Smart metering is a technology for remote, digital measurement of electricity, gas, heat and water consumption using connected meters that transmit data to billing and analytics systems in near real time. For network operators and utilities, smart metering means the end of manual readings, more accurate billing and new demand-management capabilities. In this guide we show how to design a reliable smart metering system in an IoT architecture: from the meter layer, through the network and protocols, to the cloud and security.
In short: smart metering is an IoT system in which AMI meters transmit interval consumption data (every 15-60 minutes) over networks such as NB-IoT or LoRaWAN, using DLMS/COSEM protocols, while a cloud platform handles billing, analytics and control. A well-designed battery-powered meter runs for 10-20 years.
What is smart metering?
Smart metering is a two-way measurement system in which the meter not only records consumption but also communicates with the operator, receives commands and reports events. Unlike a classic induction meter, a smart meter contains a microcontroller, a communication modem, a real-time clock and load-profile memory.
The core of deployments is the AMI (Advanced Metering Infrastructure) architecture, which connects meters, data concentrators and a management platform. AMI enables interval readings, remote power connection and disconnection, time-of-use tariffs and the detection of irregularities such as tampering attempts or voltage outages.
The business benefits are measurable. Remote reading eliminates meter-reader costs and manual-entry errors, shortens the billing cycle from months to days, and 15-minute profiles allow the detection of non-technical losses reaching 5-15 percent of volume in some networks. EU regulations (including the EED directive) require operators to make consumption data available to consumers, which further drives smart metering deployments in Poland and Europe.
What layers make up a smart metering system?
A smart metering system has four layers: the meter (edge device), the communication network, the concentration layer and the cloud. Each of them requires separate design decisions regarding hardware, firmware and integration.
- Meter layer – metrology, a microcontroller (e.g. STM32 or ESP32), a modem and a secure cryptographic element.
- Network layer – long-range, low-power radio (LPWAN) or cellular communication.
- Concentration layer – gateways and concentrators aggregating data from hundreds of meters.
- Cloud layer – billing, analytics, fleet management and APIs.
When designing the edge device, the energy budget is critical. The principles of designing battery-powered IoT devices for multi-year operation translate directly into the cost of running a meter fleet.
Which protocols are used in smart metering?
DLMS/COSEM is the dominant, standardized (IEC 62056) data model and application protocol for energy meters, ensuring interoperability between vendors. Alongside it, Wireless M-Bus (EN 13757) is used in water and heat metering, together with lightweight IoT transport protocols.
In modern deployments, telemetry often reaches the cloud via the MQTT protocol with QoS and Last Will mechanisms, which copes well with an unstable radio link. The choice of protocol affects packet size, and thus energy consumption and the cost of data transmission.
For resource-constrained devices, it is worth considering CoAP over UDP or compact serialization formats (CBOR), which reduce header overhead. In design practice, every byte counts: a single transmission 100 bytes shorter, repeated four times a day over 15 years of operation, genuinely extends battery life and lowers the transmission bill on an IoT tariff.
Which network to choose: NB-IoT, LoRaWAN or LTE-M?
The choice of network is the most important architectural decision in smart metering, because it determines coverage, cost and battery life. Meters usually transmit from a few dozen bytes to a few kilobytes per day, so what matters is building penetration and power draw, not throughput.
A comparison of LPWAN and cellular technologies – NB-IoT, LoRaWAN and LTE-M – shows that NB-IoT offers deep penetration (link gain up to 20 dB) and operator support, LoRaWAN provides independence and low OPEX, and LTE-M works where greater throughput or mobility is needed. For meters in basements, it is precisely signal penetration that is crucial.
How to process meter data in the cloud and at the edge?
A scalable smart metering system combines edge processing with a cloud data pipeline. A concentrator can pre-filter readings and detect anomalies locally, reducing traffic and latency – a classic use case for edge computing in IoT.
On the cloud side, data enters a streaming pipeline. Proven patterns are described in our guide to building an IoT data pipeline on Azure, where raw readings are validated, aggregated into 15-minute profiles and made available to billing systems. A data layer built this way handles millions of meters and allows the detection of non-technical losses and energy theft.
How to secure smart meters?
Smart metering security rests on device identity and data encryption from the meter to the cloud. A meter is a piece of critical infrastructure, so an attack on the fleet can disrupt billing and even load control.
The foundation of trust is strong authentication – we recommend X.509 certificates and a zero-trust model for IoT devices, complemented by secure boot, signed firmware and secure OTA updates. Data in transit is protected by TLS, and keys are stored in a hardware secure element (Secure Element).
Frequently asked questions (FAQ)
How does AMR differ from AMI in smart metering?
AMR (Automatic Meter Reading) is one-way, periodic meter reading, usually for billing purposes. AMI (Advanced Metering Infrastructure) provides two-way communication: remote interval readings every 15-60 minutes, control commands, dynamic tariffs and OTA firmware updates. AMI is today the standard for smart metering.
Which network to choose for meters: NB-IoT or LoRaWAN?
NB-IoT works well in dense urban development and for meters in basements thanks to deep building penetration (gain up to 20 dB) and operator support. LoRaWAN can be cheaper to maintain in rural areas and with private infrastructure. The choice depends on coverage, the cost model and the required reading frequency.
How long does a battery-powered IoT meter last?
A well-designed battery-powered meter achieves 10-20 years of operation on a lithium battery (e.g. LiSOCl2) with a few transmissions per day. The key is sleep modes below 5 µA, efficient radio protocols and minimizing the number and size of telemetry packets.
Summary and key takeaways
Smart metering is a mature IoT technology whose success depends on a coherent design of all layers: an energy-efficient meter, the right LPWAN network, interoperable DLMS/COSEM protocols and a secure, scalable cloud. Key takeaways: go for an AMI architecture with interval readings, match the network to coverage conditions, and plan security from the first day of the project.
FSS designs and deploys complete IoT systems – from meter hardware and firmware, through industrial integrations, to the cloud and analytics. If you are planning a smart metering deployment or a modernization of an existing fleet, check out our offering for designing connected devices and let's talk about your project.
{“@context”: “https://schema.org”, “@type”: “Article”, “headline”: “Smart Metering: An IoT Guide”, “description”: “Smart metering in IoT: AMI architecture, DLMS/COSEM protocols, NB-IoT and LoRaWAN networks and meter security.”, “image”: “https://fss.cc/wp-content/uploads/2026/06/inteligentne-opomiarowanie-smart-metering-iot-1.png”, “inLanguage”: “en-US”, “author”: {“@type”: “Organization”, “name”: “FSS”}, “publisher”: {“@type”: “Organization”, “name”: “FSS”}, “mainEntityOfPage”: {“@type”: “WebPage”, “@id”: “https://fss.cc/inteligentne-opomiarowanie-smart-metering-iot/”}, “keywords”: “smart metering, AMI, NB-IoT, LoRaWAN, DLMS/COSEM, IoT”}{“@context”: “https://schema.org”, “@type”: “FAQPage”, “mainEntity”: [{“@type”: “Question”, “name”: “How does AMR differ from AMI in smart metering?”, “acceptedAnswer”: {“@type”: “Answer”, “text”: “AMR (Automatic Meter Reading) is one-way, periodic meter reading, usually for billing purposes. AMI (Advanced Metering Infrastructure) provides two-way communication: remote interval readings every 15-60 minutes, control commands, dynamic tariffs and OTA firmware updates. AMI is today the standard for smart metering.”}}, {“@type”: “Question”, “name”: “Which network to choose for meters: NB-IoT or LoRaWAN?”, “acceptedAnswer”: {“@type”: “Answer”, “text”: “NB-IoT works well in dense urban development and for meters in basements thanks to deep building penetration (gain up to 20 dB) and operator support. LoRaWAN can be cheaper to maintain in rural areas and with private infrastructure. The choice depends on coverage, the cost model and the required reading frequency.”}}, {“@type”: “Question”, “name”: “How long does a battery-powered IoT meter last?”, “acceptedAnswer”: {“@type”: “Answer”, “text”: “A well-designed battery-powered meter achieves 10-20 years of operation on a lithium battery (e.g. LiSOCl2) with a few transmissions per day. The key is sleep modes below 5 µA, efficient radio protocols and minimizing the number and size of telemetry packets.”}}]}