quartz/content/notes/06-challenges-for-ubicomp-and-intro-to-sensors.md
2023-03-16 09:27:27 +13:00

2.8 KiB
Raw Blame History

title tags
06-challenges-for-ubicomp-and-intro-to-sensors
lecture
info305

Challenges

  • Whats difficult about Ubiquitous Computing?
    • Requires understanding of sensors and new sensing technologies
      • Noisy inputs
      • Sensor fusion
  • Wireless communication channels
  • Power consumption
  • Context is only a proxy for human intent
    • Unpredictable ussage context in particular for mobile devices
  • Require novel user interfaces, limited/different I/O capabilities
    • Lack of standardization in interface patterns
  • Privacy & Security

Wireless Communication Channels

  • https://www.ntia.doc.gov/files/ntia/publications/january_2016_spectrum_wall_chart.pdf
  • Unpredictable availability and throughout
  • Many physical issues:
  • Refraction
  • Absorption
  • Antenna Geometry
  • Power Limits
  • Wireless spectrum is highly contested
  • Many sources of interference
  • Limited bandwith available
  • Often complext SW/HW stacks

Power Consumption

  • Biggest energy consumers
    • Display (backlight)
    • Wireless modules
    • GPS receiver, cameras
    • Sensors (touch, accelerometer, IMU)
  • Limited power supply
    • Tradeoff: capacity vs. size/weight/portability
    • Energy consumption becomes important
  • Two solutions:
    • Increase energy in relation to size/weight
      • Requires chemistry & physics knowledge
    • Decrease energy consumption
      • Requires CS/IS/EE knowledge
  • graph

Privacy and Security

  • Context awareness requires huge amounts of private & personal data (see current generation mobile devices)
    • Emails, contacts
    • Sensor data, movement profiles
    • PIN/TAN codes
    • Photos
  • Many people want to access data
    • Google, Facebook, Microsoft (> ads)
    • NSA, GCHQ, BND, Five Eyes (> crime prevention)
    • Hackers (> extorsion, stealing your money)
  • Problem 1: No pervasive encryption
  • Problem 2: Voluntary use of cloud services, interconnected services

Sensors

need to capture context information to build context aware systems

smartphone sensors

  • Smartphones house highly integrated sensors for sensing the environment:
  • Proximity Sensors (e.g. NFC, Cameras)
  • Location sensors (e.g. Cell,. AGPS, WiFi, GSM, Cameras)
  • Orientation Sensors (e.g. Magnetometer, Accelerometer, Gyroscope, Camera)
  • 3D sensors (LiDAR, Cameras)
  • Microphone
  • Lux-meter
  • Future smartphones likely to add more:
  • Kinect (e.g., Primesense Capri, Motorola Tango)
  • RTK GPS
  • Health sensors (glucose sensor, heart rate monitor,..)

head mounted displays

  • Head-mounted displays (e.g. Hololens, Snap Spectacles) house highly integrated sensors for sensing the environment:
  • Location sensors (e.g. 2 x cameras and depth camera)
  • Orientation Sensors (e.g. Magnetometer, Accelerometer, Gyroscope, Camera)
  • Microphone
  • Lux-meter
  • Eye tracking cameras

Location