Internet of Things

Question 1

IoT & Open Data

Answer 1

Much Overlap

• Open Data datasets often come from the IoT
• Examples:
  
  • US Weather Data
  • London Traffic Flow Data

Question 2

Describe IoT

Answer 2

The network of physical items embedded with electronics, software, sensors, actuators, and network connectivity which enable these objects to connect and exchange data.

Aka.

• Pervasive computing
• Ubiquitous computing
• Machine to Machine Connectivity
• Fourth Industrial Revolution

Question 3

Industrial Revolutions

Answer 3

• First – 18/19C –iron, steam
• Second - 1870 to 1914 –steel, oil, electricity
• Third – 1980s onwards –PC, Internet, ICT
• Fourth – now: robotics, AI, nanotech, biotech, IoT, 3D printing, autonomous vehicles

Question 4

Drivers of IoT

Answer 4

• cheap sensors
• more powerful processing
• increased networking abilities
• convergence of technologies enable applications
• social acceptance high (many early adopters)

Question 5

5 components of an IoT product

Answer 5

• Sensors, actuators (or other things that can be connected)
• Connectivity (to Internet)
• Platform to process the data (typically cloud-based)
• Data Analytics/BI
• User Interface

Question 6

Supervisory control and data acquisition (SCADA)

Answer 6

A control system architecture that uses computers, networked data communications and graphical user interfaces for high-level process supervisory management, but uses other peripheral devices such as programmable logic controllersand discrete PID controllers to interface to the process plant or machinery. The operator interfaces which enable monitoring and the issuing of process commands, such as controller set point changes, are handled through the SCADA supervisory computer system. However, the real-time control logic or controller calculations are performed by networked modules which connect to the field sensors and actuators.

The SCADA concept was developed as a universal means of remote access to a variety of local control modules, which could be from different manufacturers allowing access through standard automation protocols. In practice, large SCADA systems have grown to become very similar to distributed control systems in function, but using multiple means of interfacing with the plant. They can control large-scale processes that can include multiple sites, and work over large distances. It is one of the most commonly-used types of industrial control systems, however there are concerns about SCADA systems being vulnerable to cyberwarfare/cyberterrorism attacks.

A proportional–integral–derivative controller (PID controller or three term controller) is a control loop feedback mechanism widely used in industrial control systems and a variety of other applications requiring continuously modulated control. A PID controller continuously calculates an error value {\displaystylee(t)} as the difference between a desired setpoint (SP) and a measured process variable (PV) and applies a correction based on proportional, integral, and derivative terms (denoted P, I, and D respectively) which give the controller its name.

In practical terms it automatically applies accurate and responsive correction to a control function. An everyday example is the cruise control on a road vehicle; where external influences such as gradients would cause speed changes, and the driver has the ability to alter the desired set speed. The PID algorithm restores the actual speed to the desired speed in the optimum way, without delay or overshoot, by controlling the power output of the vehicle’s engine.

Question 7

Zigbee

Answer 7

This is what the IoT, or pervasive computing, began with

• a low cost, low power mesh network standard
• wireless monitoring and control applications
• chips integrated with radios, data storage
• works only at short ranges

But IoT needs is a high power, low energy solution

Question 8

The Emerging Wireless Connectivity Ecosystem

Answer 8

• Long-range high-data rate systems: 3G and 4G
• Short-range high-rate: Wifi;
• Short-range low-rate: Zigbee, Bluetooth.
• IoT needs loww-rate long-range: 5G

Question 9

Low power Wifi

Answer 9

• Sensors sleep when not working – saves power
• Typical network configurations:
  
  • 6,000 sensors attached to a single access point
  • Communicating at 100kbps
• Applications:
  
  • Smartmetering
  • Industrialprocess management and control
  • Healthand social care
• Could work for IoT if power consumption reduced enough

Question 10

Low Power Wide Area Networks

Answer 10

• Large coverage – spread spectrum techniques
  
  • Typical urban ranges of 5km
  • Difficulturban ranges of 1-2 km
• Low cost:
  
  • Radiochip solution
  • Subscription fees are low
  • $1 per chip and $1 per annum per subscription

Question 11

Cellular–3G, 4G, 5G

Answer 11

• Biggest potential
• Wide coverage
• Low deployment costs
• Good security
• Dedicated spectrum
• Simple management
• Standardised (by 3GPP)

Question 12

3 key factors of connectivity

Answer 12

Reliability

• Can the technology meet the performance requirements
• Resilience to interference, data delivery guarantees, low system outages, etc.

Availability

• How well can the technology be applied elsewhere
• Guarantee of coverage, ability to support mobility and roaming, critical mass in rollout, etc.

Viability (i.e. cost)

• Is there a business case
• Can the business be viable in the near future.
• Cost pertains to the Total Cost of Ownership which covers acquisition costs (the smallest cost), maintenance costs and running/management costs

Question 13

ideal IoT data plattform characteristics

Answer 13

Processes IoT Data

• Native raw data support
  
  • both data ingestion and processing
  • Hadoop, MapRData
• Support for a variety of workload types
  
  • deal with low-latency queries against semi-structured data items, at scale
  • support stream processing
• Business Continuity
  
  • IoT applications usually come with SLAs in terms of availability, latency and disaster recovery metrics (Recovery Point Objective/Recovery Time Objective)​
• Security and Privacy
  
  • ensure a secure end-to-end operation
  • authentication and authorization systems
  • user privacy must be warranted

Question 14

Importance of Data Analytics in IoT (+ Machine/Deep Learning)

Answer 14

• Machine Learning can produce predictive analytics
  
  • ​supermarket demand prediction
  • supply chain optimisation
• Deep Learning can uncover patterns previously unknown to humans
  
  • may reveal why certain products fly off the shelves and others don’t

IoT and Data Analytics

• detect anomalies in real time
• trigger an alert of something going wrong or about to happen
  
  • a product ran out an alert
• long-term trends
  
  • ​allows us to construct certain policies
    
    • some products are only popular during winter

Question 15

IoT & Privacy

Answer 15

People now leave a huge digital trace

• preferences
• location via mobile phone & social media
• CCTV cameras
• Microlocationing–where you are inside store

Cons

• anxiousness
• privacy invasion

Pros

• offer people services in better ways
• to reduce crime

Assurance of privacy and fair use of data.

• Privacy Impact Assessment

Question 16

The 7 Privacy Principles

Answer 16

1. Think ahead — Proactive not Reactive

• Clear commitment at the highest levels
• Employ methods to recognise poor privacy design
• Anticipate poor privacy practices and outcomes before they affect your business

2. Privacy as the default setting

• Personal data automatically protected
• No action required

3. Privacy by Design

• Embedded in the design of IT systems and business processes
• Delivered without diminishing functionality

4. Full functionality – positive sum

• It is possible to have privacy AND achieve other business benefits

5. Full life-cycle, end-to-end security

• Privacy and security must be embedded from start to finish
• Securely retained
• Securely destroyed

6. Visibility and transparency

• Assure all stakeholders that you are operating to stated promises and objectives
• Offer independent verification
• Trust but verify

7. Respect for user privacy

• Keep it user-centric
• Strong privacy defaults
• Appropriate notice
• User-friendly options

Question 17

Security vulnerabilities of IoT

Answer 17

• Hardware usually sound; software often not so
• Vulnerabilities –SQL injection, Web APIs
• Operating system patching

SQL injection –some web applications have text boxes that let users login to thewebsite. Behind the scenes the website is querying the back end database using thelanguage called SQL to check if the user’s login details are valid. It’s often possible to type certain characters in the login box that let a malicious user inject his or her own code into the SQL query. That can do things like delete the records from the database or reveal other users’ password hashes. This is made possible by the fact that SQL,like other languages, has its own syntax. For example, quotation marks are used todenote strings. If the user is allowed to type the quotation marks in the form where the input is astring, they can break out the string and onto the actual code of the query itself.

Inthe context of embedded systems like Internet of Things devices, we often find that the same kind of vulnerabilities exist. Vendors tend to make great hardware, but putvery little effort into the software.For example, routers often have an internal web application that the user can access to configure the router and those applicationsare very often vulnerable to SQL injection, and with some ingenuity, can be exploitedremotely.

An Internet of Things device also often provide Web APIs to allow other devices tocommunicate with it and send it commands. We often find that these APIs are alsovulnerable to the same kind of flaws. We also find, particularly with embedded devices,is that the operating system that runs on the device, which is typically Linux, isinsecurely configured. For example, user programs run as admin or there’s nopassword to access the remote admin terminal.

Question 18

IoT Security Measures

Answer 18

Use a Trusted Platform Module (TPM) for authentication. A TPM is a dedicated microprocessor that integrates cryptographic keys into devices to uniquely identify and authenticate them. Each device then has its own identifier that is encrypted by the keys. This will prevent hackers from hacking and impersonating a device to gain access to home, enterprise, or government networks.

Use the Trusted Network Connect (TNC) standards to check for malicious software or firmware. The TNC standards offer a way to check devices for malicious software or firmware whenever they try to access networks or other devices. This would help prevent hackers from using hacked devices to upload spyware or other malicious software to networks or other devices.

Isolate and remediate (fix) infected devices with security software and protocols. If a device is infected with malware or other malicious programs, it needs to be quarantined. The IF-PEP protocol can isolate an infected machine from other devices and networks. There are numerous solutions from security software vendors for clearing the device of the infection once its isolated.

Layered security can limit the damage a hacker can do once device is hacked. A Mandatory Access Control system limits access to certain functions or files on a device for a given user. This acts as a choke point that can prevent hackers from gaining sensitive information through the hacked device. This is like Paul Jennings’ defencein depth

Data encryption is a must. This should go without saying, but data needs to be encrypted when stored on a device or in transit. The post recommended using a read-only mechanism to obstruct hackers’ efforts to tamper with data on a device.

Secure legacy systems through industrial control systems. To reach their full potential, IoTdevices and systems have to be integrated with legacy machines or appliances that were never built to be connected or secured against hacking. Industrial Control Systems can segment that legacy hardware from other systems and secure communications between them with encryption. This, for instance, could prevent a hacker who has infiltrated the network of a connected factory from then taking control of the machinery on the assembly line.

Question 19

User interface challenges

Answer 19

• Prepare devices for changes in functionality
  
  • e.g. tesla cars software updates
  • design the physical interface to take account changing functionality
• Enable remote control –change settings remotely
  
  • Need to design devices to reflect remote changes
  • Interusability - components of a systems understanding each other
  • Latency - there cannot be a delay in certain situations (e.g. turning on the light with your phone)