Lecture Slides
Lecture5-1-Data Processing-SAX.pptx
Key Characteristics of IoT Devices
- We need to find a way to optimize our bandwidth because inexpensive sensors or actuators are usually equipped with only a radio transceiver for various applications. These typically have low data rates (10-250 kbps), with CCTV and video/sound data requiring higher bandwidth data rates.
Beyond Conventional Sensors
- Human sensors (tweeting real-world data and/or events)
- Software sensors (software agents/services generating/representing data)<
Data Processing in IoT
- IoT data processing requires solutions that are real-time, reliable, efficient (for low-power and resource-limited nodes), and scalable.
- Use cases:
- Convert billions of smart meter readings to improve power consumption prediction and balancing
- Monitor patients, provide in-home care, and support various healthcare applications
- Analyze data from thousands of sensors—including traffic, pollution, weather, congestion, public transport, and events—to enhance traffic management and smart city operations
IoT Data Access
- Data requests for sensory data are usually multi-dimensional, including:
- Location
- Type
- Time
- One of the above plus a range or a combination of quality of information attributes
- Complex data type
- For example, a temperature value of 15 degrees becomes more meaningful when described with spatial (e.g., Guildford town centre), temporal (e.g., 09:15 AM GMT, 09-11-2020), and unit (e.g., Celsius) attributes.
In-network processing
- Mobile Ad-hoc Networks vs. Wireless Sensor Networks (WSNs):
- Mobile Ad-hoc Networks: These networks focus on transmitting data from one point to another, much like a traditional communication network.
- WSNs: These networks are more sophisticated, providing processed information rather than just raw data. This means they can interpret and analyze data before sending it.
- Key Advantage of WSNs:
- In-Network Processing: WSNs can manipulate and process data within the network itself, which is a significant advantage over traditional networks.