IoT & smart-city data: telemetry, air quality and occupancy

Cities and connected infrastructure now generate continuous streams of sensor data. Used well, that data improves air quality, mobility, energy use and public services. This guide explains the IoT and smart-city data landscape, the connectivity and standards that shape it, and how to source and deliver telemetry reliably and lawfully.

Why IoT data matters now

Sensors are cheap, connectivity is ubiquitous, and the EU Data Act now gives users a clearer right to access data generated by connected products. The result is a fast-growing supply of telemetry, from air-quality monitors to occupancy counters, that can be combined with other layers to run services and inform decisions. The challenge is less about scarcity and more about consistency, quality and lawful use.

The IoT & smart-city data landscape

• Environmental sensors: air quality, noise and weather microclimate.
• Mobility and parking: parking occupancy, traffic sensors and bike or micromobility usage.
• Buildings and energy: occupancy, building energy use and street-lighting consumption.
• Municipal operations: waste-bin fill levels, asset tracking and connected-infrastructure status.
• Device health: sensor uptime and calibration metadata that underpins data quality.

Connectivity and standards

IoT data arrives over a range of transports, with lightweight messaging protocols such as MQTT common for telemetry. Smart-city interoperability efforts (for example, NGSI-LD context information and OGC standards for sensors) aim to make data from different vendors comparable. For a buyer, the practical questions are how a feed is delivered, how it is structured, and whether device metadata is available to judge quality.

Common use cases

Applications include air-quality and environmental monitoring, smart parking and mobility management, building and street-lighting energy optimisation, waste-collection routing, and predictive maintenance of connected assets. Many of these need continuous or near-real-time data rather than periodic extracts.

Sourcing considerations specific to IoT

Telemetry quality varies with sensor calibration, placement and uptime, so device-health metadata matters as much as the readings. Standards adoption is uneven, so harmonising units, timestamps and identifiers is often necessary. And some IoT data is personal (for example, where occupancy or movement can identify individuals), bringing the GDPR into play alongside the EU Data Act.

Delivery and cadence

Live operational use cases favour streaming or MQTT and near-real-time delivery; analytics and planning use scheduled batches in formats such as Parquet or CSV. Because volumes can be very high, sourcing should consider aggregation and edge pre-processing to deliver only what the use case needs.

Governance and privacy

Where sensor data can identify people or premises, anonymisation or aggregation is typically required, and the GDPR applies. The EU Data Act shapes access rights to connected-product data. Provenance, licensing and security practices aligned with NIS2 and ISO/IEC 27001 principles support critical-infrastructure and public-sector use.

Protocols and payloads

IoT data reaches you over a handful of transports, and each shapes what you get. MQTT is the dominant lightweight publish/subscribe protocol for telemetry, efficient over constrained networks. HTTP/REST APIs suit request-driven access and integration. CoAP and LwM2M appear in constrained-device contexts. Payloads vary from compact binary to JSON, and standards such as NGSI-LD aim to make context data interoperable across vendors. For a buyer, the practical questions are how the feed is delivered, how the payload is structured, and whether you receive the metadata needed to interpret it.

Edge, cloud and aggregation

IoT volumes can be enormous, and not all of it is worth moving. Increasingly, pre-processing happens at the edge, close to the sensor, aggregating, filtering and summarising before transmission, so only what the use case needs travels to the cloud. When sourcing IoT data, decide the right granularity deliberately: raw per-reading data for detailed analytics, aggregated rollups for dashboards and trends. Over-collecting raw telemetry inflates cost and, where it touches individuals, privacy risk.

An IoT sourcing checklist

• What transport and payload format does the feed use, and is it documented?
• Is device-health metadata (calibration, uptime) available to judge quality?
• What granularity do you actually need, raw or aggregated?
• Could the data identify people or premises, requiring aggregation under the GDPR?
• How does the EU Data Act affect access to this connected-product data?
• What latency, and stream or batch, fits the use case?

Sources & further reading

• European Commission: The Data Act (Regulation (EU) 2023/2854), connected-product data.
• European Environment Agency: air-quality data and indicators.
• OGC and ETSI (NGSI-LD): smart-city and sensor interoperability standards.
• EUR-Lex: Regulation (EU) 2016/679 (GDPR).