BANs pbrandao poster

1
Body Area Sensor Networks Do you feel it?Pedro
BrandãoSup. Prof Jean Bacon Opera Group
Computer Laboratory University of Cambridge

• What are the problems?
• BANs are currently application driven
• A myriad of different applications
• A myriad of different sensors in a BAN
• Lack of an abstraction for apps
• No framework to optimize resources for apps.

• Aim
• Define a middle layer that enables
• Sensor plug and play (self-configuring and
  capability advertisement)
• Applications to discover services with
  specification of requirements
• Optimization of resources to fulfil requirements.
• Myriad of Apps ? Middleware ?Myriad of Sensors

Applications

• Real Time
• Fidelity
• Data Updates
• Determinism

requirements
Middleware

• Qos
• Time Sync
• Measurement
• Nr. Measures

metrics

• Next Objectives
• Communication/architectural Models
• What are req., metrics, resources?
• Cost Function
• Optimization Algo.
• Advertisement/Discovery Protocol

information
control

• Net. Architect.
• Nodes Net interf.
• Nodes processing
• Sensors Measure

Resources

• A tetraplegic patient controls a wheelchair in
  virtual reality, Leeb et al
• Brain Computer Interface, EEG

• Blood pressure (intra-body)

• BANs proj _at_CL
• Caregrid
• Sesame
• MESSAGE

• Given Imaging PillCam
• Oesophagus video capturing

Base Station (BS)

• VivoMetrics LifeShirt
• ECG, breath rate, activity, posture

• Closed Loop diabetes syst.
• Glycaemia reading and insulin injector

• Radiotelemetry in a lab-in-a-pill format,
  Johannessen, et al
• pH and temp. real time measure in
  gastrointestinal tract

• T-Cell count with position info.

• Nike AirID
• Rate of footsteps

Microsoft Research Summer School 2007