28  WSN Review: Architecture and Design

In 60 Seconds

WSN architecture follows a four-tier hierarchy: sensor nodes (collect), cluster heads (aggregate, reducing traffic 85-95%), gateways (bridge to IP networks), and cloud (store and analyze). Topology selection is the most consequential early decision – star topologies suit fewer than 50 nodes with reliable power, mesh suits large-area deployments needing self-healing, and cluster-tree balances energy efficiency with scalability for 100-10,000 node networks.

28.1 Learning Objectives

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

• Diagram WSN Architecture: Illustrate the hierarchical organization of wireless sensor networks from physical to application layers
• Select Appropriate Topologies: Choose optimal network topologies based on deployment constraints
• Evaluate Energy Trade-offs: Analyze energy consumption patterns and quantify optimization strategies
• Apply Architecture Decision Trees: Use structured decision frameworks to guide WSN deployment choices

MVU: Minimum Viable Understanding

If you only have 5 minutes, focus on these architecture essentials:

1. Four-layer architecture: Physical (sensor nodes) -> Network (cluster heads) -> Gateway (protocol bridge) -> Application (cloud)
2. Topology decision: If mains-powered gateways are available, use Star topology. If sensor spacing >100m, use hierarchical clustering. Otherwise, use star with repeaters
3. Energy trade-offs: Sleep mode (1-10 uA) vs. idle listening (15 mA) vs. transmission (20 mA) – idle listening is the silent battery killer

The architecture selection decision tree in this chapter is the single most valuable reference for WSN deployment planning.

Sensor Squad: Architecture Review Time!

Sammy the Sensor is building a diagram of the whole team!

“Let me draw our organization chart,” says Sammy: - Bottom: Sensor nodes like me, Lila, Max, and Bella – we collect data! - Middle: Cluster heads – they are like team captains who gather everyone’s reports - Top-ish: The Gateway – it translates our sensor language into Internet language - Top: The Cloud – where all the data gets stored and analyzed

Bella adds: “And the most important thing to remember? Choosing the RIGHT architecture saves TONS of battery! When some engineers used mesh networking for 200 sensors, batteries died in 14 HOURS. When they switched to star topology with 10 gateways, batteries lasted 36 DAYS!”

Max says: “That is 60x longer! Always ask: can we add more gateways instead of using mesh?”

28.2 Prerequisites

Required Chapters:

• WSN Overview Fundamentals - Core WSN concepts
• Wireless Sensor Networks - WSN architecture
• Sensor Fundamentals - Sensor basics

Technical Background:

• Sensor node architecture
• Energy constraints in WSNs
• Data aggregation concepts

WSN Architecture Layers:

Layer	Function	Example
Application	User services	Environmental monitoring
Transport	Reliability	Congestion control
Network	Routing	RPL, AODV
Data Link	MAC access	CSMA/CA, TDMA
Physical	Radio	802.15.4

Estimated Time: 15 minutes

Related Chapters

WSN Review Series:

• WSN Overview Review (Index) - Series overview
• WSN Review: Knowledge Checks - Quick assessment questions
• WSN Review: Scenario Analysis - Detailed scenario walkthroughs
• WSN Review: Comprehensive Assessment - Advanced topics and summary

WSN Overview Series:

• WSN Overview Fundamentals - Core WSN concepts
• Wireless Sensor Networks - Architecture and design

Advanced Topics:

• WSN Coverage Fundamentals - Area and target coverage
• WSN Tracking Fundamentals - Mobile target tracking

For Beginners: How to Use This Review

What is this chapter? This review chapter consolidates Wireless Sensor Network (WSN) architecture concepts. It’s designed to test and reinforce your understanding of WSN design decisions.

Key Concepts to Master:

Concept	Why It Matters
Sensor Nodes	Basic building blocks of WSNs
Network Topology	How sensors connect and communicate
Data Aggregation	Reducing network traffic efficiently
Energy Efficiency	Maximizing battery life

Recommended Path:

1. Study WSN Overview Fundamentals first
2. Review WSN Implementations
3. Return here to test your understanding

Common Misconception: “More Mesh Routing = Better Reliability”

The Misconception: Many developers assume full mesh topologies with multi-hop routing provide superior reliability and resilience compared to simple star topologies. After all, mesh networks have multiple redundant paths—what could go wrong?

The Reality - Quantified:

A 2019 smart building deployment compared mesh vs. star architectures for 200 temperature sensors:

Full Mesh Deployment (AODV routing):

• Initial promise: “Multiple paths ensure reliability!”
• Reality after 3 months:
  • Battery life: 14 hours (nodes drained overnight)
  • Packet delivery ratio: 67% (routing loops, collisions, overhead)
  • Network failures: Daily (nodes dying disrupted routing tables)
  • Operational cost: $182,500/year (daily battery replacements at $50/visit)
  • Root cause: Routing overhead (route discovery broadcasts) + relay burden consumed 3.3 Ah/day per node vs. 0.055 Ah/day for direct transmission

Star Topology with 10 Mains-Powered Gateways:

• Each gateway serves 20 sensors in direct single-hop communication
• Results after 12 months:
  • Battery life: 36+ days (60x improvement)
  • Packet delivery ratio: 99.2% (no routing, minimal collisions)
  • Network failures: Rare (gateway redundancy, no battery cascade failures)
  • Operational cost: $12,000/year (monthly battery service)
  • Total system cost: $7,000 hardware vs. mesh $9,000 (mesh needs more RAM/CPU for routing)

Why Mesh Failed:

1. Energy asymmetry: Inner nodes relayed traffic for ~10 neighbors, consuming 60x more energy than sensing alone
2. Routing overhead: AODV route discovery broadcasts flooded network every time a node moved or failed
3. Collision amplification: 200 competing transmitters created exponential backoff waste
4. The hotspot problem: Nodes near sinks died in days, creating coverage holes that isolated outer nodes

The Correct Principle: “Always ask: Can I add more gateways instead of using mesh?” The answer is usually yes, and it’s dramatically cheaper in both energy and operational cost. Mesh routing is a solution for situations where gateway placement is truly impossible—not a default architecture choice.

When Mesh Makes Sense (Rare Cases):

• Sensor spacing >200m AND terrain prevents gateway placement (dense forest, mountains)
• Temporary deployments where infrastructure installation is impractical (military, disaster response)
• Mobile sensor networks (vehicular, UAV swarms) where fixed infrastructure doesn’t exist

Real-world Lesson: “We spent $40K on custom mesh firmware optimization, achieving 28-hour battery life. Then someone suggested adding 5 PoE gateways for $1,000. Battery life jumped to 5+ years. Sometimes the best algorithm is: don’t route.”

28.3 WSN System Architecture

⏱️ ~10 min | ⭐ Foundational | 📋 P05.C33.U01

WSN System Architecture (Hierarchical Organization)

Layer	Component	Function	Key Features
Physical Layer	Sensor Nodes (6 nodes shown)	Sense & transmit data	MCU + Radio + Sensors, Battery powered
Network Layer	Cluster Heads (2 nodes)	Data aggregation & relay	Reduce traffic with statistical summaries (min/max/avg)
Gateway Layer	Gateway/Sink	Protocol bridge	Network coordinator, Wi-Fi/Ethernet/Cellular
Application Layer	Cloud Platform	Storage & analytics	ML, user interface, command distribution

Data Flow (Upward): Sensor Nodes → (multi-hop, 802.15.4) → Cluster Heads → (aggregated data) → Gateway → (Internet) → Cloud

Control Flow (Downward): Cloud → Gateway → Cluster Heads → Sensor Nodes (commands, configuration, actuation)

WSN hierarchical architecture with four layers

Figure 28.1: WSN hierarchical architecture showing four layers: Physical layer with battery-powered sensor nodes (temperature, humidity, motion, light, pressure, gas), Network layer with cluster heads performing data aggregation (min/max/avg), Gateway layer providing protocol bridge (Wi-Fi/Ethernet/Cellular), and Application layer with cloud platform for analytics, ML, and user interface. Data flows upward through multi-hop 802.15.4 to cluster heads to gateway to cloud; control commands flow downward.

Figure 28.2: Alternative View: Architecture Selection Decision Tree: While the layered architecture diagram above shows WSN components, this decision tree helps practitioners select the right components based on deployment constraints. Start with power availability (the most critical constraint), then evaluate data rate, range, and latency requirements. Each path leads to specific technology recommendations with expected outcomes. This decision-oriented view transforms abstract architecture knowledge into actionable deployment guidance. Key insight: most architecture failures stem from ignoring the first question (power availability) and jumping directly to protocol selection.

28.4 WSN Application Domains

WSN Application Domains

Domain	Applications
Environmental	Forest Fire Detection, Air Quality Monitoring, Water Quality, Weather Stations, Flood Detection
Industrial	Process Monitoring, Equipment Health, Factory Automation, Supply Chain Tracking, Predictive Maintenance
Infrastructure	Structural Health, Bridge Monitoring, Pipeline Leak Detection, Smart Grid, Traffic Monitoring
Agriculture	Precision Farming, Soil Moisture, Crop Health, Livestock Tracking, Greenhouse Automation
Healthcare	Patient Monitoring, Vital Signs, Medication Tracking, Fall Detection, Elderly Care
Military	Battlefield Surveillance, Target Tracking, Intrusion Detection, Chemical Detection, Perimeter Security
Smart City	Parking Management, Waste Management, Street Lighting, Noise Monitoring, Public Safety
Wildlife	Animal Tracking, Habitat Monitoring, Migration Patterns, Population Studies, Conservation

WSN application domains across eight sectors

Figure 28.3: WSN application domains mindmap showing eight major deployment areas: Environmental monitoring (fire, air, water quality), Industrial automation (process monitoring, predictive maintenance), Infrastructure health (bridges, pipelines, smart grid), Agriculture (precision farming, soil moisture, livestock), Healthcare (patient monitoring, fall detection), Military (surveillance, intrusion detection), Smart City (parking, waste, lighting), and Wildlife conservation (animal tracking, habitat monitoring).

28.5 WSN Energy Trade-offs

Understanding energy consumption patterns is critical for WSN deployment success:

WSN energy consumption states and duty cycling solutions

Figure 28.4: WSN energy consumption breakdown showing four primary energy states: Sleep mode (1-10 µA, excellent for 5+ year battery life), Idle listening (15 mA continuous, poor - only 5 days battery life), Transmission (20 mA at 1% duty cycle, 4.8 mAh/day), and Sensing (5 mA at 0.1%, 1.2 mAh/day). Solutions include duty cycling protocols (S-MAC synchronized, X-MAC asynchronous, Wake-up radio) that enable deep sleep 99.9% of time while maintaining network responsiveness.

Quick Check: Energy States

Which WSN operating mode is the “silent battery killer” that consumes nearly as much power as active transmission?

28.6 WSN Energy Mode Calculator

Use this calculator to compare how different radio operating modes affect battery lifetime.

viewof duty_cycle_pct = Inputs.range([0.1, 100], {value: 1, step: 0.1, label: "Active duty cycle (%)"})

viewof battery_cap_arch = Inputs.range([200, 10000], {value: 2800, step: 100, label: "Battery capacity (mAh)"})

{
  const dc = duty_cycle_pct / 100;
  const sleep_ua = 5;       // µA deep sleep
  const idle_ma  = 15;      // mA idle listening
  const tx_ma    = 20;      // mA transmission
  const sense_ma = 5;       // mA sensing

  // Daily consumption for each mode (mAh)
  const daily_always_idle  = idle_ma * 24;
  const daily_duty_cycled  = ((tx_ma * dc) + (sleep_ua / 1000 * (1 - dc))) * 24;
  const daily_always_sleep = (sleep_ua / 1000) * 24;

  const life_idle   = battery_cap_arch / (daily_always_idle * 365);     // years
  const life_cycled = battery_cap_arch / (daily_duty_cycled * 365);
  const life_sleep  = battery_cap_arch / (daily_always_sleep * 365);

  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>Battery Lifetime Comparison</strong> (${battery_cap_arch} mAh battery)</p>
    <table style="width:100%; border-collapse: collapse;">
      <thead><tr style="background:#2C3E50; color:white;"><th style="padding:6px 8px; text-align:left;">Mode</th><th style="padding:6px 8px; text-align:right;">Daily (mAh)</th><th style="padding:6px 8px; text-align:right;">Lifetime</th></tr></thead>
      <tbody>
        <tr><td style="padding:4px 8px;">Always idle (no duty cycling)</td><td style="padding:4px 8px; text-align:right;">${daily_always_idle.toFixed(1)}</td><td style="padding:4px 8px; text-align:right; color:#E74C3C;">${(life_idle * 365).toFixed(1)} days</td></tr>
        <tr style="background:#f0f0f0;"><td style="padding:4px 8px;">Duty cycled at ${duty_cycle_pct}%</td><td style="padding:4px 8px; text-align:right;">${daily_duty_cycled.toFixed(3)}</td><td style="padding:4px 8px; text-align:right; color:${life_cycled >= 5 ? '#16A085' : '#E67E22'};">${life_cycled.toFixed(1)} years</td></tr>
        <tr><td style="padding:4px 8px;">Pure sleep (reference)</td><td style="padding:4px 8px; text-align:right;">${daily_always_sleep.toFixed(4)}</td><td style="padding:4px 8px; text-align:right; color:#16A085;">${life_sleep.toFixed(0)} years</td></tr>
      </tbody>
    </table>
    <p style="margin-top:0.5rem; font-size:0.9em; color:#7F8C8D;">Duty cycling improvement over always-idle: <strong>${(daily_always_idle / daily_duty_cycled).toFixed(0)}×</strong></p>
  </div>`
}

28.7 WSN Topology Selection Decision Tree

Choose the right topology based on deployment constraints:

WSN topology selection decision tree

Figure 28.5: WSN topology selection decision tree starting with mains power availability: If YES → Star topology (20-50 sensors per gateway, 5+ year battery, simple routing, best for buildings/infrastructure); If NO → Check sensor spacing: If >100m → Hierarchical clustering (LEACH rotation, 3-6 month battery for cluster heads, best for remote/forest deployments); If <100m → Star with repeaters (2-3 mains repeaters, 3+ year battery, minimal overhead, best for accessible campuses/farms/industrial sites).

28.8 Test Your Understanding

Test Your Understanding

Question 1: In the WSN hierarchical architecture, what is the primary function of cluster heads at the Network Layer?

1. Sensing environmental data and transmitting raw readings
2. Data aggregation and relay, reducing traffic with statistical summaries (min/max/avg)
3. Providing Wi-Fi connectivity to end users
4. Running machine learning models on sensor data

Answer

b) Data aggregation and relay, reducing traffic with statistical summaries (min/max/avg). Cluster heads sit at the Network Layer between sensor nodes and the gateway. They receive data from multiple sensor nodes, aggregate it into summaries (reducing 20 individual messages to 1 summary), and relay the compressed data to the gateway. This reduces total network traffic by 70-90%.

Putting Numbers to It

Aggregation ratio determines traffic reduction:

\[\text{Traffic Reduction} = 1 - \frac{1}{N} = \frac{N-1}{N}\]

where \(N\) is the number of sensor readings aggregated.

Worked example: A cluster head receives 20 sensor readings (each 8 bytes) and computes min/max/avg (24 bytes total): - Without aggregation: \(20 \times 8 = 160\) bytes transmitted - With aggregation: \(24\) bytes transmitted - Reduction: \((160-24)/160 = 85\%\) traffic savings

Question 2: A WSN deployment has mains power available at gateway locations and sensor spacing is within 50m. According to the topology decision tree, which topology should be used?

1. Full mesh with AODV routing
2. Hierarchical clustering with LEACH
3. Star topology with 20-50 sensors per gateway
4. Hybrid TDMA + CSMA/CA mesh

Answer

c) Star topology with 20-50 sensors per gateway. The topology decision tree starts with: “Can you deploy mains-powered gateways?” – YES leads to Star topology. Since sensor spacing is within 50m (direct communication range), no multi-hop routing is needed. Star gives 5+ year battery life with no relay burden, at the lowest operational cost.

Question 3: Which WSN application domain typically requires the highest sensing frequency and lowest latency tolerance?

1. Environmental monitoring (forest fire, air quality)
2. Smart agriculture (soil moisture, crop health)
3. Industrial automation (vibration analysis, process monitoring)
4. Smart city (parking, waste management)

Answer

c) Industrial automation. Industrial monitoring requires sensing frequencies up to 1000 Hz (for vibration analysis) and latency tolerance below 100ms (for safety-critical gas detection). Environmental monitoring tolerates minutes-to-hours latency with 5-60 minute sensing intervals. Agriculture needs 10-30 minute intervals. Smart city applications typically sense every 1-15 minutes with seconds-to-minutes latency tolerance.

28.9 Worked Example: WSN Architecture Selection for a Bridge Structural Health Monitor

Scenario: Transport Scotland needs to monitor the Forth Road Bridge (2.5 km span) for structural integrity. The system must detect micro-vibrations, tilt changes, and corrosion indicators across the bridge’s suspension cables, deck sections, and tower foundations.

Requirements:

• 80 accelerometer nodes (vibration: 200 Hz sampling)
• 40 inclinometer nodes (tilt: 1 sample/minute)
• 20 corrosion sensors (electrochemical: 1 sample/hour)
• Data must reach the monitoring centre within 5 seconds for accelerometers, 60 seconds for others
• Deployment lifetime: 10 years (bridge inspection cycles)
• No mains power on the bridge deck – solar + battery only

Architecture Options Evaluated:

Criterion	Star (single sink)	Flat Mesh	3-Tier Hierarchical
Range coverage (2.5 km)	Needs 10+ repeaters	Self-healing, but 15+ hops	Cluster heads every 200 m
Accelerometer data rate	200 Hz x 2 bytes x 3 axes = 1.2 KB/s	Overloads intermediate nodes	Cluster head aggregates locally
Latency (accelerometer)	50 ms (1 hop)	800 ms (15 hops)	120 ms (2 hops to CH, 1 to sink)
Node battery life	3.2 years (high TX power)	1.8 years (relay burden)	5.1 years (short-range to CH)
Single point of failure	Base station = total loss	None (self-healing)	CH failure loses 1 cluster
Cost (140 sensors + infra)	GBP 42,000	GBP 38,000	GBP 52,000

Selected Architecture: 3-Tier Hierarchical

• Tier 1 (Sensor nodes): 140 low-power nodes with 50 m radio range, sleeping 95% of time
• Tier 2 (Cluster heads): 13 solar-powered cluster heads (one every 200 m), each managing 10-11 sensors
• Tier 3 (Base station): 2 redundant base stations (one at each tower) with 4G backhaul

Energy Budget for Accelerometer Node:

Mode	Duration/hour	Current	Energy/hour
Sampling (200 Hz)	60 min	8 mA	26.4 mWh
TX to cluster head (50 m)	36 sec (1.2 KB/s bursts)	25 mA	0.83 mWh
Sleep	0 min (always sampling)	0.005 mA	0.017 mWh
Total			27.2 mWh

• Battery: 19,000 mWh (2x D-cell lithium, rated for -20C to +60C)
• Solar panel: 50 mW (mini panel, 4 hrs effective sun/day in Scotland) = 200 mWh/day
• Net daily energy: +200 - (27.2 x 24) = +200 - 652.8 = -452.8 mWh/day (solar insufficient alone)
• Battery-only lifetime: 19,000 / 652.8 = 29 days (unacceptable)

Solution: Duty-cycled accelerometers sample at 200 Hz for 10 seconds every 5 minutes (3.3% duty cycle):

• Revised sampling energy: 27.2 x 0.033 = 0.9 mWh/hr
• Daily consumption: (0.9 + 0.83 + 0.017) x 24 = 41.9 mWh/day
• Solar surplus: 200 - 41.9 = +158.1 mWh/day (net positive, infinite lifetime)
• Event-triggered continuous mode activates if vibration exceeds threshold, draining battery at 652.8 mWh/day for up to 29 days of continuous monitoring during anomalies

Key Insight: The architecture selection was driven by two constraints: (1) the 2.5 km bridge span requires multi-hop communication, eliminating star topology, and (2) the high data rate of accelerometers (1.2 KB/s continuous) would collapse a flat mesh through relay burden. Hierarchical architecture with solar-powered cluster heads solves both by keeping high-traffic paths short (50 m) while solar panels at cluster heads provide indefinite operation.

28.10 Concept Relationships

Concept	Relates To	Relationship Type	Significance
Hierarchical Architecture	Cluster Heads	Compositional	Network Layer uses cluster heads to aggregate sensor data, reducing traffic by 70-90%
Star Topology	Mains Power	Prerequisite	Star topology requires mains-powered gateways; without them, hierarchical clustering is needed
Energy Trade-offs	Idle Listening	Dominant Factor	Idle listening (15 mA continuous) drains batteries 15,000x faster than sleep mode (1 µA)
Cluster Head Rotation	LEACH Protocol	Energy Balancing	Random rotation ensures no single node bears aggregation burden, extending lifetime 8x
Data Aggregation	Transmission Energy	Reduction	5:1 aggregation reduces long-distance transmissions by 80%, exploiting quadratic distance-energy relationship
Topology Decision Tree	Power Availability	Primary Constraint	Mains power availability is the first decision point that determines all downstream architecture choices
Mesh Routing	Battery Life	Trade-off	Mesh routing enables self-healing but creates energy hotspots where relay nodes die 60x faster than edge nodes

28.11 See Also

Prerequisites:

• WSN Overview Fundamentals - Core concepts for wireless sensor networks
• Wireless Sensor Networks - Detailed architecture and design patterns

Related Review Chapters:

• WSN Review: Knowledge Checks - Quick assessment questions for self-testing
• WSN Review: Scenario Analysis - Detailed deployment scenario walkthroughs

Advanced Topics:

• WSN Coverage Fundamentals - Area coverage and K-coverage algorithms
• WSN Routing Fundamentals - Data-centric and hierarchical routing protocols

Common Pitfalls

1. Prioritizing Theory Over Measurement in WSN Review: Architecture and Design

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

28.12 Summary

This chapter covered the foundational architecture concepts for WSN review:

• Hierarchical Architecture: Four-layer organization from physical sensors to cloud applications
• Architecture Selection: Decision tree approach based on power, data rate, range, and latency constraints
• Application Domains: Eight major deployment areas spanning environmental, industrial, and smart city applications
• Energy Trade-offs: Understanding sleep vs. idle vs. transmission power consumption
• Topology Selection: Choosing between star, mesh, and hierarchical based on constraints

28.13 Knowledge Check

Quiz: WSN Review: Architecture and Design

Match Architecture Components to Their Functions

Order the WSN Architecture Selection Process

28.14 What’s Next

Topic	Chapter	Description
Scenario Analysis	WSN Review: Scenario Analysis	Quantified deployment scenario walkthroughs
Comprehensive Assessment	WSN Review: Comprehensive Assessment	Advanced topics and complete design guidance
Coverage Fundamentals	WSN Coverage Fundamentals	Area coverage and K-coverage algorithms
Routing Fundamentals	WSN Routing Fundamentals	Data-centric and hierarchical routing protocols