10  WSN Sensor Nodes and Hardware

In 60 Seconds

A sensor node has four subsystems – sensing, processing, communication, and power – where the radio dominates energy at 3-30x more than the others combined. Multi-hop routing saves energy but creates hotspots near the sink where relay nodes die first. WSNs are data-centric (not address-centric), resource-constrained, and rely on duty cycling to extend battery life from days to years.

Minimum Viable Understanding

• A sensor node has four subsystems – sensing, processing, communication, and power – where the communication radio dominates energy consumption (3-30x more than the other subsystems combined).
• Multi-hop routing saves energy by relaying data through intermediate nodes instead of transmitting directly over long distances, but creates a hotspot problem near the sink where relay nodes deplete batteries much faster.
• WSNs differ fundamentally from traditional networks: they are data-centric (not address-centric), operate under extreme resource constraints, and use duty cycling to extend battery life from days to years.

Sensor Squad: Inside a Tiny Sensor Node

Sammy the Sensor looked at all the parts that made up his body. “I have four main parts,” he explained to his friends.

“My sensing unit is like my eyes and ears – I can feel temperature, see light, and detect motion!”

Max the Microcontroller chimed in: “And I’m the brain! I take what Sammy senses and turn it into numbers a computer can understand. But I only have a tiny amount of memory – like trying to remember things with a very small notebook.”

“Don’t forget about me!” said the radio transceiver. “I’m like a walkie-talkie that lets Sammy talk to other sensor nodes. But talking uses SO much energy – way more than sensing or thinking!”

Bella the Battery sighed. “That’s why I’m always tired. The radio uses up most of my energy. It’s like how shouting across a room takes way more effort than whispering to someone next to you. That’s why we pass messages through nearby friends instead of shouting all the way to the base station – it’s called multi-hop routing and it saves me a LOT of energy!”

10.1 Learning Objectives

By the end of this chapter, you will be able to:

• Analyze Node Architecture: Differentiate the four subsystems (sensing, processing, communication, power) and their interactions
• Calculate Energy Trade-offs: Apply the radio power law to multi-hop vs direct transmission decisions
• Compare WSN to Traditional Networks: Identify fundamental differences in scale, power, and protocols
• Design for Constraints: Work within memory, computation, and communication limitations

For Beginners: WSN Sensor Nodes and Hardware

Wireless sensor networks are groups of small, battery-powered devices that cooperate to monitor their surroundings. Think of them as a team of tiny reporters spread across a field, a building, or even a forest, each measuring things like temperature, vibration, or motion and relaying their observations to a central point. Understanding WSNs is fundamental to many IoT applications.

10.2 Prerequisites

• WSN Introduction: Understanding of WSN fundamentals, history, and network topologies
• Sensor Fundamentals: Knowledge of sensor operation and characteristics

Related Chapters

Previous:

• WSN Introduction - WSN history and network architectures

Continue Learning:

• WSN Swarm Behavior - Distributed intelligence patterns
• WSN Communication Patterns - N-to-1 data flow
• WSN Energy Management - Power optimization

Deep Dives:

• Energy-Aware Considerations - Battery lifetime analysis
• Prototyping Hardware - Sensor node platforms

---

10.3 Characteristics of Sensor Nodes

⏱️ ~12 min | ⭐⭐ Intermediate | 📋 P05.C31.U02

Key Concepts

• Core Concept: Fundamental principle underlying WSN Sensor Nodes and Hardware — understanding this enables all downstream design decisions
• Key Metric: Primary quantitative measure for evaluating WSN Sensor Nodes and Hardware performance in real deployments
• Trade-off: Central tension in WSN Sensor Nodes and Hardware design — optimizing one parameter typically degrades another
• Protocol/Algorithm: Standard approach or algorithm most commonly used in WSN Sensor Nodes and Hardware implementations
• Deployment Consideration: Practical factor that must be addressed when deploying WSN Sensor Nodes and Hardware in production
• Common Pattern: Recurring design pattern in WSN Sensor Nodes and Hardware that solves the most frequent implementation challenges
• Performance Benchmark: Reference values for WSN Sensor Nodes and Hardware performance metrics that indicate healthy vs. problematic operation

Sensor nodes are the fundamental building blocks of WSNs, each comprising several key components that enable autonomous sensing, processing, and communication.

10.3.1 Hardware Components

10.3.1.1 Sensor Node Architecture

A sensor node consists of four key subsystems working together to enable autonomous sensing and communication:

Subsystem	Components	Power Draw	Function
Sensing	ADC, sensors	1-10 mW	Data acquisition from environment
Processing	MCU, memory	10-100 mW	Local computation, protocol execution
Communication	Radio, antenna	50-300 mW	Data transmission/reception
Power	Battery, harvester	N/A	Energy supply and management

Power Distribution Reality

The communication subsystem typically consumes 3-30× more power than sensing and processing combined. This fundamental asymmetry drives WSN protocol design toward minimizing radio usage through duty cycling, data aggregation, and local processing.

Putting Numbers to It

Power breakdown for a Tmote Sky sensor node (typical WSN platform):

Subsystem	Active Power	Duration per Cycle	Energy per Cycle
Sensing (temperature + humidity)	5 mW	20 ms	\(5 \times 20 = 100 \text{ µJ}\)
Processing (8 MHz MSP430)	3 mW	50 ms	\(3 \times 50 = 150 \text{ µJ}\)
Radio TX (CC2420 at 0 dBm)	52.2 mW	5 ms	\(52.2 \times 5 = 261 \text{ µJ}\)
Radio RX (listening)	56.4 mW	10 ms	\(56.4 \times 10 = 564 \text{ µJ}\)
Sleep	0.015 mW	915 ms	\(0.015 \times 915 = 13.7 \text{ µJ}\)

Total energy per 1-second cycle: \(100 + 150 + 261 + 564 + 13.7 = 1{,}088.7 \text{ µJ} \approx 1.09 \text{ mJ}\)

Percentage breakdown: Radio (TX + RX) = 75.8%, Processing = 13.8%, Sensing = 9.2%, Sleep = 1.3%

Key insight: Despite being active for only 15 ms out of 1,000 ms (1.5% duty cycle), the radio consumes 76% of the energy. This is why WSN protocols obsess over reducing transmissions – a single extra transmission per cycle increases energy consumption by 24%.

Figure 10.1: Sensor node architecture block diagram showing four subsystems: Sensing Unit (sensors and ADC), Processing Unit (microcontroller and memory), Communication Unit (radio transceiver), and Power Unit (battery).

Figure 10.2: Energy Timeline View: This diagram illustrates a typical 1-second duty cycle. Based on Tmote Sky measurements, the node sleeps for 915ms (consuming 0.014 mJ), wakes to sense for 20ms at 5mW (0.1 mJ), processes for 50ms at 3mW (0.15 mJ), and transmits for 5ms at 52.2mW (0.26 mJ). Despite transmission occupying only 0.5% of cycle time, radio TX+RX together consume over 75% of total energy. This visualization helps students understand why WSN protocols focus obsessively on minimizing radio transmissions.

Figure 10.3: Basic components of a sensor node - microcontroller, radio transceiver, sensors, memory, and power supply

Academic Reference: WSN Node Architecture (IIT Kharagpur NPTEL)

Wireless sensor node architecture showing key components

Source: IIT Kharagpur - NPTEL Introduction to Internet of Things

Figure 10.4: Examples of sensor nodes showing various form factors and capabilities

Figure 10.5: Single source single object detection in WSN - one sensor detecting one target

Figure 10.6: Single source multiple object detection - one sensor tracking multiple targets simultaneously

Figure 10.7: Multiple source single object detection - collaborative sensing of one target by multiple sensors

Figure 10.8: Multiple source multiple object detection - distributed sensing network tracking multiple targets

Figure 10.9: Multiple source multiple object detection scenario II - advanced collaborative target tracking

Sensing Unit:

• Sensors: Transduce physical phenomena into electrical signals (temperature, humidity, light, pressure, accelerometer, etc.)
• Analog-to-Digital Converter (ADC): Converts analog sensor readings to digital values for processing
• Multiple Sensor Integration: Modern nodes often integrate multiple sensor types (multi-modal sensing)

Processing Unit:

• Microcontroller (MCU): Executes application logic, protocols, and data processing
• Typical Specifications: 8-16 bit processors, MHz clock speeds, KB-MB RAM, KB-MB flash memory
• Examples: ARM Cortex-M series, TI MSP430, Atmel AVR, Arduino platforms

Communication Unit:

• Radio Transceiver: Enables wireless communication with other nodes
• Common Standards: IEEE 802.15.4, Bluetooth Low Energy (BLE), LoRa, Sub-GHz radios
• Antenna: Internal or external for signal transmission/reception
• Typical Range: 10-100 meters (short-range), up to several kilometers (long-range LPWAN)

Power Supply:

• Primary Batteries: Non-rechargeable (AA, coin cells), long lifetime but limited capacity
• Rechargeable Batteries: Li-ion, Li-polymer, enabling solar or energy harvesting recharge
• Energy Harvesting: Solar panels, vibration harvesters, thermal energy, RF energy harvesting
• Typical Capacity: mAh to few Ah, determining operational lifetime

Optional Components:

• Storage: SD cards or flash memory for local data logging
• Positioning: GPS modules for location awareness (power-intensive)
• Actuators: Enable control functions (valves, motors, switches)
• Security: Cryptographic co-processors for secure communication

10.3.2 Node Capabilities

Sensing Capabilities:

• Measurement accuracy and precision
• Sampling frequency and resolution
• Multi-modal sensing (combining multiple sensor types)
• Calibration and drift management

Processing Capabilities:

• Data preprocessing and filtering
• Feature extraction and compression
• Local decision making
• In-network data aggregation

Communication Capabilities:

• Data transmission rates (typically kbps to hundreds of kbps)
• Communication range (meters to kilometers)
• Multi-hop routing support
• Protocol support (MAC, routing, transport layers)

Energy Capabilities:

• Battery capacity and type
• Power consumption profiles (active, sleep, deep sleep modes)
• Energy harvesting potential
• Expected operational lifetime

10.3.3 Multi-hop Communication

Multi-hop routing is a fundamental characteristic of WSNs, enabling nodes beyond direct radio range to communicate by relaying messages through intermediate nodes.

Figure 10.10: Multi-hop routing path in WSN showing data forwarding from source to sink through intermediate nodes

10.3.3.1 Why Multi-hop? The Radio Power Law

Radio transmission power follows the path loss model where required power scales with distance:

\[P_{tx} \propto d^n\]

where: - \(P_{tx}\) = transmission power - \(d\) = distance to receiver - \(n\) = path loss exponent (typically 2-4 for wireless propagation)

Energy Benefit Example:

Consider transmitting data 100 meters to a sink:

Approach	Path Loss (n=2)	Path Loss (n=4)	Relative Energy
Direct: 1 hop × 100m	\(100^2 = 10,000\)	\(100^4 = 100,000,000\)	1× (baseline)
Multi-hop: 10 hops × 10m	\(10 \times 10^2 = 1,000\)	\(10 \times 10^4 = 100,000\)	10× savings (n=2) 1000× savings (n=4)

viewof total_dist = Inputs.range([50, 500], {value: 100, step: 10, label: "Total distance to sink (m)"})

viewof num_hops = Inputs.range([1, 20], {value: 5, step: 1, label: "Number of hops"})

viewof path_loss_n = Inputs.range([2, 4], {value: 2, step: 0.5, label: "Path-loss exponent n"})

{
  const hop_dist = total_dist / num_hops;
  const direct_loss = Math.pow(total_dist, path_loss_n);
  const multihop_loss = num_hops * Math.pow(hop_dist, path_loss_n);
  const savings = direct_loss / multihop_loss;

  return html`<div style="background: var(--bs-light, #f8f9fa); padding: 1rem; border-radius: 8px; border-left: 4px solid #3498DB; margin-top: 0.5rem;">
    <p><strong>Per-hop distance:</strong> ${hop_dist.toFixed(1)} m</p>
    <p><strong>Direct path loss factor:</strong> ${direct_loss.toExponential(2)}</p>
    <p><strong>Multi-hop total path loss factor:</strong> ${multihop_loss.toExponential(2)}</p>
    <p><strong>Energy savings from multi-hop:</strong> <span style="color:#16A085; font-size:1.2em;">${savings.toFixed(1)}×</span></p>
    <p style="font-size:0.85em; color:#7F8C8D;">Higher n (obstructed environments) gives greater multi-hop benefit. Savings = (total_dist)^n / (hops × (hop_dist)^n)</p>
  </div>`;
}

Putting Numbers to It

Concrete energy calculation for 100m transmission: Using the Friis transmission equation with realistic WSN parameters:

\[P_{\text{rx}} = P_{\text{tx}} \times G_{\text{tx}} \times G_{\text{rx}} \times \left(\frac{\lambda}{4\pi d}\right)^2\]

For CC2420 radio at 2.4 GHz (λ = 12.5 cm), antenna gain G = 0 dBi (unity), receiver sensitivity = -95 dBm, and path loss exponent n = 3.5 (urban environment):

Direct transmission (100m):

• Required TX power: \(P_{\text{tx}} = -95 \text{ dBm} + 10 \times 3.5 \times \log_{10}(100) = -95 + 70 = -25 \text{ dBm} = 0.32 \text{ mW}\)
• Packet duration: 4 ms for 50-byte packet at 250 kbps
• Energy: \(0.32 \text{ mW} \times 4 \text{ ms} = 1.28 \text{ µJ}\) (unrealistic – CC2420 minimum is 0 dBm = 1 mW)
• Actual CC2420 at 0 dBm: \(52.2 \text{ mW} \times 4 \text{ ms} = 208.8 \text{ µJ}\)

Multi-hop (5 hops × 20m each):

• Required TX power per hop: \(-95 + 10 \times 3.5 \times \log_{10}(20) = -95 + 45.5 = -49.5 \text{ dBm}\) (well below -25 dBm minimum)
• Use CC2420 at -10 dBm power level: 31.2 mW
• Energy per hop: \(31.2 \times 4 = 124.8 \text{ µJ}\)
• Total for 5 hops: \(5 \times 124.8 = 624 \text{ µJ}\)

Counterintuitive result: Multi-hop uses 3× MORE energy than direct transmission (624 µJ vs 208.8 µJ) because CC2420’s minimum power level (-25 dBm) is already sufficient for 100m, and the 5-hop approach wastes energy on radio startup/shutdown overhead and per-packet processing at each intermediate node. Multi-hop saves energy ONLY when direct distance exceeds the radio’s maximum range or when using adaptive power control that can reduce per-hop power below the hardware minimum.

The Multi-hop Energy Paradox

While multi-hop routing dramatically reduces per-hop transmission power, it introduces three critical trade-offs:

1. Latency Increase: Each hop adds processing and queuing delay. A 10-hop path with 10ms per-hop delay = 100ms total vs. 10ms for direct transmission.

2. Intermediate Node Burden: Nodes near the sink relay massive traffic from all downstream nodes, depleting batteries 10-100× faster (the “hotspot problem”).

3. Routing Overhead: Maintaining multi-hop routes requires periodic control messages, topology discovery, and route repair mechanisms that consume energy and bandwidth.

Solution: Use multi-hop routing strategically—essential for long-range communication beyond single-hop radio range, but avoid unnecessary hops. Deploy multiple sinks to reduce average hop count, implement energy-aware routing to balance load, and use adaptive transmission power control to optimize per-hop distances.

10.3.3.2 Multi-hop Trade-offs

Advantages:

• Extended Range: Coverage beyond single-hop radio limits
• Lower Transmission Power: Reduced per-hop energy consumption
• Redundancy: Multiple paths increase reliability
• Scalability: Supports large deployments

Challenges:

• Higher Latency: Each hop adds delay (10-50 ms typical per hop)
• Increased Complexity: Routing protocols, topology maintenance
• Hotspot Problem: Nodes near sink relay all traffic
• Overhead: Routing messages consume energy and bandwidth

Practical Design Guidelines:

1. Minimize Hop Count: Deploy sufficient gateways to keep average path length < 5 hops
2. Balance Energy: Use energy-aware routing to avoid overloading specific nodes
3. Adaptive Transmission Power: Adjust radio power based on required hop distance
4. Hierarchical Topologies: Cluster-based architectures reduce long multi-hop chains

10.3.4 WSN vs. Traditional Networks

Wireless Sensor Networks differ fundamentally from conventional computer networks in scale, economics, and operational constraints:

Aspect	Traditional Network	WSN
Node Cost	$100-1000+ per device	$1-10 per node
Node Count	10s-100s devices	100s-10,000s nodes
Deployment	Planned, structured	Ad-hoc, dense, unplanned
Power Source	Wired AC or rechargeable	Battery (months-years), harvesting
Data Flow	Any-to-any, peer-to-peer	Many-to-one (convergecast to sink)
Lifetime	Years with maintenance	Months-years, unattended
Bandwidth	Mbps-Gbps	kbps-100s kbps
Processing	GHz multi-core CPUs	MHz 8-32 bit MCUs
Memory	GB RAM, TB storage	KB RAM, MB flash
QoS	Guaranteed latency/bandwidth	Best-effort, graceful degradation
Security	Centralized PKI, strong crypto	Lightweight crypto, resource-constrained
Mobility	Limited (laptops, phones)	Mostly static, some mobile sinks/nodes
Address Assignment	DHCP, static IP	Auto-configuration, hierarchical IDs
Network Discovery	Manual configuration	Self-organizing, distributed

Why These Differences Matter for Design

Traditional network assumptions break in WSNs:

1. IP-Based Protocols Too Heavy: Full TCP/IP stack requires 50-100 KB flash + KB RAM—exceeding many sensor node capabilities. Solutions: 6LoWPAN (compressed IPv6), CoAP (lightweight HTTP alternative).

2. Always-On Radio Assumption: Traditional protocols assume devices listen continuously. WSNs use duty cycling (1-10% radio active) to extend battery life 10-100×, requiring asynchronous MAC protocols.

3. Reliability Through Retransmission: TCP’s reliable delivery via ACKs and retransmissions is energy-prohibitive in WSNs where radio dominates power. WSNs use redundancy (multiple sensors measure same phenomenon) and in-network aggregation instead.

4. End-to-End Addressing: Traditional networks address individual devices. WSNs often use data-centric routing (query “get temperature > 30°C from Region A”) where specific node identity is irrelevant—only data matters.

5. Centralized Management: Traditional networks rely on SNMP, centralized controllers. WSNs require distributed, autonomous operation due to limited connectivity to management infrastructure.

Design for WSN constraints: minimize radio transmissions, use application-specific protocols, embrace statistical aggregation over perfect reliability, deploy redundancy at the sensing layer (cheap nodes) rather than network layer (expensive protocols).

10.3.5 Resource Constraints

Energy Constraints: The most critical limitation in WSNs. Radio communication typically consumes the most energy, followed by sensing and processing.

Typical Power Consumption:

• Active mode (sensing + transmitting): 10-50 mW
• Idle listening: 1-20 mW
• Sleep mode: 1-100 μW
• Deep sleep: < 1 μW

Memory Constraints:

• RAM: 2 KB - 64 KB (working memory)
• Flash/ROM: 32 KB - 512 KB (program and data storage)
• Impact: Limits code complexity, buffering, and local storage

Computational Constraints:

• Processor Speed: 1-48 MHz typical
• Computation: Simple arithmetic, limited cryptography, basic signal processing
• Impact: Complex algorithms must be simplified or offloaded

Communication Constraints:

• Bandwidth: Typically 20-250 kbps (802.15.4), up to 1 Mbps (BLE)
• Range: Limited by power and frequency
• Reliability: Subject to interference, obstacles, and environmental factors

Idle Listening Drains Batteries Faster Than You Think

Many WSN deployments fail because developers underestimate idle listening power consumption. A sensor node with its radio in idle listening mode (waiting to receive) consumes 10-20 mA continuously–nearly as much as actively transmitting. A 2000 mAh battery at 15 mA idle current lasts only 5.5 days, but with proper duty cycling (1% active, deep sleep at ~10 µA) extends to over 13 months. See the detailed worked example below for protocols and code-level fixes.

Worked Example: Multi-Hop Energy Hotspot Analysis

Scenario: A forest monitoring WSN has 100 temperature sensors arranged in a 10x10 grid, each 50 meters apart. The gateway is at one corner. After 6 months, nodes near the gateway fail while edge nodes still have 70% battery. Analyze the energy hotspot problem and design a mitigation strategy.

Given:

• Grid size: 10×10 nodes, 50m spacing (500m × 500m field)
• Gateway at position (0,0)
• Radio range: 75m (allows diagonal hops)
• Data rate: Each node transmits 1 packet/hour (own data)
• Packet size: 50 bytes
• TX energy: 50 µJ/byte, RX energy: 30 µJ/byte
• Battery: 2000 mAh at 3.3V = 23.76 kJ

Steps:

1. Calculate hop count for each node position:
   • Node at (9,9): 6 hops to gateway (diagonal path)
   • Node at (0,9): 9 hops (vertical path)
   • Average hop count: ~5 hops across network
2. Calculate traffic load at gateway-adjacent nodes:
   • Nodes at (0,1), (1,0), (1,1) relay traffic from entire network
   • Total network packets: 100 nodes × 1 packet/hour = 100 packets/hour
   • Node (1,1) relays approximately 60% of traffic = 60 packets/hour
   • Node’s own packet: 1 packet/hour
   • Total: 61 packets/hour for hotspot node vs. 1 packet/hour for edge node
3. Calculate energy consumption difference:
   • Edge node energy/hour: 1 packet × 50 bytes × 50 µJ = 2.5 mJ
   • Hotspot node energy/hour:
     • TX own data: 50 bytes × 50 µJ = 2.5 mJ
     • TX relayed: 60 × 50 × 50 µJ = 150 mJ
     • RX relayed: 60 × 50 × 30 µJ = 90 mJ
     • Total: 242.5 mJ/hour (97× more than edge node)
4. Calculate battery lifetime disparity:
   • Edge node lifetime: 23,760,000 mJ / 2.5 mJ/hour = 9,504,000 hours = 1,085 years (theoretical)
   • Hotspot node lifetime: 23,760,000 mJ / 242.5 mJ/hour = 97,979 hours = 11.2 years
   • With realistic 10% duty cycle overhead: Edge = 10+ years, Hotspot = 1.1 years
5. Design mitigation: Deploy 4 gateways:
   • Place gateways at (0,0), (0,9), (9,0), (9,9)
   • Maximum hop count reduced from 6 to 3
   • Maximum relay load per node: 25% of network = 25 packets/hour
   • New hotspot energy: ~65 mJ/hour (74% reduction)

Result: Adding 3 additional gateways reduces worst-case relay burden by 74%, extending hotspot node lifetime from 1.1 years to 4+ years while edge nodes remain unaffected.

Key Insight: In convergecast WSNs, nodes near the sink experience traffic proportional to the entire network size. Gateway placement and quantity are critical design decisions - the cost of additional gateways is often far less than the cost of replacing depleted batteries in hard-to-access hotspot locations.

Worked Example: Choosing Between Battery and Energy Harvesting

Scenario: An environmental agency needs to deploy 50 air quality sensors on streetlights in a city center for 5-year continuous monitoring. Compare battery-only versus solar harvesting approaches and recommend the most cost-effective solution.

Given:

• Deployment: 50 sensors on streetlight poles (partial shading)
• Required lifetime: 5 years (43,800 hours)
• Average power consumption: 250 µW (after optimization with 0.5% duty cycle)
• Location: Northern European city (limited winter sunlight: 2 hours/day December)
• Solar panel efficiency in urban canyon: 40% of rated due to shading
• Maintenance cost: $150/visit (cherry picker + technician)

Steps:

1. Calculate battery-only requirements: \[\text{Energy needed} = 250 \mu W \times 43800 \text{ hours} = 10.95 \text{ Wh}\]

   • Primary lithium battery: 3.6V × 19Ah = 68.4 Wh (Tadiran TL-5930)
   • Cost per battery: $45
   • Available energy margin: 6.2× safety factor (accounts for self-discharge, temperature)

2. Calculate solar harvesting system:

   • Winter minimum: 2 hours × 40% efficiency = 0.8 effective sun-hours/day
   • Required harvest rate: 250 µW × 24h / 0.8h = 7.5 mW panel output
   • With 50% harvesting efficiency: 15 mW panel needed
   • Small solar panel (50×50mm): $8, provides 100 mW peak = 10 mW urban effective
   • Need 2× panels: $16
   • Battery for 3 days backup: 250 µW × 72h = 18 mWh → 50 mAh LiPo: $5
   • Charge controller: $12
   • Total harvesting BOM: $33

3. Calculate 5-year total cost comparison:

   Approach	Initial Cost	5-Year Maintenance	Total per Node
   Battery-only	$45	$0 (no visits)	$45
   Solar harvest	$33	$150 × 0.5 visits (panel cleaning)	$108

4. Factor in failure modes:

   • Battery failure rate: 1% over 5 years (lithium primary very reliable)
   • Solar system failure rate: 15% (panel damage, connector corrosion, controller failure)
   • Battery-only: 50 × $45 + 0.5 replacement visits × $150 = $2,325
   • Solar: 50 × $33 + 7.5 failures × $150 repair + 25 cleanings × $150 = $5,655

Result: For this 5-year deployment with low power consumption (250 µW), battery-only is 2.4× more cost-effective than solar harvesting. The high-quality primary lithium battery provides reliable, maintenance-free operation.

Key Insight: Energy harvesting is not always the best choice. For deployments under 10 years with very low power budgets (<1 mW average), high-capacity primary batteries often provide better reliability and lower total cost of ownership than harvesting systems. Reserve harvesting for: (1) >10 year deployments, (2) >1 mW average power, or (3) locations where battery replacement is physically impossible.

10.3.6 Node Types and Roles

Regular Sensor Nodes:

• Most common and numerous
• Perform sensing and basic routing
• Minimal configuration and capabilities

Cluster Head Nodes:

• Coordinate groups of sensor nodes
• Aggregate and forward data
• More capable hardware and energy resources
• May be rotated among nodes to balance energy consumption

Gateway/Sink Nodes:

• Interface between WSN and external networks
• More powerful processing and communication
• Often mains-powered
• Bridge different protocols and networks

Sink Node Architecture

Figure 10.11: Sink node architecture illustrating the gateway’s role in bridging sensor networks to external infrastructure.

Actuator Nodes:

• Include actuation capabilities
• Respond to sensed conditions or commands
• Examples: irrigation valves, motors, heaters

Mobile Nodes:

• Mounted on vehicles, robots, or carried by people
• Dynamic topology challenges
• Data collection through mobility (data mules)

10.4 Test Your Understanding

Knowledge Check

Question 1: A sensor node’s radio transceiver consumes 200 mW during transmission and the node transmits for 20 ms every second. What percentage of the duty cycle is the radio active, and why is this important?

1. 2% – minimizing radio-on time is the primary strategy for extending battery life
2. 20% – the radio is on most of the time
3. 50% – sensing and transmitting share equal time
4. 98% – the node is mostly transmitting

Answer

a) 2% – minimizing radio-on time is the primary strategy for extending battery life. The radio is active for only 20 ms out of each 1000 ms cycle (2%). Since radio communication dominates energy consumption (3-30x more than other subsystems), keeping the radio duty cycle low is the single most impactful technique for extending network lifetime from days to months or years.

Question 2: In a 10x10 grid WSN with the gateway at one corner, why do nodes near the gateway fail first even though all nodes have identical batteries?

1. The gateway interferes with nearby node radios
2. Nearby nodes must relay traffic from the entire network, consuming energy 10-100x faster
3. Nodes far from the gateway never transmit any data
4. The gateway drains power from adjacent nodes

Answer

b) Nearby nodes must relay traffic from the entire network, consuming energy 10-100x faster. In convergecast (many-to-one) data flow, all sensor data must pass through nodes adjacent to the gateway. A node one hop from the sink may relay 60+ packets per hour from downstream nodes while edge nodes transmit only their own data (1 packet/hour). This unequal relay burden is called the “energy hole” or “hotspot” problem.

Question 3: Which of the following is the most significant difference between WSNs and traditional computer networks for protocol design?

1. WSNs have fewer nodes
2. WSNs use wireless instead of wired connections
3. WSN protocols are data-centric rather than address-centric, and must minimize radio usage
4. WSNs have faster processors

Answer

c) WSN protocols are data-centric rather than address-centric, and must minimize radio usage. Traditional networks address individual devices (IP addresses), but in WSNs, queries target data characteristics (“get temperature > 30C from Region A”) regardless of which specific node provides it. Additionally, the extreme energy asymmetry (radio uses 3-30x more power) means WSN protocols are designed around minimizing transmissions, unlike traditional protocols that assume always-on connectivity.

Worked Example: Multi-Hop vs. Direct Transmission Energy Trade-off

Scenario: Sensor node 300m from sink. Compare direct transmission vs. 3-hop multi-hop routing.

Radio Power Model: \(P_{tx} = P_{elec} + P_{amp} \times d^n\) where n = path loss exponent (n=2 free space, n=4 obstructed)

Given Parameters:

• Electronics energy: \(P_{elec}\) = 50 µJ/byte (fixed cost for radio circuits)
• Amplifier energy: \(P_{amp}\) = 0.1 µJ/byte/m² (for n=2) or 0.0013 µJ/byte/m⁴ (for n=4)
• Packet size: 100 bytes
• Receive energy: 50 µJ/byte

Direct Transmission (300m, 1 hop):

Path loss n=2 (free space): \[E = 100 \times (50 + 0.1 \times 300^2) = 100 \times (50 + 9,000) = 905,000 \mu J = 905 \text{ mJ}\]

Path loss n=4 (urban/forest): \[E = 100 \times (50 + 0.0013 \times 300^4) = 100 \times (50 + 10,530,000) = 1,053,000,000 \mu J = 1,053 \text{ J} (!)\]

3-Hop Multi-Hop (100m per hop):

Each hop transmit (n=2): \[E_{tx} = 100 \times (50 + 0.1 \times 100^2) = 100 \times 1,050 = 105,000 \mu J = 105 \text{ mJ}\]

Each hop receive (intermediate nodes): \[E_{rx} = 100 \times 50 = 5,000 \mu J = 5 \text{ mJ}\]

Total for 3 hops (n=2): - Hop 1 TX: 105 mJ - Hop 1 RX + Hop 2 TX: 5 + 105 = 110 mJ - Hop 2 RX + Hop 3 TX: 5 + 105 = 110 mJ - Total: 325 mJ

Total for 3 hops (n=4): - Each hop TX: \(100 \times (50 + 0.0013 \times 100^4) = 100 \times (50 + 130{,}000) = 13{,}005{,}000 \text{ µJ} = 13{,}005 \text{ mJ}\) - With RX overhead: 3 × (13,005 + 5) = 39,030 mJ ≈ 39 J

Energy Comparison Table:

Approach	n=2 (free space)	n=4 (obstructed)	Savings
Direct (1 hop × 300m)	905 mJ	1,053,000 mJ	Baseline
Multi-hop (3 hops × 100m)	325 mJ	39,030 mJ	2.8× (n=2), 27× (n=4)

Key Insights:

1. Multi-hop provides 2.8–27× energy savings depending on environment (n=2 to n=4)
2. Savings increase with path loss exponent: higher n means obstructed environments benefit more from shorter hops
3. In obstructed environments (forests, urban), direct transmission at 300m with n=4 consumes 1,053 J per packet—essentially impossible for battery-powered nodes transmitting frequently
4. The electronics cost (50 µJ/byte) becomes dominant in very short-hop scenarios, setting a floor on multi-hop efficiency

Practical Implication: Direct 300m transmission at n=4 would drain a 2000 mAh battery (26.6 kJ) in approximately 25 packets (~2.6 minutes). With 3-hop routing at n=4 (39 J per packet transfer): ~680 packets (~11 hours). At n=2 with 3 hops (325 mJ/packet): ~81,800 packets (~57 days). This illustrates why multi-hop routing is essential in obstructed environments, and why reducing path loss through node density and placement is a critical deployment strategy.

Decision Framework: Sensor Node Platform Selection

Use Case	MCU Type	Radio	Battery	Cost Target	Example Platform
Research/Prototyping	32-bit ARM, 64KB+ RAM	802.15.4 + BLE	Rechargeable Li-ion	$40-80	nRF52840-DK, ESP32
Low-cost deployment (<$10/node)	8-bit AVR/MSP430, 2KB RAM	802.15.4 only	2× AA primary	$5-10	Arduino ProMini + RFM69
Ultra-long lifetime (>10 years)	MSP430, deep sleep <1µA	Sub-GHz LoRa	2× AA lithium primary	$15-25	Moteino, RFM95
High-data-rate (video/audio)	ARM Cortex-A, Linux	Wi-Fi or 4G	Mains or large Li-ion	$100-200	Raspberry Pi + camera
Extreme environment (industrial)	Industrial-grade MCU	PoE Ethernet or WirelessHART	Mains (PoE)	$200-500	Industrial IoT gateway
Mobile/wearable	Ultra-low-power ARM	BLE 5.0	Coin cell CR2032	$20-40	Nordic Thingy:91, nRF52

Selection Process:

1. Determine power budget: Battery life requirement → average power budget → radio duty cycle → MCU sleep modes
2. Calculate memory needs: Protocol stack (6LoWPAN: 40KB, Zigbee: 50KB, BLE: 30KB) + application + buffers
3. Match radio to range/throughput: <100m + low data → 802.15.4. <1km + ultra-low-power → LoRa. <10m + mobile → BLE.
4. Cost-vs-effort trade-off: $5 custom board (high development effort) vs. $40 dev kit (rapid prototyping)

Common Mistake: Underestimating Idle Listening Power Consumption

The Mistake: Calculating battery life assuming “radio is off during sleep,” but configuring radio in “idle” mode (ready to receive) instead of true deep sleep.

Why It’s Wrong: Many radios have multiple “sleep” states: - Deep sleep: Radio powered off, 1-10 µA consumption, ~5 ms wake-up time - Idle listening: Radio receiver on, ready for incoming packets, 10-20 mA consumption, 100 µs wake-up time

Beginners often leave the radio in idle listening (because it seems necessary to “hear incoming messages”) without realizing this uses 1000-20,000× more power than deep sleep.

Real-World Failure: Forest monitoring deployment calculated 5-year battery life based on “1% duty cycle with 20 mA active, 10 µA sleep.” Actual configuration left radio in idle listening mode (thinking nodes need to “always listen for coordinator beacons”). Result: 10 mA × 99% + 20 mA × 1% = 10.09 mA average (not 0.21 mA calculated). Battery life: 8 days instead of 5 years (227× shorter).

Power Breakdown Comparison:

Intended design (deep sleep):
- Sleep 99%: 10 µA × 0.99 = 0.0099 mA
- Active 1%: 20 mA × 0.01 = 0.2 mA
- Total: 0.21 mA → 2000 mAh / 0.21 mA = 9,523 hours (13 months)

Actual deployment (idle listening):
- Idle 99%: 10 mA × 0.99 = 9.9 mA
- Active 1%: 20 mA × 0.01 = 0.2 mA
- Total: 10.1 mA → 2000 mAh / 10.1 mA = 198 hours (8 days)

The Fix:

1. Use synchronized MAC protocols (S-MAC, RI-MAC) that coordinate sleep schedules so nodes wake simultaneously – no idle listening needed
2. Implement preamble sampling (B-MAC, X-MAC) where senders transmit long preambles and receivers wake periodically for brief channel samples
3. Use wake-up radio: Add secondary ultra-low-power radio (10 µA) that triggers main radio wake-up, eliminating idle listening entirely

Code-Level Fix (ESP32 example):

// WRONG: Radio left in idle mode
WiFi.mode(WIFI_STA);  // Station mode, radio active, ~80 mA

// CORRECT: Radio fully powered down
WiFi.mode(WIFI_OFF);   // Radio off, <10 µA
esp_wifi_stop();       // Ensure radio is completely disabled

Rule of Thumb: Measure actual current consumption with a multimeter during “sleep” – if you see >1 mA, the radio is NOT in deep sleep. Track down the idle listening culprit and eliminate it.

Common Pitfalls

1. Prioritizing Theory Over Measurement in WSN Sensor Nodes and Hardware

Relying on theoretical models without profiling actual behavior leads to designs that miss performance targets by 2-10×. Always measure the dominant bottleneck in your specific deployment environment — hardware variability, interference, and load patterns routinely differ from textbook assumptions.

2. Ignoring System-Level Trade-offs

Optimizing one parameter in isolation (latency, throughput, energy) without considering impact on others creates systems that excel on benchmarks but fail in production. Document the top three trade-offs before finalizing any design decision and verify with realistic workloads.

3. Skipping Failure Mode Analysis

Most field failures come from edge cases that work in the lab: intermittent connectivity, partial node failure, clock drift, and buffer overflow under peak load. Explicitly design and test failure handling before deployment — retrofitting error recovery after deployment costs 5-10× more than building it in.

🏷️ Label the Diagram

Code Challenge

10.5 Summary

This chapter covered the fundamental architecture and constraints of wireless sensor nodes:

• Node Architecture: Four subsystems – sensing (ADC + sensors), processing (MCU + memory), communication (radio transceiver), and power (battery/harvester) – with communication consuming 3-30x more power than all others combined.
• Multi-Hop Communication: Reduces per-hop transmission energy significantly (2.8× savings for n=2 free space, 27× for n=4 obstructed, over a 300m path split into 3 hops) but introduces latency, routing overhead, and the critical hotspot problem near the sink.
• Resource Constraints: Sensor nodes operate under extreme limitations – KB of RAM, MHz processors, kbps bandwidth, and mAh batteries – requiring specialized lightweight protocols.
• Node Types: Regular sensors, cluster heads, gateway/sink nodes, actuator nodes, and mobile nodes each serve distinct roles in the network hierarchy.
• WSN vs Traditional Networks: Data-centric design, duty-cycled radios, statistical reliability through redundancy, and distributed autonomous operation distinguish WSNs from conventional networks.

10.6 Knowledge Check

Quiz: WSN Sensor Nodes and Hardware

Match the WSN Component to Its Role

Order the Steps: Designing a Battery-Powered Sensor Node

10.7 What’s Next

Topic	Chapter	Description
WSN Swarm Behavior	WSN Swarm Behavior and IoT	How simple local rules create complex network-wide behaviors
WSN Communication	WSN Communication Patterns	N-to-1 data flow patterns and convergecast
WSN Energy Management	WSN Energy and Duty Cycling	Power optimization techniques for long network lifetime
Sensor Fundamentals	Sensor Fundamentals	Sensor operation principles and characteristics
Energy-Aware Design	Energy-Aware Considerations	Battery lifetime analysis and harvesting strategies