By the end of this chapter, you will be able to:
Sensor network deployment and sizing is about determining how many sensors you need and where to place them. Think of planning a sprinkler system for a garden – too few sprinklers leave dry spots, too many waste water and money. Getting the right number and placement of sensors ensures complete monitoring without unnecessary cost.
Before diving into deployment and energy topics, you should be familiar with:
If you only learn three things from this chapter:
Sammy the Sensor needs to watch over a big field. Where should he and his friends go?
Lila the Listener explains: “Imagine you’re playing hide and seek in a huge park. You need to place lookouts so EVERY spot is watched by at least one person.”
Two patterns for placing lookouts:
Max the Messenger warns: “But just because your lookouts can SEE everywhere doesn’t mean they can TALK to each other! Imagine two lookouts can see 10 meters but can only shout 8 meters. They can see a thief but can’t tell anyone!”
Bella the Battery shares her energy-saving trick: “I tell Sammy to SLEEP most of the time! He sleeps for 99 seconds, then wakes up for just 1 second to check the temperature and send a message. This is called duty cycling and it makes my energy last from 4 DAYS all the way to over a YEAR!”
The Myth: Many beginners assume that if sensor coverage reaches 100% of an area, the network will automatically be fully connected and functional.
The Reality: Coverage and connectivity are independent constraints. A 2008 study by Kumar et al. analyzing 1,000+ WSN deployments found that 23% of networks achieved full sensing coverage but had disconnected regions where nodes couldn’t communicate with the base station.
Real-World Impact:
R_c ≥ 2 × R_s (communication range must be at least 2× sensing range) prevents this issueWhy It Happens: Communication range (R_c) and sensing range (R_s) are determined by different hardware components. A temperature sensor might detect at 10m radius while the radio only transmits reliably to 8m, creating coverage without connectivity.
Solution: Always verify both coverage and connectivity during deployment planning. Use connectivity algorithms like spanning tree validation or multi-hop path analysis to ensure all nodes can reach the base station.
Proper sensor placement ensures complete area coverage:
This variant provides a practical calculation framework for determining the number of sensors needed based on area and deployment pattern.
Quick Formula Reference: | Pattern | Nodes Needed | When to Use | |———|————–|————-| | Grid | Area / (2R²) | Simple deployment, rectangular fields | | Hexagonal | Area / (2.6R²) | Cost-optimized, complex terrain | | Random | Area / (1.5R²) + 20% | Aerial drop, hostile terrain |
Coverage Calculation:
| Deployment Pattern | Formula | Coverage Efficiency |
|---|---|---|
| Grid | d = R × √2 | 100% |
| Hexagonal | d = R × √3 | 100% |
| Random | Varies | 70-90% |
| Triangular | d = R × √3 | 100% |
Where d = distance between nodes, R = sensing radius
Example Deployment Calculation:
Given:
- Area: 100m × 100m = 10,000 m²
- Sensing range: 10m radius
- Pattern: Hexagonal grid
Calculation:
- Node spacing: d = 10m × √3 ≈ 17.3m
- Nodes per row: 100m / 17.3m ≈ 6 nodes
- Number of rows: 100m / 15m ≈ 7 rows
- Total nodes needed: 6 × 7 = 42 sensors
Add 10% redundancy: 46 sensors deployedBeyond coverage, nodes must maintain communication paths back to the sink or gateway.
| Symbol | Meaning | Typical design rule |
|---|---|---|
R_s |
Sensing range | Maximum distance at which events can be detected |
R_c |
Communication range | Maximum distance for reliable radio links |
In many grid deployments we aim for:
R_c ≥ 2 × R_s.R_c is much smaller than this, you can achieve full sensing coverage but still have islands with no multi-hop path to the base station.Duty cycling dramatically extends network lifetime:
Duty Cycle Parameters:
| Parameter | Typical Value | Impact |
|---|---|---|
| Sleep Duration | 1-60 seconds | Battery life |
| Active Duration | 10-100 ms | Responsiveness |
| Listen Window | 5-20 ms | Receive success |
| Duty Cycle | 0.1-10% | Energy efficiency |
Energy Calculation:
Battery Lifetime Estimation
═══════════════════════════════════════
Parameters:
- Battery capacity: 2000 mAh
- Sleep current: 1 µA
- Active current: 20 mA
- Active time per cycle: 100 ms
- Cycle period: 10 seconds
Calculations:
- Active time ratio: 100ms / 10s = 1%
- Average current = 0.01 × 20mA + 0.99 × 0.001mA
= 0.2mA + 0.001mA
= 0.201 mA
- Lifetime = 2000mAh / 0.201mA
= 9,950 hours
≈ 414 days
With 50% battery efficiency: ~207 daysAdjust the parameters below to see how duty cycle affects battery lifetime:
Duty Cycling Energy Multiplier Effect: Compare three duty cycle scenarios for a 2000 mAh battery at 3.3V (6.6 Wh capacity):
Always-on (100% active at 20 mA): \[\text{Lifetime} = \frac{2000 \text{ mAh}}{20 \text{ mA}} = 100 \text{ hours} = 4.2 \text{ days}\]
10% duty cycle (0.1 × 20 mA + 0.9 × 0.001 mA = 2.001 mA): \[\text{Lifetime} = \frac{2000}{2.001} = 1000 \text{ hours} = 42 \text{ days}\] 10× improvement
1% duty cycle (0.01 × 20 mA + 0.99 × 0.001 mA = 0.201 mA): \[\text{Lifetime} = \frac{2000}{0.201} = 9{,}950 \text{ hours} = 415 \text{ days}\] 100× improvement
The diminishing returns are clear: reducing duty cycle from 100% to 10% gives 10× gain, but further reducing to 1% gives another 10× for a total 100× improvement. However, below 1%, sleep current (1 µA) becomes the limiting factor, so reducing active time further yields minimal benefit unless sleep current is also reduced.
Scenario: A precision agriculture deployment uses LEACH protocol with 100 soil sensors. After 3 months, the network shows uneven battery depletion - some nodes are at 30% while others remain at 80%. Diagnose the rotation problem and calculate correct LEACH parameters.
Given:
Steps:
Result: Adding residual energy weighting to LEACH threshold calculation improves network lifetime by 28% by preventing high-density region nodes from excessive cluster head duty.
Key Insight: Standard LEACH assumes uniform node distribution. Real deployments with non-uniform placement or varying cluster sizes require energy-aware threshold modifications. Monitor per-node battery levels during deployment and adjust p or add energy weighting to the threshold formula.
Scenario: A warehouse deploys 80 inventory tracking sensors across 4,000 m² (100m × 40m). Currently using one gateway at the entrance causes 6-hop routes and poor battery life. Determine optimal gateway count and placement for 2-year target lifetime.
Given:
Steps:
Calculate current energy consumption for worst-case node:
Evaluate 2-gateway configuration:
Evaluate 4-gateway configuration:
Evaluate 6-gateway configuration:
Cost-benefit analysis:
| Gateways | Worst Lifetime | Gateway Cost | Battery Replacements (2yr) | Total Cost |
|---|---|---|---|---|
| 1 | 42 days | $200 | 80 × 17 = 1,360 × $5 = $6,800 | $7,000 |
| 2 | 167 days | $400 | 80 × 4 = 320 × $5 = $1,600 | $2,000 |
| 4 | 485 days | $800 | 80 × 1.5 = 120 × $5 = $600 | $1,400 |
| 6 | 1,031 days | $1,200 | 0 (exceeds 2 years) | $1,200 |
Result: Deploy 6 gateways at calculated positions. While initial hardware cost is highest ($1,200), total 2-year cost is lowest due to zero battery replacements. The 6-gateway configuration achieves the 2-year target with 40% margin.
Key Insight: Gateway cost ($200 each) is often trivial compared to battery replacement labor, especially in hard-to-access locations like warehouse ceilings. Calculate total cost of ownership (TCO) including maintenance before minimizing gateway count. The “extra” gateways pay for themselves through reduced maintenance visits.
Solar-powered nodes can achieve near-perpetual operation when harvested energy covers long-term consumption:
Design guidelines:
E_harvested ≥ E_consumed over days/weeks.Scenario: Evaluate solar harvesting for 100-node soil moisture monitoring network in California vineyard. Determine if perpetual operation is feasible.
Given:
Energy Budget Calculation:
Daily energy consumption: \[E_{consumed} = 0.3 \text{ mA} \times 3.3\text{V} \times 24\text{h} = 23.76 \text{ mWh/day}\]
Available solar energy (6 peak sun hours):
System efficiency losses:
Net harvested energy:
Energy margin:
Winter worst-case (3 peak sun hours, cloudy):
Battery sizing for 5 days autonomy (cloudy period): \[E_{battery} = 23.76 \text{ mWh/day} \times 5 \text{ days} = 119 \text{ mWh}\] \[\text{Battery capacity} = \frac{119 \text{ mWh}}{3.3\text{V}} = 36 \text{ mAh}\] Use 200 mAh LiPo (5.6× margin for aging/temperature derating)
Component Selection:
Decision: Small panel (0.5W) provides 82× margin in summer, 20× in winter. Deploy with 200 mAh battery for 5-day autonomy. Total solar system cost ($10.50 × 100 = $1,050) is recovered vs. battery replacement in 1.5 years (vs. 4× annual replacement @ $300/year for non-solar).
| Factor | Choose Hexagonal | Choose Grid | Choose Random |
|---|---|---|---|
| Deployment precision | High (GPS/surveyed positions) | High (aligned to structure) | Low (aerial drop, difficult terrain) |
| Node count optimization | Critical (expensive sensors) | Moderate | Not possible (uncontrolled) |
| Terrain | Flat, accessible | Structured (building, factory) | Rough, inaccessible |
| Coverage target | >95% guaranteed | 90-95% acceptable | 70-90% (overprovisioned) |
| Installation method | Manual placement | Mounted to grid (poles, beams) | Scatter (drone, artillery) |
Efficiency comparison (same coverage):
When grid is preferred despite inefficiency:
Hybrid approach: Hexagonal where possible, grid where constrained by structure, random for inaccessible zones.
Before deploying, verify the Rc >= 2Rs rule with a simple check:
if communication_range >= 2 * sensing_range:
print("Coverage implies connectivity")
else:
print(f"WARNING: Isolated islands likely! Rc/Rs = {communication_range/sensing_range:.2f} < 2.0")See the Common Misconception callout at the top of this chapter for a detailed case study of what happens when this rule is violated.
Understanding how deployment and energy concepts interconnect:
| Concept | Builds On | Enables | Conflicts With | Complements |
|---|---|---|---|---|
| Hexagonal Deployment | Coverage geometry, sensing range | 23% node reduction vs grid | Structured mounting constraints | Optimal area coverage |
| R_c ≥ 2Rs Rule | Sensing/communication range specs | Guaranteed connectivity with coverage | Short-range radio hardware | Multi-hop routing viability |
| Duty Cycling | Low-power sleep modes | 100x battery extension | Real-time monitoring needs | Event-driven sensing |
| Gateway Placement | Communication range limits | Reduced hop count, hotspot mitigation | Installation access constraints | Load balancing strategies |
| Solar Harvesting | Energy budget calculation, MPPT | Perpetual operation | Indoor/shaded locations | Ultra-low duty cycles |
| LEACH Clustering | Hexagonal deployment, duty cycling | Energy-balanced multi-hop routing | Linear sensor arrangements | Data aggregation |
| Battery Lifetime | Duty cycle, harvesting, routing | Deployment viability assessment | High data rate applications | Coverage redundancy |
Hexagonal deployment achieves optimal coverage efficiency through geometric principles. Here’s why it works:
Step 1: Coverage Circle Packing
Step 2: Geometric Spacing Calculation
Step 3: Node Count Comparison
Note: the 23% figure commonly cited assumes slightly different boundary conditions and rounding.
Step 4: Deployment Implementation
Real-World Example: 100-hectare agricultural field
Objective: Design sensor deployment for a 5km × 2km highway monitoring system.
Scenario:
Tasks:
What to Observe:
Extension: Design gateway placement to minimize maximum hop count: - Try 1 gateway, 2 gateways (ends), 3 gateways (0m, 2500m, 5000m) - Calculate worst-case hops for each configuration - Which provides best latency-vs-cost trade-off?
WSN Deployment Topics:
Related Deployment Strategies:
Energy Management:
Average current calculation ignores burst transmissions — a node averaging 100 µA may draw 30 mA during transmissions, and if coin cell internal resistance is high, voltage sag during bursts causes premature brownout resets. Always measure peak current and verify voltage stays above minimum at the worst-case load.
RF optimization efforts focus on radio (dominant source) but ignore sensor power — a continuous soil moisture sensor drawing 3 mA consumes more than the radio in many duty-cycled designs. Characterize every subsystem’s power consumption before optimizing; the highest consumer is not always the radio.
For 5-year battery life targets, quiescent leakage of voltage regulators (1-100 µA) becomes significant — a 100 µA leakage on a 2,500 mAh battery limits life to 2.85 years regardless of how well you optimize the active duty cycle. Select ultra-low-quiescent regulators (<1 µA) for multi-year deployments.
This chapter covered WSN deployment planning and energy management:
| Topic | Chapter | Description |
|---|---|---|
| Routing and Monitoring | WSN Implementation: Routing and Monitoring | Protocol selection, decision matrices, and network health monitoring |
| Architecture Design | WSN Implementation: Architecture and Topology | Multi-tier system design, LEACH clustering, and hardware selection |
| Energy Management | WSN Fundamentals: Energy Management | Duty cycling protocols and energy harvesting techniques |