Skip to content
Permalink
main
Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?
Go to file
 
 
Cannot retrieve contributors at this time

Applications

Learning outcomes

Understand

  • the different parts that make up a typical IoT application
  • ethical issues that may arise
  • key constraints on what IoT can do

Parts of an IoT Application

The mote or node

figures/tmote.jpg

  • Generally, all motes receive the same software

The border router

  • Also known as sink, gateway, or root node
  • Generally consists of a mote connected to the USB port of a PC
  • Can also be a Raspberry PI and mote
  • Converts protocols between mote network (e.g., RPL or 6LowPAN) and ordinary IP network
  • May require special software that is different from all other motes

Message Queue

  • The Message Queue (e.g., MQTT) provides a connector between the source (motes) and sink (back-end database)
  • Publish / subscribe architecture is used
  • Security may be an issue and consider configuring passwords early in the design

Web-server (CoAP)

  • CoAP is the Constrained Application Protocol (RFC7252)
  • It can be thought of as a REST web server approach
  • The CoAP server provides a HTTP style service and sits on a PC
  • Motes make CoAP requests using GET, PUT, POST, DELETE verbs
  • The server responds with responses
  • See IoT in 5 days for more info.

Rule-based systems (Node-RED, IFTTT)

  • Systems like Node-RED can subscribe to messages from MQTT
  • They can then transform and act on those messages
  • Node-RED is good for wiring systems together

Ethical issues

Considering privacy

  • If you are performing an `experiment’, then standard ethical tests apply
    • participants must have informed consent
    • they must be able to opt out at any time
  • Information can often be derived indirectly:
    • Bathroom humidity might indicate when shower is in use
    • Bedroom CO2 sensors might identify sleep patterns or sexual activity
  • Participants should be anonymised

Consider GDPR issues

  • Data protection provides guidance about what data should be stored and processed
  • Reprocessing data for a different purpose than the originally stated purpose is not allowed
  • Further reading is available

Who owns the data?

  • Data ownership is often overlooked
  • Your contract may state that the data is owned by the company who employed you
  • When considering ownership, keep in mind GDPR reprocessing issues

It shouldn’t be up to the individual to decide

  • As a researcher or an employee, neither you nor your superior should decide what is ethical
  • There must always be a higher, independent authority
  • This might be a professional association, such as the IEEE, or a university ethics board

Constraints

Radio limitations

  • Maximal transmission / reception distances are a factor of:
    • transmitter power
    • receiver power / sensitivity
    • radio frequency
    • building materials for any structures
    • interference
  • Overall, expect typical IoT devices to have a range of around 20m indoors and about 100m outdoors

Transmission power

  • IoT devices tend to operate with much less power than laptop WiFi and thus will have a smaller range
  • Transmission power can be adjusted to help reduce power consumption

Radio frequency

  • As with all electromagnetic waves, higher frequencies are blocked by solid objects whereas lower frequencies tend to pass through
  • WiFi, BlueTooth, ZigBee and other IEEE 802.15.4 use the 2.4GHz ISM band

Building materials and shape

  • Electrically conductive materials (water or metal) tend to absorb RF better than non-conductive materials (wood or air)
  • In some cases, structures may help to extend the range slightly (e.g., in a long corridor)

Interference

  • Many systems use the same ISM bands and interference is common
  • Channel hopping approaches may help to bypass interference - particularly when operating near heavy machinery

Network lifetime

  • How long a network lasts for without changing the batteries is an important factor for many applications
  • Power consumption is an important consideration and has been discussed in another lecture.
  • Energy harvesting may allow indefinite extension of the network life
  • Specialist batteries can improve lifetime also
  • Software approaches are another avenue for improving lifetime

Battery technology

  • Specialist batteries may help extend the life of a system
  • Main consideration for the battery is volume / weight and mAh rating
  • For energy harvesting consider lead-acid batteries (although these can be dangerous in some situations)
  • Rechargeable batteries will tend to have shorter lifetimes per charge cycle but last longer overall

Energy harvesting

  • Solar panels are inexpensive and reliable but you must also have a battery that you can charge to continue to operate overnight
  • Wind is less reliable
  • Supercapacitors can be used to smooth out small fluctuations

Memory

  • Most motes have very limited memory and thus will restrict
    • code size
    • stored data
  • It may be possible to store more data in non-volatile flash memory
    • e.g., Telos mote has 1Mb flash

Time accuracy

  • Motes generally have inaccurate clocks
  • Protocols such as TSCH (Time Synchronized Channel Hopping) may help
  • Assume that your mote clock is several seconds out