3 Edge Acquisition: Control Paths and Bandwidth

3.1 Start With the Decision

A cloud command and an edge alarm do not share the same safe delay. Map both paths before deciding where data and control should live.

3.2 Route Overview

This is part 2 of 3. Review Edge Acquisition: Device Architecture for the preceding evidence.

3.3 Learning Objectives

• Trace cloud and edge control paths with failure states.
• Calculate bandwidth saved by local aggregation.

3.4 Chapter Roadmap

• Cloud and Edge Control Paths
• Checkpoint: Device Categories
• Data Generation Patterns
• Mobile figure summary
• Mobile comparison summary
• Data Volume by Device Type
• Inertial Measurement Example
• Mobile figure summary
• Putting Numbers to It
• Edge Bandwidth Calculator

3.5 Cloud and Edge Control Paths

Before assigning computation, inspect Figure 3.1 to see why a cold-store control loop can become dependent on its wide-area path.

Read TEMP SENSOR through CLOUD RULE to ACTUATOR in Figure 3.1. The DESIGN INFERENCE explains that central history is simple, but a WAN failure interrupts the decision unless a safe fallback exists.

Inspect Figure 3.2 to compare the same threshold when it is evaluated before the event leaves the site.

Read Figure 3.2 from EDGE GATEWAY through LOCAL ACTUATOR, while EVENT RECORD carries rule and quality evidence to CLOUD ANALYTICS. That split improves continuity but makes gateway update and replay ownership explicit.

Inspect Figure 3.3 next to classify events before choosing local or remote processing and to expose the evidence each placement requires.

Trace RFID and SPEED into AUTHENTICATE and NORMALISE in Figure 3.3, then compare LOCAL RESPONSE with REMOTE PATH. The latency chips and continuity ledger show why urgency, sensitivity, and replay needs—not data volume alone—set placement.

Finally, inspect Figure 3.4 to see how the same responsibilities scale across multiple physical zones without erasing their failure boundaries.

Read Figure 3.4 across ZONE A · ORCHARD and ZONE C · HOUSE, which retain bounded gateway failure domains while FARM PLATFORM coordinates irrigation plans and approvals. command + expiry and audit + outcome keep control and evidence paired as the estate grows.

3.6 Data Generation Patterns

Understanding data generation patterns is essential for designing efficient edge acquisition systems. Different device types produce vastly different data volumes and require different handling strategies. Begin with the device-class diagram in Figure 3.5, then use Figure 3.6 to isolate the effect of sampling frequency.

Read Figure 3.5 across each device row. Compare typical volume with transmission rhythm and connectivity: continuous, high-volume sources place a different burden on the edge than sparse event-driven sensors, even before fleet size is applied. Then compare rows vertically to see why direct IP connectivity does not automatically make raw upload economical. This carries the category model into the running acquisition narrative by turning device capability into a bandwidth, storage, and processing obligation.

The sampling-rate comparison diagram in Figure 3.6 holds record size constant so you can see the cost of changing only the acquisition interval.

Read Figure 3.6 down the sampling-rate column and compare the daily totals. Each tenfold increase in frequency produces ten times as many 20-byte records, so the visual growth is arithmetic rather than a property of a particular sensor. Carry that daily number forward to the fleet and retention period before choosing a link. This connects sample timing to the chapter’s running design decision: faster acquisition must be justified by signal bandwidth or decision latency, then paid for with storage, radio energy, or local reduction.

Read Figure 3.5 across each device row: typical data rate, transmission rhythm, and connectivity together determine whether raw forwarding is plausible. Then read Figure 3.6 down the sampling-rate column while holding the 20-byte record size constant. Each tenfold increase in sample frequency produces a tenfold daily volume, so a modest single-node stream becomes a fleet storage and radio problem. The two comparisons advance the running narrative from device category to a measurable acquisition budget that can justify filtering, aggregation, or a different link.

Data Volume by Device Type

This view shows how data generation rates vary dramatically by device type, driving different edge processing strategies. Inspect the comparison diagram in Figure 3.7 to locate the largest raw stream before selecting an edge reduction method.

Read Figure 3.7 from low-volume Non-IP sensors toward high-volume IP devices such as cameras. The widening orders-of-magnitude gap means one forwarding policy cannot fit every source: sparse events may travel intact, periodic measurements may be batched, and continuous media often needs local selection or inference. The comparison connects the chapter’s device categories to its running budget argument: preserve raw evidence only where the downstream decision needs it, and measure the bandwidth, storage, energy, and privacy cost of that choice.

3.9 Inertial Measurement Example

High-frequency sensors like accelerometers and gyroscopes demonstrate why edge aggregation is critical. Inspect the paired traces in Figure 3.8 before reducing them so the retained summary remains tied to the motion evidence.

Read Figure 3.8 from the three accelerometer axes in the upper panel to the three gyroscope axes below, then move left to right through the shared time interval. Accelerometer changes combine motion and gravity; gyroscope changes record rotation. A motion event can therefore span several axes and both sensor types, which is why a single average may erase useful structure. This connects the example to the running volume argument: edge windows should retain the features required by the decision while replacing a continuous six-axis stream with an auditable summary.

At 100 Hz sampling across 6 axes (3 accelerometer + 3 gyroscope), an IMU generates 600 samples/second. Transmitting raw data as 16-bit integers would require ~1.2 KB/s — unsustainable for battery-powered devices on LPWAN networks. Edge aggregation reduces this to statistical summaries at 1 Hz.

3.11 Edge Bandwidth Calculator

Use the sliders below to explore how sampling rate, number of axes, and aggregation window affect raw versus aggregated data rates. Observe how quickly raw data exceeds LPWAN capacity.

3.12 Continue to the Next Part

Carry this evidence into Edge Acquisition: Power Budgets and Pipelines, which begins with Power Budget Decision Framework.