Lesson 1: Introduction to Sensors and Wireless Sensor Networks

How physical events become useful digital information

A wireless sensor network (WSN) connects the physical world to computing systems. Sensors observe temperature, light, movement, pressure, sound, humidity, location, or other conditions. Small embedded devices convert these measurements into digital data, process them, and communicate them wirelessly. A gateway or base station then delivers the information to applications that store, analyse, display, or act on it.

This first lesson establishes the foundation for the whole course: what a sensor is, how common sensors work, what a mote or sensor node contains, why multiple nodes form a WSN, what makes WSN design difficult, and where these networks are used.

Core idea: A WSN is an end-to-end measurement and decision system—not merely a collection of wireless devices.

Learning outcomes

After completing this lesson, you should be able to:

1. Define sensing, sensor, transducer, actuator, mote, sensor node, and WSN.
2. Explain the complete data-acquisition chain from a physical phenomenon to digital information.
3. Describe how accelerometers, capacitive humidity sensors, thermistors, thermocouples, photodiodes, and LDRs work.
4. Classify sensors using the measured property, energy source, output, contact, and operating principle.
5. Distinguish a conceptual sensor node from a physical mote platform.
6. Explain single-hop and multi-hop WSN communication.
7. Describe why energy, self-management, wireless links, decentralization, limited resources, and security are major WSN challenges.
8. Compare WSNs with traditional networks and mobile ad-hoc networks.
9. Explain ubiquitous and calm computing in relation to WSNs.
10. Identify major WSN applications and propose a suitable sensing system for a real problem.

1. Course direction

The Wireless Sensor Networks course develops theoretical and practical knowledge of sensor technology, embedded measurement, wireless networking, and IoT integration. Its major themes include:

• WSN architectures;
• wireless communication principles;
• medium access, routing, and energy-efficient communication;
• IoT fundamentals;
• signal processing and data collection;
• analogue and digital sensors;
• calibration and electrical measurement;
• sensor-network applications; and
• evaluation of performance, energy, robustness, sustainability, security, and societal impact.

The expected practical abilities include selecting sensors, configuring network nodes, integrating sensor data, measuring system behaviour, and planning, documenting, and presenting a group project. The most important engineering attitude is to justify each technical choice instead of selecting a sensor or radio only because it is popular.

2. Sensing, sensors, transducers, and actuators

2.1 Sensing

Sensing is the process of gathering information about a physical object, process, system, or area. A useful measurement must relate to the real quantity of interest—not only produce a changing number.

Human biology provides familiar examples:

• eyes capture optical information;
• ears capture acoustic information;
• the nose responds to chemical/olfactory information; and
• skin captures touch, pressure, texture, and temperature information.

Electronic sensing follows the same broad pattern: a receptor responds to physical energy and produces a signal that can be interpreted.

2.2 Sensor and transducer

A sensor detects a physical, chemical, or biological quantity and produces a usable signal. A transducer converts energy from one form to another. In WSN discussions, a sensor is commonly described as an input transducer because it converts physical-world energy into an electrical signal.

Examples:

• a thermistor converts temperature change into resistance change;
• a photodiode converts incident light into electrical current;
• a microphone converts sound pressure into an electrical signal;
• a piezoelectric element converts force or vibration into charge/voltage.

2.3 Actuator

An actuator performs the reverse system role: it receives electrical control and changes the physical world. Examples include a motor, relay, valve, heater, fan, buzzer, LED, and servo.

A closed-loop system can therefore be written as:

Physical process → Sensor → Controller → Actuator → Physical process

Example: A greenhouse temperature sensor reports 31 °C. The controller compares it with a 28 °C limit and activates a ventilation fan. The next measurements confirm whether the temperature decreases.

3. The sensing and data-acquisition chain

Figure 1. A real measurement passes through several stages before it becomes reliable digital information.

3.1 Physical phenomenon

The measurand is the quantity being measured: temperature, acceleration, relative humidity, light intensity, pressure, gas concentration, and so on. Clearly defining the measurand prevents a common mistake—choosing a sensor before understanding the actual problem.

3.2 Sensor conversion

The sensor changes an electrical property such as resistance, capacitance, current, voltage, charge, or frequency. The raw change may be extremely small and easily disturbed.

3.3 Signal conditioning

Signal conditioning prepares the sensor output for measurement. It may include:

• amplification: enlarge a small signal;
• attenuation: reduce a signal that is too large;
• filtering: remove unwanted frequencies and noise;
• linearization: compensate for a nonlinear sensor response;
• level shifting: move the signal into an ADC’s input range;
• excitation: supply current or voltage to a resistive/capacitive sensor;
• isolation and protection: prevent damage and reduce ground-related errors.

3.4 Analogue-to-digital conversion

An ADC performs two essential operations:

1. Sampling: measuring the analogue signal at particular times.
2. Quantization: mapping each sample to one of a finite number of digital codes.

For an ideal $N$-bit ADC with reference voltage $V_{ref}$, the approximate step size is:

$$\text{LSB size}=\frac{V_{ref}}{2^N}$$

For a 12-bit ADC with a 3.3 V reference:

$$\frac{3.3}{4096}\approx0.000806\text{ V}=0.806\text{ mV per count}$$

This number is resolution, not guaranteed accuracy. Reference error, noise, offset, gain error, nonlinearity, sensor tolerance, wiring, and calibration all affect the final result.

3.5 Digital processing and output

The microcontroller can calibrate, filter, timestamp, compress, classify, store, or transmit the data. It may also control an actuator through a DAC, PWM output, digital pin, or communication bus.

Worked example: temperature alarm

1. An NTC thermistor changes resistance with temperature.
2. A voltage divider converts resistance into voltage.
3. The ADC converts voltage into a code.
4. Firmware converts the code to resistance and then temperature.
5. A calibration correction is applied.
6. If temperature exceeds the limit, the node sends an alarm.
7. A gateway logs the event and activates a local fan or relay.

4. Sensor classification

The physical property being monitored is the first guide to sensor selection.

Sensors can also be classified as:

• active/self-generating: produce an electrical output from the measured energy, such as a thermocouple or photovoltaic sensor;
• passive/modulating: need external excitation, such as a thermistor or strain gauge;
• analogue: produce a continuous electrical signal;
• digital: contain conversion/logic and communicate coded values;
• contact: must touch the object or medium;
• non-contact: use light, sound, radiation, magnetic fields, or other remote effects;
• absolute, relative, or differential: depending on the reference used.

Important: “Digital sensor” does not mean “perfect sensor.” It still has tolerance, drift, response time, environmental limits, and calibration requirements.

5. How selected sensors work

Figure 2. Different physical effects eventually become measurable electrical changes.

5.1 MEMS capacitive accelerometer

A micro-electromechanical system (MEMS) accelerometer contains a tiny proof mass suspended by flexible structures. Acceleration causes the mass to move. Its movement changes the distances between capacitor plates, producing opposite changes in two capacitances, often written as $C_1$ and $C_2$.

Electronics measure the differential capacitance and output acceleration data. A stationary accelerometer also measures the effect of gravity, which is why a smartphone can determine orientation and control a maze/rolling-ball application.

Key points:

• measures acceleration along one or more axes;
• includes gravity unless the system estimates and removes it;
• bandwidth and sample rate determine what motion can be observed;
• offset, scale-factor error, vibration, mounting, and temperature affect accuracy.

5.2 Capacitive humidity sensor (hygrometer)

A capacitive humidity element places a moisture-sensitive dielectric between electrodes. Water molecules entering the material change its dielectric permittivity, which changes capacitance. An AC measurement circuit avoids simple DC effects and converts the capacitance into relative humidity (RH).

Humidity readings are affected by temperature, contamination, condensation, hysteresis, response time, and long-term drift. Outdoor installations need a protective membrane that allows water vapour to pass but reduces liquid water and dust exposure.

5.3 Thermistor

A thermistor is a semiconductor resistor whose resistance changes strongly with temperature.

• NTC: resistance decreases as temperature increases.
• PTC: resistance increases as temperature increases over its intended region.

An NTC is often connected in a voltage divider. Its response is nonlinear, so firmware may use a lookup table or the Beta/Steinhart-Hart model. Thermistor self-heating occurs when measurement current warms the component; lower excitation or short measurement pulses can reduce the error.

5.4 Thermocouple

A thermocouple uses two dissimilar conductors joined at a measurement junction. A voltage related to the temperature difference between the measurement junction and reference junction is produced by the thermoelectric (Seebeck) effect. The system therefore needs cold-junction compensation. Thermocouples suit wide temperature ranges but produce small voltages requiring careful amplification and noise control.

5.5 Photodiode

Photons absorbed in a semiconductor junction create electron-hole pairs. The junction’s electric field separates the charges, producing a current related to light intensity. A transimpedance amplifier commonly converts this small current to voltage.

Photodiodes can be fast and accurate but require attention to dark current, wavelength sensitivity, optical geometry, amplifier noise, and ambient light.

5.6 LDR

A light-dependent resistor changes resistance with illumination. It is simple and inexpensive, but generally slower and less precisely characterized than a photodiode. It is useful for basic day/night or brightness detection, not every high-speed optical measurement.

6. Selecting a sensor correctly

Ask these questions before buying a sensor:

1. What exact measurand and unit are required?
2. What is the minimum and maximum range?
3. What accuracy, precision, and resolution are necessary?
4. How quickly must the sensor respond?
5. What sample rate and bandwidth are required?
6. Is the output analogue, I²C, SPI, UART, pulse, or another interface?
7. What voltage, current, warm-up time, and sleep behaviour does it have?
8. What temperature, humidity, vibration, chemical, dust, or water exposure will it face?
9. Does it require calibration, compensation, or periodic replacement?
10. What happens when the sensor fails or produces an impossible value?

Accuracy means closeness to the true value. Precision means repeatability. A sensor can be precise but consistently wrong because of offset; calibration can often correct that offset.

7. Motes and sensor nodes

7.1 What is a sensor node?

A sensor node is a functional WSN device that senses, processes, and wirelessly transmits data. Its usual blocks are:

• sensor and analogue front end;
• ADC or digital sensor interface;
• microcontroller and memory;
• radio transceiver and antenna;
• power supply and power management; and
• optional storage, localization, or actuators.

7.2 What is a mote?

A mote is a small, low-power physical hardware platform implementing a sensor node. Classic examples include Mica2/MicaZ, TelosB/Tmote Sky, IRIS, and Imote2. Berkeley’s Smart Dust research and mote platforms helped make large, low-power WSN experiments practical.

The terms are often used interchangeably in practice, but the distinction helps: a mote is one implementation of the wider sensor-node concept.

7.3 Classic platforms from the lecture

• Mica2/MicaZ: widely used in early WSN research; MicaZ added an IEEE 802.15.4-compatible 2.4 GHz radio with up to 250 kbit/s in its common configuration.
• TelosB/Tmote Sky: low-power academic platform with integrated sensing options.
• IRIS: improved radio capabilities compared with earlier Mica-family platforms.
• MTS310 sensor board: example peripherals included acceleration, magnetic field, light, temperature, acoustics, and a sounder.
• SHIMMER: a wearable platform using a TI MSP430, CC2420 IEEE 802.15.4 radio, triaxial accelerometer, rechargeable Li-polymer battery, and microSD storage.

8. What is a wireless sensor network?

A WSN consists of multiple sensor nodes—sometimes hundreds or thousands—that cooperatively monitor a physical area or system. Data is communicated to a base station or sink and then delivered to remote devices for storage, processing, analysis, visualization, or control.

Figure 3. Nodes can sense and relay data; the sink collects it; the gateway connects the sensor field to applications.

A WSN may include:

• ordinary sensing nodes;
• relay/router nodes;
• cluster heads or aggregation nodes;
• a sink/base station;
• a gateway or border router;
• edge processing;
• Internet/cloud services; and
• user applications and actuators.

9. Brief history of WSNs

Figure 4. WSNs evolved as sensing, microelectronics, radios, networking, and computing became smaller and more energy efficient.

• 1940s–1980s: radar, telemetry, distributed sensing, and microelectronics created foundations.
• 1990s: distributed sensor-network research, MEMS, DARPA programs, and Smart Dust concepts advanced.
• Early 2000s: practical motes, TinyOS, energy-aware routing, and IEEE 802.15.4 enabled large experiments.
• 2006–2010: real deployments expanded in environmental, industrial, and healthcare fields; aggregation and harvesting gained importance.
• 2010–2015: 6LoWPAN/IPv6 and cloud platforms connected WSNs to the IoT.
• 2016–2020: LPWAN options such as LoRaWAN and NB-IoT supported wider-area systems.
• 2021 onward: edge computing, TinyML, AI-assisted operation, 5G integration, energy harvesting, and battery-less research increase local intelligence and sustainability.

The overall direction is:

Miniaturization → Low power → Connectivity → Intelligence → Sustainability

10. WSN communication

Traditional Wi-Fi (IEEE 802.11 family) is common and provides high data rates, but it may consume too much energy for tiny nodes that must operate for years. It can still be suitable when mains power, rechargeable batteries, or high bandwidth are available.

IEEE 802.15.4 was designed for low-rate, low-power, low-complexity wireless personal-area networking. Common WSN traffic consists of short, infrequent readings rather than video-sized data. A widely used 2.4 GHz PHY provides up to 250 kbit/s, while other bands/PHYs can differ. IEEE 802.15.4 defines lower-layer functions; higher technologies such as Zigbee, Thread, 6LoWPAN, or custom stacks build additional networking behaviour on top.

Radio reality

Wireless range is influenced by transmit power, antenna, frequency, obstacles, multipath, interference, receiver sensitivity, and data rate. A datasheet’s line-of-sight range is not a guarantee inside a building, near the ground, or beside metal machinery.

11. Single-hop and multi-hop networks

Figure 5. A star is simple; a mesh uses relays to increase coverage but needs routing.

Single-hop/star

Every sensor communicates directly with the base station.

Advantages:

• simple routing and management;
• low forwarding delay;
• ordinary nodes do not spend energy relaying other nodes’ traffic.

Limitations:

• distant nodes may need high transmit power;
• weak direct links may make large geographic coverage impossible;
• the base station is a central traffic and failure point.

Multi-hop/mesh

Nodes relay data for neighbours.

Advantages:

• shorter radio links can reduce the energy needed per transmission;
• coverage can extend beyond one radio hop;
• alternate paths may improve resilience.

Limitations:

• routing is necessary;
• every added hop can increase latency and loss probability;
• relay nodes spend energy forwarding other traffic;
• sleeping/duty-cycled nodes complicate path availability.

Multi-hop is not automatically more energy efficient. The total cost includes transmission, reception, idle listening, routing control, retransmissions, and uneven relay load.

12. Energy: the dominant WSN constraint

Nodes may be battery-powered, recharged, supplied by harvested solar/vibration/thermal energy, replenished through wireless power transfer, or discarded when depleted. When physical access is difficult, the node must survive for its mission duration.

12.1 CPU energy model from the lecture

The simplified CPU energy expression is:

$$E_{CPU}=E_{switch}+E_{leakage}$$ $$E_{CPU}=C_{total}V_{dd}^{2}+V_{dd}I_{leak}\Delta t$$

where:

• $C_{total}$ is effective switched capacitance;
• $V_{dd}$ is supply voltage;
• $I_{leak}$ is leakage current; and
• $\Delta t$ is the execution or idle interval.

Switching energy occurs as transistors charge and discharge capacitance. Its strong $V^2$ relationship explains why lower-voltage operation can save dynamic energy, within safe hardware limits.

Leakage energy exists even when logic is not switching. It accumulates with time and can matter during long idle periods.

This is a teaching model. Real MCU energy also depends on clock tree, memories, peripherals, regulator efficiency, sleep-state transitions, temperature, and workload.

12.2 Whole-node energy

The CPU is only one consumer. A practical budget includes:

$$E_{node}=E_{sense}+E_{process}+E_{radio}+E_{storage}+E_{idle}+E_{conversion\ losses}$$

The radio often dominates because receiving and idle listening can consume significant current. Energy-saving methods include duty cycling, event-triggered reporting, local aggregation, adaptive sampling, efficient routing, and turning off unused peripherals.

13. Self-management and unattended operation

WSNs may be deployed without precise planning—for example, dropped in a remote area, installed quickly during a disaster, or carried by moving robots or animals. A node may need to determine location, discover neighbours, configure parameters, find a route to the sink, and initiate its sensing role.

Figure 6. A self-managing network continually observes its state and adapts.

• Self-organization: adapt configuration based on system and environmental state.
• Self-optimization: monitor and improve use of limited energy, bandwidth, memory, and processing.
• Self-protection: recognize and resist intrusions and attacks.
• Self-healing: detect disruptions, identify failures, and restore useful operation.

Unattended operation also requires adaptation to changing topology, node density, link quality, traffic, and failures.

14. Wireless, decentralization, and design constraints

14.1 Attenuation and multi-hop effects

Radio power decreases with distance and obstacles. Multipath may strengthen or cancel the signal at different points. Multi-hop can extend coverage but adds queueing delay, retransmission opportunities, and dependence on sleeping relays.

14.2 Centralized versus decentralized decisions

A base station with global knowledge could calculate an “optimal” route, but collecting frequent global updates consumes energy and becomes slow in a changing network.

In a decentralized system, each node chooses from limited local information—such as neighbour quality and advertised path cost. The result may not be globally optimal, but it can be cheaper and more adaptable.

14.3 Hardware and software constraints

Common limitations include:

• low CPU speed to reduce energy;
• small RAM/flash to reduce cost and size;
• limited I/O and absence of GPS;
• small packet sizes and low data rates;
• limited operating-system features;
• restricted debugging and physical access; and
• harsh environmental conditions.

Good WSN software uses bounded memory, short messages, failure-aware state machines, power-safe timing, and graceful degradation.

15. WSN security

WSNs may monitor homes, healthcare, factories, transport, agriculture, and critical infrastructure. Unattended nodes and broadcast wireless links increase exposure.

Threats include:

• unauthorized access;
• eavesdropping;
• spoofed nodes or messages;
• replay of an old valid command;
• physical node capture and key extraction;
• data or firmware tampering;
• routing attacks; and
• denial of service through traffic, interference, or battery exhaustion.

Required protections include lightweight but appropriate encryption, authentication, integrity checks, key management, replay protection, secure boot, signed firmware updates, access control, safe commissioning, logging, intrusion detection, and a method to revoke a compromised node.

Security has an energy and memory cost, but omitting it can invalidate every measurement and control decision. Design security into the architecture rather than adding it after deployment.

16. Traditional networks, ad-hoc networks, and WSNs

Traditional networks versus WSNs

Sensor networks versus mobile ad-hoc networks

WSNs can contain far more densely deployed nodes. Sensor nodes are usually more constrained in energy, computation, and memory; are more failure-prone; may mainly broadcast; may not have globally meaningful IDs; and often generate data toward one or a few sinks. Traditional ad-hoc networks more often emphasize peer-to-peer communication among more capable devices.

17. Ubiquitous and calm computing

Ubiquitous computing—also called pervasive computing—places computing and sensing throughout the environment so services become context-aware and available where needed. WSNs act as a sensing backbone, enabling environments to perceive, reason, and respond without continuous human input.

Mark Weiser described a future where computers blend into everyday life. Calm technology informs people without constantly demanding attention.

Example: A smart office silently adjusts lighting and ventilation from occupancy, light, temperature, and air-quality sensors. It interrupts the user only when action is necessary, such as unsafe CO₂ or a system fault.

18. Computing trends that enabled WSNs

Several technology trends made sensor networks practical:

• processors became cheaper, faster, smaller, and more energy efficient;
• flash and other storage became denser and smaller;
• networking expanded from global infrastructure to local, ad-hoc, low-power links;
• displays and mobile devices improved visualization and interaction;
• MEMS enabled tiny mechanical sensing structures;
• cloud and edge computing made large-scale processing accessible.

Moore’s Law describes the historical growth in integrated-circuit transistor counts, while Bell’s Law of computer classes describes the repeated appearance of lower-cost computing classes and new applications. Neither is a promise that every system improves automatically; engineering must still manage energy, heat, cost, reliability, and software complexity.

Sensors can be treated as physical information servers: each exposes information about the real world. Gateways and standard interfaces help integrate them with controllers, enterprise systems, IP networks, web applications, and data analytics.

19. WSN applications

19.1 Military and public safety

• friendly-force, equipment, and ammunition monitoring;
• reconnaissance and terrain observation;
• battlefield surveillance and damage assessment;
• nuclear, biological, and chemical hazard detection.

These applications require reliability, security, concealment, ruggedness, and tolerance of node loss.

19.2 Healthcare and wearables

• remote physiological monitoring;
• tracking patients, clinicians, and equipment;
• medication support;
• vital-sign monitoring, fall/accident detection, and elderly care;
• high-fidelity motion analysis using wearable platforms such as SHIMMER/Mercury.

Healthcare systems require privacy, safe alarms, clinical validation, comfortable wearability, and clear handling of missing or corrupted data.

19.3 Environmental monitoring

• forest-fire and flood detection;
• habitat and biodiversity observation;
• precision agriculture and soil moisture;
• weather, temperature, pressure, and humidity monitoring.

The Great Duck Island project used a large sensor deployment to relay environmental measurements to a central device and onward through a satellite/Internet link.

19.4 Wildlife tracking: ZebraNet

GPS-equipped collars collected animal movement data. Peer-to-peer exchanges allowed data to move between zebras, so contact with a limited number of animals could retrieve information accumulated across the network. This illustrates delay-tolerant networking: data may wait until useful contacts occur instead of requiring continuous infrastructure.

19.5 Smart homes, offices, and museums

• appliances, lighting, heating, and ventilation control;
• occupancy and air-quality monitoring;
• interactive exhibits and visitor guidance;
• safety and security monitoring.

19.6 Transport, parking, and asset protection

• road sensors for traffic flow and real-time route updates;
• smart parking occupancy detection;
• vehicle and asset tracking;
• theft detection and alerting;
• accident and road-condition monitoring.

19.7 Industrial and commercial systems

• crop and process-condition monitoring;
• inventory and in-process part tracking;
• automated fault reporting;
• equipment condition and predictive maintenance;
• RFID-supported theft deterrence and traceability.

Industrial systems must handle metal, electrical noise, vibration, strict downtime requirements, legacy interfaces such as RS-485, and safe control behaviour.

20. Practical design example: smart soil-moisture network

Requirement

Monitor a large field and irrigate only dry zones.

Possible design

1. Capacitive soil-moisture probes connect to low-power motes.
2. Each node conditions and samples the sensor through an ADC.
3. Temperature compensation and calibration convert ADC counts to a useful moisture estimate.
4. Nodes send periodic summaries and immediate low-moisture alarms.
5. Multi-hop communication reaches a field gateway where direct links are unreliable.
6. The gateway stores data locally during Internet outages.
7. A server displays trends and recommends irrigation.
8. Valve commands are authenticated and confirmed by actuator feedback.

Questions the designer must answer

• How will probe readings vary with soil type and salinity?
• What measurement current avoids corrosion and self-heating?
• How often is sampling truly necessary?
• How are failed nodes distinguished from adequately wet soil?
• What happens when a relay, gateway, or Internet connection fails?
• How are keys installed and replaced?
• Does water saved justify node production, batteries, and maintenance?

21. Common beginner mistakes

• Confusing resolution with accuracy.
• Connecting a raw sensor directly to an ADC without checking voltage range or conditioning.
• Ignoring calibration, response time, drift, and environmental protection.
• Assuming a digital output has no measurement error.
• Choosing Wi-Fi when a low-rate, multi-year battery node is required without completing an energy budget.
• Believing multi-hop always saves energy.
• Calculating only CPU current and ignoring sensors, radio, regulator, and idle listening.
• Treating silence as “normal” instead of detecting node failure.
• Sending sensitive or actuator data without authentication and replay protection.
• Designing only the node and forgetting gateway, data storage, user interface, maintenance, and end-of-life.

22. Lesson summary

• Sensing converts information from the physical world into electrical signals.
• The data-acquisition chain includes the sensor, conditioning, ADC, digital processing, communication, and sometimes an actuator.
• Accelerometers and humidity sensors can use capacitance changes; thermistors use resistance; thermocouples use thermoelectric voltage; photodiodes use photon-generated current.
• A sensor node is a functional concept; a mote is a small physical implementation.
• A WSN combines many nodes with a sink/base station, gateway, applications, and possibly actuators.
• Single-hop is simple; multi-hop extends coverage but introduces routing, relay energy, delay, and reliability challenges.
• Energy, unattended self-management, wireless variability, decentralization, limited resources, and security shape every WSN decision.
• WSNs support military, healthcare, environmental, wildlife, home, transport, industrial, and commercial applications.
• A good design begins with the measurand and application requirement, then justifies the sensor, electronics, network, security, and energy strategy.

23. Knowledge check

1. What is the difference between a sensor and an actuator?
2. Why is signal conditioning used before an ADC?
3. What are sampling and quantization?
4. What is the ideal LSB size of a 10-bit ADC using a 3.3 V reference?
5. How does a capacitive accelerometer detect acceleration?
6. Why does a capacitive humidity sensor use a moisture-sensitive dielectric?
7. What is the usual relationship between temperature and resistance in an NTC thermistor?
8. Why does a thermocouple require cold-junction compensation?
9. What is the difference between a sensor node and a mote?
10. Give two advantages and two disadvantages of multi-hop communication.
11. Explain switching and leakage energy in the CPU model.
12. Name the four self-management abilities.
13. Why might decentralized routing be preferred even if it is not globally optimal?
14. Give four WSN security threats.
15. What does calm technology mean in a smart environment?

Show answers

1. A sensor measures the physical world and produces information; an actuator receives control and changes the physical world.
2. To amplify, filter, protect, linearize, excite, or level-shift the signal into a usable ADC range.
3. Sampling measures at discrete times; quantization maps each sample to a finite digital code.
4. $3.3/1024≈3.22$ mV per count.
5. Acceleration moves a proof mass, changing differential capacitances that electronics convert to an acceleration value.
6. Absorbed water changes dielectric permittivity, and therefore capacitance, in relation to relative humidity.
7. Temperature increases while resistance decreases.
8. It measures a temperature difference; the reference-junction temperature must be known or compensated.
9. Sensor node is the functional concept; mote is a specific small, low-power physical platform implementing it.
10. Advantages: greater coverage, shorter links, alternate paths. Disadvantages: routing complexity, relay energy, added delay, additional loss opportunities.
11. Switching energy charges/discharges capacitances and depends strongly on voltage; leakage consumes energy over time even without useful switching.
12. Self-organization, self-optimization, self-protection, and self-healing.
13. It avoids the communication cost and delay of continuously collecting global information, especially in a large/changing network.
14. Any four: eavesdropping, spoofing, replay, node capture, tampering, routing attack, unauthorized access, denial of service.
15. Computing informs and assists without unnecessarily demanding the user’s attention.

24. Practical activity

Choose one application: smart classroom, elderly-care wearable, forest-fire detector, industrial motor monitor, smart parking, or precision agriculture.

Prepare a design containing:

1. measurand and required range;
2. selected sensor and operating principle;
3. signal-conditioning and ADC/interface needs;
4. mote/node blocks;
5. single-hop or multi-hop topology with justification;
6. sampling and message policy;
7. energy-saving plan;
8. node/gateway failure behaviour;
9. three security controls;
10. calibration and test plan; and
11. sustainability and societal-impact discussion.

Further reading

• Course sources: Lecture 1 Part 1 - Course Introduction, Lecture 1 Part 2 - Fundamentals of WSNs, and Intro Sensor Technologies, Kristianstad University, 2026.
• R. Faludi, Building Wireless Sensor Networks, O’Reilly/CRC Press, 2017.
• G. C. M. Meijer, Smart Sensor Systems, Wiley, 2008.
• IEEE 802.15.4 standards overview.
• NIST: Security for IoT Sensor Networks.
• Zephyr ADC documentation and examples for a practical embedded ADC software perspective.