Middleware for Adaptive Multimedia Communication

1
Middleware for Adaptive Multimedia Communication
Objectives key problems
Participants Partners

• Build on the success of voice
• The killer app for mobile communication
• Interesting enhancements
• Hi-fi quality, Video, Augmented/virtual
  realities
• Support applications with simplest possible
  infrastructure
• Robust adaptive source coding for low service
  requirements
• Simple edge-based traffic controls for quality
  of service
• Form a generic middleware between application
  and infrastructure
• Follow the Internet design principle
• Functions placed at low levels of a system may be
  redundant or of little value when compared with
  the cost of providing them at that low level

• KTH/IMIT
• Teletraffic Systems (Karlsson, proj. leader)
• KTH/S3
• Communication Theory (Skoglund)
• Sound and Image Processing (Kleijn)
• Partner projects
• Audiovisual Mobile Internetworking (VINNOVA)
• Winternet Traffic control (SSF)
• Towards a self-regulating Internet (VR and
  SCINT)
• Industry partners
• Ericsson
• Telia

2
Middleware for Adaptive Multimedia Communication

• Robust Adaptive
• Source Coding
• Speech, soundimage and video
• Robustness towardpacket loss and biterrors
• Multiple descriptions
• Seamless rate and quality adaptability
• Cross-layer issues

• Traffic Control
• Self-admission control
• Single-service qualitydifferentiation
• Combination of control functions FEC,
  smoothing, buffering
• Cross-layer issues

• Experiments
• Voice over IP measurement data
• used within the project
• retrieved more than 100 times by outsiders

Robust image transmission to combat bit errors
and lossesleft traditional technique, right new
technique
Gain of multiple-description coding over FEC,
different channel conditions