Pervasive Data Access (PDA) Research Group

1
Pervasive Data Access (PDA) Research Group

• Mei Li
• Pennsylvania State University

2
Overview

• Investigate various research issues involving
  data access and management in a wide spectrum of
  pervasive/mobile computing and networking
  environments
• peer-to-peer networks
• mobile ad-hoc networks
• wireless sensor networks
• Also investigate object-oriented databases, XML,
  security, and information integration and
  retrieval

3
Peer-to-Peer (P2P) Systems

• Peer-to-Peer computing paradigm has changed the
  ways large-scale distributed systems are built
  and operated.
• Goal of our research
• design data management systems to support various
  applications ranging from distributed research
  locating to network attack detection
• We Investigate
• Information management infrastructure
• Complex queries processing
• Knowledge discovery (clustering, outlier
  detection)
• We address various system issues
• overlay maintenance, load balancing, resilient to
  dynamic changes, adaptivity to query locality

4
Location-aware Wireless Sensor Networks (WSNs)

• WSNs have a wide range of applications
• object tracking, environmental monitoring,
  warehouse inventory, and health care.
• Goal of our research
• build data management systems in support of data
  collection, aggregation, dissemination,
  in-network query processing and query
  optimization.
• We investigate
• location tracking of moving objects
• location-based routing
• complex query processing (window, KNN, and
  aggregation queries)
• Spatio-temporal modeling and estimation of
  wireless link quality in sensor networks

5
Mobile and Pervasive Data Access

• Important for a wide range of mobile businesses
  and applications.
• We are one of the first few pioneers who started
  research in this area
• Indexing
• Caching
• Accessing
• Data aggregation
• Data dissemination
• Channel allocation
• Query processing
• Our work has extensive coverage
• from simple data to spatial data
• from single cell to multiple cells
• from stationery clients to moving clients

6
Location-Based Services (LBSs)

• LBS is one of the killer applications for mobile
  computing and wireless data services.
• We investigate
• indexing and caching spatial data to support the
  processing of location based spatial queries
  (LBSQs)
• point query, window query, k nearest neighbor
  (KNN) search, continuous nearest neighbor (CNN)
  search, spatial join, and other complex spatial
  queries
• We address various issues
• large index search space, large index size,
  linear streaming property of wireless data
  broadcast, continuous movement and requests of
  mobile users, and cache replacement/invalidation

7
Contact Information

• Dr. Wang-Chien Lee
• E-Mail wlee_at_cse.psu.edu
• http//www.cse.psu.edu/wlee/pda/

• Mei Li
• E-Mail meli_at_cse.psu.edu
• http//www.cse.psu.edu/meli