17  Topology Comparison

Key Concepts
  • Topology Fundamentals Review: A structured consolidation of star, bus, ring, mesh definitions, properties, and trade-offs before assessment
  • Connection Count Formulas: Star n-1, ring n, full mesh n(n-1)/2 — the three formulas most frequently tested in assessments
  • Fault Tolerance Ranking: Bus < Ring < Star < Partial Mesh < Full Mesh in order of increasing resilience
  • Logical vs Physical Topology Distinction: A key conceptual distinction: physical describes cable/radio layout, logical describes data flow paths
  • IoT Protocol-Topology Mapping: Which protocols naturally implement which topologies (LoRaWAN→star, Zigbee→mesh, BACnet MS/TP→ring)
  • Cost-Resilience Trade-off: The fundamental tension between minimising connections (lowering cost) and maximising redundancy (raising resilience)
  • Prerequisite Knowledge Check: Verifying that fundamental concepts are solid before attempting higher-level analysis topics

17.1 In 60 Seconds

This chapter compares star, mesh, tree, and ring topologies side by side using quantified metrics: link counts, redundancy ratios, and network diameter. You will learn that more mesh links do not always improve performance, and that topology selection depends on balancing fault tolerance, scalability, and cost for your specific IoT application requirements.

This chapter compares different network topologies to help you make informed design decisions:

  • We analyze star, mesh, tree, and ring topologies side by side.
  • You will learn to compute link counts, redundancy, and diameter for different designs.
  • We connect these concepts to IoT performance (latency, reliability, cost).

If you need to review basic topology concepts first, see Topologies Fundamentals.

17.2 Learning Objectives

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

  • Compare Topologies Quantitatively: Assess star, tree, mesh, and hybrid topologies using link count, redundancy, and diameter metrics
  • Evaluate Trade-offs: Weigh fault tolerance, scalability, and cost for each topology in specific deployment contexts
  • Debunk Common Misconceptions: Explain why adding more mesh links beyond 60% connectivity degrades performance
  • Justify Topology Selection: Defend a topology choice based on measurable IoT application requirements

17.3 Prerequisites

Required Chapters:

Technical Background:

  • Basic understanding of nodes and links
  • Network redundancy concepts

Estimated Time: 20 minutes

17.4 Topology Comparison Overview

Time: ~15 min | Level: Intermediate | Unit: P07.C26.U01

Understanding how different topologies compare is essential for IoT network design. The figure below shows a structural comparison:

Comparison of four network topologies showing structural differences. Star topology has navy central hub with four orange devices connected radially, demonstrating single point of failure. Mesh topology shows five teal devices interconnected with seven links providing multiple redundant paths. Tree topology displays hierarchical structure with navy root connecting to two teal routers, each supporting two orange end devices, showing layered organization. Ring topology presents five gray devices connected in circular pattern where each device connects to exactly two neighbors, forming closed loop. IEEE color palette emphasizes role differences: navy for critical nodes, teal for routing capable devices, orange for end devices, gray for equal-role nodes.
Figure 17.1: Comparison of Star, Mesh, Tree, and Ring Network Topologies

This variant presents topology selection as a decision process based on application requirements:

Decision tree flowchart for IoT topology selection. Starting from top, first decision node asks whether high reliability is required. Yes branch leads to second node asking whether power budget is constrained. If power-constrained, tree topology is recommended; otherwise mesh topology is recommended. No branch from reliability question leads to node asking whether network is small-scale. Small-scale leads to star recommendation; larger scale leads to ring or hybrid recommendation. IEEE color palette with navy decision nodes and teal outcome boxes.

This decision tree helps you select the right topology by prioritizing reliability requirements first, then optimizing for power or simplicity.

This variant compares topologies using quantitative metrics for n=10 nodes:

Side-by-side metrics comparison table rendered as a visual chart for star, mesh, tree, and ring topologies with n equals 10 nodes. Columns represent each topology; rows show link count, redundancy ratio, network diameter, and cost index. Color coding uses green for favorable metrics and orange for trade-offs. Star shows 10 links and diameter 2 in green but low redundancy in orange. Mesh shows 45 links and high redundancy in green but diameter and cost disadvantages in orange. Tree shows 9 links and low cost in green but medium redundancy in orange. Ring shows 10 links in green but high diameter in orange.

Green indicates favorable metrics, orange indicates trade-offs. Notice how each topology excels in some areas but compromises in others.

Key Trade-offs:

Aspect Star Mesh Tree Ring
Fault Tolerance Low (hub SPOF) High (self-healing) Medium (subtree isolation) Medium (link break recovers via reverse path)
Scalability Good (hub limited) Excellent (distributed) Excellent (hierarchical) Limited (sequential)
Latency Low (max 2 hops) Variable (multi-hop) Low (few levels) High (sequential)
Installation Easy (to hub) Complex (routing) Medium (hierarchy) Medium (ring closure)
Cost Low (minimal links) High (many links) Medium (backbone) Low (simple loop)
IoT Use Smart home Smart city Building automation Industrial control
Cross-Hub Connections: Test Your Topology Knowledge

Ready to validate your understanding? Connect with these learning resources:

Interactive Learning:

Video Resources:

  • Videos Hub - Watch topology visualization animations and real-world deployment examples
  • Search for “IoT Network Topologies” and “Mesh vs Star Comparison”

Knowledge Gaps:

  • Knowledge Gaps Hub - Common topology misconceptions and troubleshooting patterns
  • Learn why physical and logical topologies differ
  • Understand when mesh overhead outweighs benefits

Knowledge Map:

  • Knowledge Map - See how topology concepts connect to protocols, routing, and architecture

Pro Tip: Before diving into the review questions, try the Topology Visualizer to see how diameter, redundancy, and connectivity ratios change as you add/remove links!

17.5 Common Misconceptions

Common Misconception: “More Mesh Links Always Improve Network Performance”

The Misconception: Many engineers assume that increasing mesh connectivity (adding more links between nodes) automatically improves network performance and reliability.

The Reality: Beyond 60% connectivity ratio, additional mesh links provide diminishing returns and can actually degrade performance due to routing overhead.

Real-World Data:

A 2019 study of Zigbee mesh networks in smart buildings revealed:

  • 30% connectivity (partial mesh): 3.2 neighbor connections per node, 98.5% packet delivery, 85ms average latency, 12-month battery life
  • 60% connectivity (dense mesh): 5.4 neighbors per node, 99.1% packet delivery, 92ms latency, 9-month battery life (25% reduction)
  • 90% connectivity (near-full mesh): 8.1 neighbors per node, 98.8% packet delivery, 127ms latency, 6-month battery life (50% reduction)
  • 100% connectivity (full mesh): 9.0 neighbors per node, 98.2% packet delivery, 156ms latency, 5-month battery life (58% reduction)

Why This Happens:

  1. Routing Protocol Overhead: Each node must maintain and update routing tables for all neighbors. 9 neighbors requires 3x memory and 3x control message traffic compared to 3 neighbors.

  2. Path Selection Complexity: With 8 alternate paths between nodes, routing algorithm takes longer to compute optimal path (50-150ms vs 10-30ms).

  3. Broadcast Amplification: Network-wide announcements flood through all links. 90% mesh creates 3x broadcast traffic vs 30% mesh.

  4. Hidden Terminal Problem: Dense mesh increases radio interference. More neighbors transmitting simultaneously = more collisions = retransmissions.

The Sweet Spot:

Industry best practice: 40-60% connectivity ratio balances redundancy against overhead:

  • Formula: For n nodes with full mesh having \(\frac{n(n-1)}{2}\) links, deploy \(0.4 \times \frac{n(n-1)}{2}\) to \(0.6 \times \frac{n(n-1)}{2}\) links
  • Example: 20 nodes, full mesh = 190 links, optimal = 76-114 links (each node connects to 4-6 neighbors)
  • Result: Multiple redundant paths without excessive routing overhead

Calculating Mesh Connectivity Ratio for IoT Deployments

For a 20-node Zigbee sensor network, let’s quantify the trade-off between full mesh and partial mesh:

Full mesh link count: \[ L_{\text{full}} = \frac{n(n-1)}{2} = \frac{20 \times 19}{2} = 190 \text{ links} \]

40% partial mesh (recommended minimum): \[ L_{40\%} = 0.4 \times 190 = 76 \text{ links} \] \[ \text{Average degree} = \frac{2 \times 76}{20} = 7.6 \text{ neighbors per node} \]

60% partial mesh (maximum before overhead dominates): \[ L_{60\%} = 0.6 \times 190 = 114 \text{ links} \] \[ \text{Average degree} = \frac{2 \times 114}{20} = 11.4 \text{ neighbors per node} \]

Battery life impact (empirical data from Zigbee networks): \[ \text{Battery life} \propto \frac{1}{\text{routing overhead}} \approx \frac{1}{\text{neighbor count}^{1.2}} \]

(This relationship is illustrative; actual exponent varies by protocol and hardware.)

At 40% connectivity (7.6 neighbors): ~10 months battery At 60% connectivity (11.4 neighbors): ~8 months battery At 100% connectivity (19 neighbors): ~5 months battery

**Cost savings (at \(100 per configured link):**\)$ _{40%} = (190 - 76) $100 = $11,400 \text{ (60% cost reduction)} $$

Try It: Mesh Connectivity Calculator

Design Rule: Start with 40% connectivity. Only increase to 60% for mission-critical networks where 99.5%+ reliability justifies battery life trade-off. Never deploy full mesh (100%) for battery-powered IoT.

Case Study: Smart factory with 50 vibration sensors. Initial design used 90% mesh (1102 links) achieving 98.9% uptime but requiring battery replacement every 4 months ($15,000/year labor cost). Redesign to 50% mesh (612 links) achieved 99.2% uptime with 10-month battery life ($6,000/year labor cost). Savings: $9,000/year + 0.3% uptime improvement.

17.6 Understanding Check

Scenario: You’re designing a smart building network with 10 critical environmental control nodes. Full mesh would require 45 links (n(n-1)/2 formula), but you’ve deployed 18 links strategically.

Think about:

  1. How does 40% connectivity (18/45 links) balance redundancy against deployment cost?
  2. What failure scenarios can your partial mesh handle versus full mesh?

Key Insight: Partial mesh with 40% connectivity provides multiple paths between most nodes while saving 60% in cabling/configuration costs. For IoT deployments, 30-60% connectivity typically delivers adequate fault tolerance: each node has 3-4 neighbors for redundancy without full mesh complexity.

Verify Your Understanding:

  • How would you determine the minimum connectivity ratio for your application’s reliability needs?
  • What happens to network resilience if connectivity drops below 30%?
Show Answer

Answer: 40% connectivity ratio

Calculation:

num_nodes = 10
num_links = 18

# Full mesh formula: n(n-1)/2
max_links = num_nodes * (num_nodes - 1) // 2  # = 45

# Connectivity ratio
connectivity = num_links / max_links  # = 18/45 = 0.4 = 40%

Interpretation:

  • 40% connectivity = Partial mesh with moderate redundancy
  • < 30% = Sparse mesh
  • 30-60% = Good balance (typical for IoT)
  • 80% = Near full mesh

  • 100% = Full mesh

Practical implications:

  • 18 links provide multiple paths between nodes
  • Much cheaper than 45-link full mesh (60% cost savings)
  • Adequate redundancy for most IoT applications

17.7 Worked Example: Choosing a Topology for a Multi-Building Campus

Scenario: A university is deploying 600 environmental sensors across 5 buildings on a campus (each building 200m apart). Each building has 3 floors with 40 sensors per floor. The sensors report temperature, humidity, and CO2 every 60 seconds. The university requires 99% uptime, 3-year battery life on AA cells, and a total infrastructure budget of $50,000. Which topology should connect the sensors?

Step 1: Evaluate Topology Options

Topology Intra-Building Links Inter-Building Links Total Infrastructure Failure Mode
Star (1 hub/floor) 15 hubs, 600 radial links 5 gateways (one per building) 15 hubs + 5 gateways = 20 devices Hub failure loses 40 sensors (1 floor)
Full Mesh (all-to-all) 600 x 599 / 2 = 179,700 links Impractical Not feasible Extreme routing overhead, 2-month battery
Partial Mesh (50%) ~90,000 links Feasible but expensive Too many routing entries for 64KB RAM sensors Complex, power-hungry
Tree (hierarchical) 15 floor routers + 5 building gateways 1 campus root 21 infrastructure nodes Subtree isolation on router failure

Step 2: Calculate Battery Impact

For a sensor sending 1 reading/minute (1,440/day):

Topology Neighbor Count Control Msgs/Day Radio-On Time/Day AA Battery Life
Star 1 (hub only) ~50 (keepalive) 45 seconds 4.2 years
Partial Mesh (50%) ~12 neighbors ~2,400 (link-state) 18 minutes 0.4 years
Tree 1 parent + 0-3 children ~200 (RPL DIO/DAO) 2 minutes 3.1 years

Star and tree meet the 3-year battery target. Partial mesh fails dramatically.

Step 3: Cost Comparison

Topology Hub/Router Hardware Gateway Hardware Cabling/Install Total
Star 15 hubs x $200 = $3,000 5 gateways x $500 = $2,500 Ethernet backbone: $4,000 $9,500
Tree (RPL) 15 floor routers x $80 = $1,200 5 building roots x $300 = $1,500 No cabling (wireless mesh) $2,700

Both fit the $50,000 budget easily. Tree is 72% cheaper.

Step 4: Reliability Analysis

Failure Scenario Star Impact Tree Impact
Single sensor fails 1 sensor lost 1 sensor lost
Floor hub/router fails 40 sensors lost (1 floor offline) All 40 sensors temporarily offline; RPL reroutes most within 30-120s; 8-12 may be permanently unreachable if too far from alternate routers
Building gateway fails 120 sensors lost (entire building offline) 120 sensors offline (same – gateway is root)
Recovery time Manual hub replacement (hours) Auto-reroute via RPL (30-120 seconds)

Tree topology is more resilient to mid-level failures because RPL’s DODAG enables automatic rerouting.

Decision: Tree topology with RPL routing. It meets all three requirements: 3.1-year battery life (exceeds 3-year target), $2,700 infrastructure cost (well under $50,000), and 99%+ uptime through RPL self-healing. The tree’s hierarchical structure naturally maps to the campus-building-floor hierarchy, and RPL’s energy-aware routing minimizes battery drain.

Try It: Topology Selection Calculator

Adjust the deployment parameters to see how the topology recommendation changes.

Common Pitfalls

Bus and ring both use a shared structure but differ fundamentally in how data flows. Fix: draw each topology type from memory and annotate data flow direction before the assessment.

Pure star, bus, ring, and mesh are textbook examples. Real IoT deployments use cluster-tree, dual-ring, or tiered star. Fix: for each basic topology, name one real-world protocol that implements a variant of it.

Reading the definitions is passive learning. Fix: close the notes and sketch each topology with n=6 nodes, label the connections, and compute the connection count from the formula.

17.9 Summary

  • Network topology selection requires balancing fault tolerance, scalability, latency, and cost
  • Star topology offers simplicity and low latency but has a single point of failure at the hub
  • Mesh topology provides self-healing and range extension but adds routing overhead and complexity
  • Tree topology scales well hierarchically and enables traffic aggregation at intermediate nodes
  • Ring topology offers simple wiring and predictable hop counts but has limited scalability and higher worst-case latency
  • Connectivity ratio of 40-60% is optimal for mesh networks, balancing redundancy against overhead
  • Hybrid topologies often provide the best cost-performance balance for large deployments

17.10 Knowledge Check

17.11 What’s Next

If you want to… Read this
Review topology analysis metrics Review Analysis
Practise with review scenarios Review Scenarios
Study the comprehensive review Comprehensive Review
Revisit the original topology types chapter Basic Types