Unique in the world!
• Potential.
• Every wireless, telecom, internet service
  provider is looking to farm out this crucial
  piece of their operations. Estimated market for
  these services 100s of millions in US per
  year. Crucial for NJ State. Interest from
  multiple VCs now.

Commission on Jobs Growth and Economic Development
36
Visualizing, Monitoring, and Analyzing Network
Data(Statistics Dept. and Avaya Labs)

• Communication networks are widespread.
• Typical data provides a partial view of flow-data
  (e.g., on links)
• Analyzing network data is important in
• network planning and design
• monitoring flaws
• measuring reliability parameters
• determining suitability of the network for
  different transmission functions (voice, data,
  voice over IP)

Commission on Jobs Growth and Economic Development
37
Visualizing, Monitoring, and Analyzing Network
Data

• Challenges
• Network data is complex and of high
  dimensionality.
• Statistical methods for analyzing network data
  are few and far between.
• Visualizing data helps us to spot trends quickly.
• Need is to develop high quality, practical,
  statistical and data analytic tools for
  understanding data from partial views and limited
  measurements.

Commission on Jobs Growth and Economic Development
38
Visualizing, Monitoring, and Analyzing Network
Data

• Results potentially useful for other kinds of
  networks transportation, social,
• Such tools of great importance in telecom.
• Research in this area already funded through NSF
  and NSA
• New methods/products should be very useful to NJ
  companies.

Commission on Jobs Growth and Economic Development
39
Self-Healing Dependable Computing

• Computer, Heal Thyself
• Scientific American, July 2003
• We need systems that
• monitor themselves
• adjust hardware and software configurations to
  match demand
• predict and diagnose problems and effect repairs
• defend against hacker attacks

Commission on Jobs Growth and Economic Development
40
Self-Healing Dependable ComputingKey Concerns

• Susceptibility to attack
• Do denial-of-service attacks and viruses cause
  problem?
• Performability
• Is system available with adequate performance
  when needed?
• Dependability
• Can you rely on correct and predictable behavior?
• Self-awareness and autonomy
• Does your system monitor and repair itself?
• Fail-safe uses
• Would you trust your computer with a
  mission-critical task?

Commission on Jobs Growth and Economic Development
41
Self-Healing Dependable Computing
State of the Art

• Information Technology is predicated on
  well-behaved, interacting machines
• but spam, viruses, and attacks are epidemic

Commission on Jobs Growth and Economic Development
42
Solutions
Self-Healing Dependable Computing

• Be realistic about computing environments
• Errors, both human and computer, will always be
  present
• Machines are only as well-behaved as their owners
• Viruses, spam, and attacks ARE part of the
  environment

• Design systems that are self-aware and
  self-healing
• Hardware is fast enough and affordable
• Establish self-administered distributed policies
• Continuously monitor, diagnose, and adapt

Commission on Jobs Growth and Economic Development
43
Self-Healing Dependable Computing Economic Impact

• On the global IT sector
• System downtime will become increasingly costly
• Without self-healing systems salaries will
  dominate IT costs
• On New Jersey build on strengths
• Two of the six NJ growth clusters are related to
  IT
• NJ is center of telecom industry
• NJ has the largest number of scientists/engineers
  per capita
• Experienced workforce is available for new
  initiatives

Commission on Jobs Growth and Economic Development
44
Self-Healing Dependable Computing Rutgers
Expertise

• Active research in several related areas
• autonomous agents, change analysis for OO
  languages, component-based scalable networks,
  database mining, distributed systems, fault
  tolerance, peer-to-peer computing, secure
  services, modeling and simulation
• 7 CS faculty are currently working on relevant
  research
• 3.5M in external grants awarded over last few
  years
• Active industrial collaboration
• Panasonic (Peer-to-peer computing)
• IBM (change analysis for OO languages)
• Telcordia and Rutgers CS are developing a joint
  initiative in this area

Commission on Jobs Growth and Economic Development
45
Multimodal Human-Machine Interface
46
Multimodal Human-Machine InterfaceReal-world
trial with NJ National Guard
47
User interface for interaction and collaboration
with robots and humans
NSF Equipment Grant EIA98-18313 Center for
Advanced Information Processing, Rutgers
University, Piscataway, NJ 08854. PI J.L.
Flanagan, co-PIs J. Wilder, I. Marsic, M. Krane
48
Portable Interactive Command Console (PICC)
Internet
HQ/VEHICLE
Flatpanel
display
Sensors
Robotic Vehicles Emergency Responder in
the Field
49
Pervasive and Autonomous Computing WinLAB, ECE,
CS
Commission on Jobs Growth and Economic Development
49
50
Pervasive and Autonomous Computing WinLAB, ECE,
CS

• Communication and computing cost and performance
  have been improving by 2x every 18 months or less
  for decades.
• Wireless now makes it possible to complete the
  last link to people, machines, sensors, etc.,
  everywhere.
• Great opportunity (and challenge) to move from
  point-to-point communication to pervasive
  communication, computing and knowledge access.

Commission on Jobs Growth and Economic Development
50
51
Pervasive and Autonomous Computing

• Computing and communication can be integrated in
  the environment
• Knowledge, information, communication always
  available, but less obtrusive
• Your personal Radar ORiley is there to help,
  wherever you are (and gone when you want privacy)
• Sensors bring realtime data that matters
• From your heartbeat to traffic jams and afternoon
  weather
• The worlds knowledge is always available when
  needed.

Commission on Jobs Growth and Economic Development
52
Pervasive and Autonomous Computing

• Massive (and interesting) Research Challenges
• Flexible system integration
• New approaches to networking at all levels
• Information-centric parallel and grid computing
• Energy efficiency at all levels
• Context awareness for communication and
  applications
• Location awareness in routing and computing
• Effective and user friendly security at all
  levels
• Integrating of Computing and Communication
  (especially wireless) is already a major
  corporate thrust at Intel, Microsoft, IBM, and
  many others.

Commission on Jobs Growth and Economic Development
53
Pervasive and Autonomous Computing

• Rutgers has expertise and ongoing programs in
  these areas.
• Communications and computing affect every aspect
  of the economy and every individual
• Recent events show the limitations of the
  existing models for Telecom. NJ could take the
  lead in changing the landscape.
• New Jersey has the right combination of
  people, expertise, facilities to make it happen.

Commission on Jobs Growth and Economic Development
54

• Trusted Computing/Authentication
• Rutgers Camden (CS)
• Security is the fastest growing sector of the
  telecommunications market today
• Security involves encryption, authentication,
  access control, identity management, user
  provisioning,
• Telecom often involves access to remote
  resources, requiring authentication of users and
  monitoring of users access privileges

Commission on Jobs Growth and Economic Development
54
55
Trusted Computing/Authentication

• Some Research Themes
• Authentication of remote users is usually done by
  passwords.
• Traditional (alphanumeric) passwords are not
  user-friendly and lead to security problems and
  increased IT costs.
• Graphical passwords user-friendly provide an
  extremely large password space (similar to a
  cryptographic key space) and thus are inherently
  more secure.
• Human-factors analysis of new password schemes

Commission on Jobs Growth and Economic Development
56
Trusted Computing/Authentication

• Impact of Research
• Passwords are the most common method for
  authentication, but also one of the most
  vulnerable to cyber as well as physical attack.
• Improved authentication will impact
  human-computer interface, security.
• Will allow users to directly use passwords as
  cryptographic keys
• Collaborations Drexel, Brooklyn Poly., Minnesota
• Collaborations Unisys
• Password research is of great interest to
  software and telecommunications industries.

Commission on Jobs Growth and Economic Development
57
Trusted Computing/Authentication

• Other Research Challenges
• Location-aware authentication/provisioning
• Dynamically changing access control and inference
  management
• Biometrics

58
Telecom and Homeland Security

• Communication security
• wireless security
• sharing data
• information privacy
• identity theft
• secure e-commerce
• Emergency Communication
• Sensor Networks for Bio/Chemical Hazard
  Monitoring

Commission on Jobs Growth and Economic Development
58
59
Telecom and Homeland Security

• Rutgers projects in communication security
  include
• tunable, programmable, adaptive filters for
  secure communication (Engineering School)
• low bit-rate coding of speech signals for secure
  communications (CAIP) (with Sarnoff)
• information privacy (DIMACS) (with HP Labs NJ,
  Telcordia, ATT Labs, NEC Labs)
• secure e-commerce (CS with Fogbreak Software)
• These projects are funded by NSF, DARPA, NJCST

Commission on Jobs Growth and Economic Development
59
60
Telecom and Homeland Security

• Rutgers projects in emergency communication
  include
• infostations for rapid wireless communication for
  first responders (WINLAB) (with Mayflower Radio)
• rapid networking at emergency locations (DIMACS
  with Telcordia)
• rapid telecollaboration (CAIP)

These projects are funded by DARPA
Commission on Jobs Growth and Economic Development
60
61
Telecom and Homeland Security

• Rutgers project in sensor networks with
  application to bio/chemical hazard monitoring
• WINLAB
• partnered with Agere, Sarnoff, Semandex, Thomson,
  JJ, Lucent

Commission on Jobs Growth and Economic Development
61
62
Telecom and Homeland Security

• Methods used in telecom are coming to be useful
  in homeland security research.
• Provides a great business opportunity for NJs
  telecom industry.
• Already, NJ telecom companies are subcontractors
  to Rutgers federal grants in this area.
• Examples are
• surveillance/detection methods
• bioterrorism sensor location

Commission on Jobs Growth and Economic Development
62
63
Telecom and Homeland Security

• Surveillance/detection
• Massive data set methods used in fraud detection,
  network intrusion detection, etc. are being used
  in bioterrorist attack detection, emerging
  disease identification.
• DIMACS, 3M from NSF, ONR, Sloan Foundation,
  Burroughs Wellcome Fund
• cooperating with ATT, Lucent, Telcordia, Merck,
  state and local health departments, CDC

anthrax
Commission on Jobs Growth and Economic Development
63
64
Telecom and Homeland Security

• Surveillance/detection
• MDS methods also used in monitoring streams of
  text messages for new events
• DIMACS, 1M from ICMIC (intelligence community)
• cooperating with ATT, Telcordia

Commission on Jobs Growth and Economic Development
64
65
Telecom and Homeland Security

• Bioterrorism sensor location

BASIS bioterroism sensor system
Commission on Jobs Growth and Economic Development
65
66
Telecom and Homeland Security

• Bioterrorism sensor location
• Network design methods useful.
• Equipment placing algorithms developed for
  broadband access at Telcordia are candidates for
  modification for sensor placement problems.
• Algorithms developed at Telcordia for placing
  regenerating equipment in transparent optical
  networks are also relevant.
• Work at DIMACS with partners from ATT Labs,
  Telcordia, Industrial Engineering, Environmental
  and Occupational Health and Safety Institute
  (joint with UMDNJ), Statistics, CS, and RUTCOR

Commission on Jobs Growth and Economic Development
66
67
Telecom and Homeland Security

• Thus, homeland security research can put NJ
  telecom back to work.

Commission on Jobs Growth and Economic Development
67