QoS levels (0, 1, 2) in IOT Protocols - Deep Dive

Overview - QoS levels (0, 1, 2)

What is it?

QoS levels define how messages are delivered between devices in IoT communication. They control the guarantee of message delivery and how many times a message is sent. There are three levels: 0 means 'at most once', 1 means 'at least once', and 2 means 'exactly once'. These levels help balance reliability and network efficiency.

Why it matters

Without QoS levels, messages could be lost or duplicated without control, causing unreliable device communication. This could lead to wrong data, missed commands, or wasted network resources. QoS ensures devices can trust the messages they receive, which is critical for automation, monitoring, and safety in IoT systems.

Where it fits

Learners should first understand basic IoT communication and messaging concepts. After mastering QoS levels, they can explore advanced IoT protocols, message brokers, and network optimization techniques.

Mental Model

Core Idea

QoS levels control how many times and how reliably a message is sent between devices to balance delivery assurance and network load.

Think of it like...

Imagine sending a letter: QoS 0 is like dropping it in the mailbox once without tracking; QoS 1 is like sending it with a receipt confirmation but possibly sending duplicates; QoS 2 is like sending it with a signed delivery confirmation ensuring it arrives exactly once.

┌───────────────┐
│   QoS Levels  │
├───────────────┤
│ 0: At most once│───► Send once, no retry
│ 1: At least once│──► Send until ack, possible duplicates
│ 2: Exactly once│───► Four-step handshake, no duplicates
└───────────────┘

Build-Up - 7 Steps

1

Foundation

Understanding Message Delivery Basics

Concept: Introduce the idea that messages can be lost or duplicated in networks.

In IoT, devices send messages over networks that can be unreliable. Sometimes messages get lost or arrive multiple times. Understanding this helps us see why controlling delivery is important.

Result

Learners grasp that message delivery is not guaranteed by default.

Knowing that networks are imperfect sets the stage for why QoS levels exist.

2

Foundation

Introducing QoS Levels Concept

Concept: Explain that QoS levels are rules to manage message delivery reliability.

QoS levels tell devices how to handle sending messages: whether to send once, retry until confirmed, or ensure no duplicates. This helps balance reliability and network traffic.

Result

Learners understand QoS as a tool to control message delivery behavior.

Recognizing QoS as a control mechanism helps learners appreciate its role in IoT communication.

3

Intermediate

QoS 0: At Most Once Delivery

🤔Before reading on: do you think QoS 0 guarantees message delivery or not? Commit to your answer.

Concept: Explain QoS 0 where messages are sent once without confirmation.

QoS 0 means the message is sent once and not retried. If lost, it won't be resent. This is fastest but least reliable.

Result

Messages may be lost but never duplicated.

Understanding QoS 0 shows the tradeoff of speed over reliability.

4

Intermediate

QoS 1: At Least Once Delivery

🤔Before reading on: does QoS 1 guarantee no duplicates or possible duplicates? Commit to your answer.

Concept: Explain QoS 1 where messages are sent until acknowledged, allowing duplicates.

QoS 1 sends a message and waits for an acknowledgment. If no ack, it resends. This ensures delivery but can cause duplicates.

Result

Messages are delivered at least once, duplicates possible.

Knowing QoS 1 balances reliability with the risk of duplicates helps in choosing it for critical but tolerant systems.

5

Intermediate

QoS 2: Exactly Once Delivery

🤔Before reading on: do you think QoS 2 uses extra steps to avoid duplicates? Commit to your answer.

Concept: Explain QoS 2 with a four-step handshake to ensure exactly one delivery.

QoS 2 uses a handshake process: sender sends, receiver acknowledges, sender confirms, receiver finalizes. This avoids duplicates but adds overhead.

Result

Messages are delivered exactly once, no duplicates, highest reliability.

Understanding QoS 2 reveals how extra communication ensures perfect delivery at a cost.

6

Advanced

Choosing QoS Levels in Real Systems

🤔Before reading on: would you use QoS 2 for all messages or only some? Commit to your answer.

Concept: Discuss how to select QoS levels based on application needs and network conditions.

Use QoS 0 for non-critical data to save bandwidth, QoS 1 for important data where duplicates are okay, and QoS 2 for critical commands needing perfect delivery. Consider network load and device power.

Result

Learners can make informed decisions on QoS use in IoT projects.

Knowing when to use each QoS level optimizes reliability and resource use.

7

Expert

Internal Protocol Handling of QoS 2

🤔Before reading on: do you think QoS 2's handshake can cause message delays? Commit to your answer.

Concept: Explore the internal message flow and state tracking for QoS 2 in MQTT protocol.

QoS 2 requires sender and receiver to track message states with packet IDs and flags. This prevents duplicates but adds latency and memory use. Handling failures in handshake steps is complex.

Result

Learners understand the tradeoffs and complexity behind QoS 2 implementation.

Understanding internal mechanics of QoS 2 helps troubleshoot and optimize IoT messaging.

Under the Hood

QoS levels work by controlling message sending and acknowledgment patterns. QoS 0 sends once without waiting. QoS 1 sends and waits for an acknowledgment, resending if needed. QoS 2 uses a four-step handshake with unique packet IDs to ensure exactly one delivery, tracking message states on both ends.

Why designed this way?

These levels were designed to balance reliability and network efficiency. Early IoT devices had limited power and bandwidth, so a simple 'send once' was fastest. But critical systems needed guaranteed delivery, so more complex handshakes were added. The design reflects tradeoffs between speed, reliability, and resource use.

Sender                      Receiver
  │                            │
  │ QoS 0: Send message once   │
  │──────────────────────────▶│
  │ (No ack, no retry)         │

  │                            │
  │ QoS 1: Send message        │
  │──────────────────────────▶│
  │                            │
  │          Ack message       │
  │◀──────────────────────────│
  │ (Resend if no ack)         │

  │                            │
  │ QoS 2: Send PUBLISH        │
  │──────────────────────────▶│
  │          PUBREC            │
  │◀──────────────────────────│
  │          PUBREL            │
  │──────────────────────────▶│
  │          PUBCOMP           │
  │◀──────────────────────────│

Myth Busters - 4 Common Misconceptions

Quick: Does QoS 0 guarantee message delivery? Commit yes or no.

Common Belief: QoS 0 guarantees the message will arrive because it is sent.

Reality: QoS 0 does not guarantee delivery; the message may be lost without retry.

Why it matters: Assuming QoS 0 is reliable can cause missing critical data or commands.

Quick: Does QoS 1 prevent duplicate messages? Commit yes or no.

Common Belief: QoS 1 ensures no duplicate messages are received.

Reality: QoS 1 can deliver duplicates because it resends until acknowledged.

Why it matters: Ignoring duplicates can cause repeated actions or data errors in applications.

Quick: Is QoS 2 always the best choice for all messages? Commit yes or no.

Common Belief: QoS 2 should be used for all messages to ensure perfect delivery.

Reality: QoS 2 adds overhead and latency, so it is not efficient for all messages.

Why it matters: Using QoS 2 unnecessarily wastes bandwidth and device resources.

Quick: Does QoS 2 guarantee zero message delay? Commit yes or no.

Common Belief: QoS 2 delivers messages instantly without delay.

Reality: QoS 2's handshake adds delay due to multiple message exchanges.

Why it matters: Expecting instant delivery with QoS 2 can lead to wrong assumptions in time-sensitive systems.

Expert Zone

1

QoS 2 requires both sender and receiver to maintain state for each message, increasing memory use on constrained devices.

2

Network interruptions during QoS 2 handshake can cause message duplication or delays if not handled carefully.

3

Some MQTT brokers optimize QoS 1 and 2 flows internally to reduce overhead, which can affect message timing.

When NOT to use

Avoid QoS 2 for high-frequency telemetry data where duplicates are acceptable; use QoS 0 or 1 instead. For very low-power devices, QoS 0 reduces battery use. Use alternatives like UDP for non-critical messages where even QoS 0 is too heavy.

Production Patterns

In production, critical commands use QoS 2 to ensure no duplicates or loss. Sensor data often uses QoS 0 or 1 to save bandwidth. Systems monitor acknowledgment failures to trigger alerts or retries. Brokers log QoS flows for auditing and debugging.

Connections

TCP vs UDP Protocols

QoS levels in MQTT are similar to TCP's reliable delivery vs UDP's best effort.

Understanding TCP and UDP helps grasp why QoS levels balance reliability and speed.

Database Transaction Isolation Levels

QoS 2's exactly-once delivery is like serializable isolation ensuring no duplicates or lost updates.

Knowing transaction isolation clarifies how QoS 2 prevents duplicate message effects.

Postal Mail Delivery

QoS levels mirror postal options: standard mail (QoS 0), registered mail (QoS 1), and certified mail with signature (QoS 2).

Real-world mail services show why different delivery guarantees exist and their costs.

Common Pitfalls

#1 Assuming QoS 0 messages always arrive.

Wrong approach:

Publish message with QoS 0 and ignore missing data errors.

Correct approach:

Use QoS 1 or 2 for critical messages needing guaranteed delivery.

Root cause:

Misunderstanding that QoS 0 does not retry lost messages.

#2 Ignoring duplicate messages from QoS 1.

Wrong approach:

Process every received message without checking for duplicates.

Correct approach:

Implement idempotent processing or duplicate detection for QoS 1 messages.

Root cause:

Belief that QoS 1 prevents duplicates.

#3 Using QoS 2 for all messages causing network overload.

Wrong approach:

Set all messages to QoS 2 regardless of importance.

Correct approach:

Select QoS level based on message criticality and resource constraints.

Root cause:

Assuming highest QoS is always best.

Key Takeaways

QoS levels control message delivery reliability and duplication in IoT communication.

QoS 0 sends messages once without guarantee, suitable for non-critical data.

QoS 1 ensures delivery but may cause duplicates, requiring careful handling.

QoS 2 guarantees exactly one delivery using a handshake, adding overhead and delay.

Choosing the right QoS level balances reliability, network load, and device resources.