The volume and reach of connected things are rapidly expanding, with 2020 marking the first time that the number of IoT connections surpassed the number of non-IoT online connections.
There were 11.7 billion IoT connections in 2020 versus 10 billion non-IoT connections -- e.g., smartphones and computers -- according to market research firm IoT Analytics. And researchers there estimated the number of IoT connections will swell to 30.9 billion by 2025.
The availability and expansion of IoT protocols, including 5G and low-power WANs, drives and supports much of that growth.
Why are IoT protocols important?
The benefit and value of IoT comes from enabling the components to communicate; this ability to communicate is what moves data from endpoint devices through the IoT pipeline to central servers.
This communication happens via IoT protocols, which ensure that data sent from endpoint devices, such as sensors, is received and understood by the next and subsequent steps in the connected environment, whether the next step for that data is to another endpoint device or a gateway or an application.
Simply put, IoT protocols are as critical to the existence of IoT as the things themselves.
Although protocols as a collective group are essential to making IoT work, protocols aren't all created equal. Not all protocols work, or work well, in every circumstance, according to Bill Ray, analyst and senior research director at Gartner.
Ray noted that some protocols work well for IoT use in buildings, some are well suited for IoT deployments spread among buildings and others work well for national or global IoT use cases.
How many protocols are there in IoT?
There are multiple IoT protocols available, with each one offering certain capabilities or combinations of features that make it preferable over other options for specific IoT deployments.
Each IoT protocol enables either device-to-device, device-to-gateway or device-to-cloud/data center communication -- or combinations of those communications.
Factors such as geographic and special location, power consumption needs, battery-operated options, the presence of physical barriers and cost determine which protocol is optimal in an IoT deployment.
What are the different layers of IoT architecture?
Networking systems are built as a stack of technologies; these are frequently visualized in a reference model -- a type of framework -- that technologists use to conceptualize how data is communicated over the entire stack.
The most well-known one is the Open Systems Interconnection (OSI) model, which lists seven layers. From bottom to top, the layers are the following:
- Data link
IoT is also expressed in a multilayer model. Although some use the OSI seven-layer model, others in use include the following:
- three-layer model: perception, network and application
- four-layer model: perception, support, network and application
- five-layer model: perception, transport, processing, application and business, or physical, data link, network, transport and application
Internet protocols in use generally vary by layer. As such, an IoT ecosystem could have multiple protocols, with different protocols enabling communication at different layers and with some protocols bridging across layers, said Scott Young, principal research advisor for infrastructure at Info-Tech Research Group.
For example, Bluetooth and wireless support communication at the lowest layers, while Data Distribution Service (DDS) and MQTT work in the application layer.
Most common protocols
Technologists can select from multiple communication protocols when building a network to serve their IoT ecosystem. The most common include the following.
Short for Advanced Message Queuing Protocol, AMQP is an open standard protocol used for more message-oriented middleware. As such, it enables messaging interoperability between systems, regardless of the message brokers or platforms being used. It offers security and interoperability, as well as reliability, even at a distance or over poor networks. It supports communications, even when systems aren't simultaneously available.
2. Bluetooth and BLE
Bluetooth is a short-range wireless technology that uses short-wavelength, ultrahigh-frequency radio waves. It had most commonly been used for audio streaming, but it has also become a significant enabler of wireless and connected devices. As a result, this low-power, low-range connectivity option is a go-to for both personal area networks and IoT deployments.
Another option is Bluetooth Low Energy, known as either Bluetooth LE or BLE, which is a new version optimized for IoT connections. True to its name, BLE consumes less power than standard Bluetooth, which makes it particularly appealing in many use cases, such as health and fitness trackers and smart home devices on the consumer side and for in-store navigation on the commercial side.
Cellular is one of the most widely available and well-known options available for IoT applications, and it is one of the best options for deployments where communications range over longer distances. Although 2G and 3G legacy cellular standards are now being phased out, telecommunications companies are rapidly expanding the reach of newer high-speed standards -- namely, 4G/LTE and 5G. Cellular provides high bandwidth and reliable communication. It's capable of sending high quantities of data, which is an important capability for many IoT deployments. However, those features come at a price: higher cost and power consumption than other options.
The Internet Engineering Task Force Constrained RESTful Environments Working Group in 2013 launched CoAP, for Constrained Application Protocol, having designed it to work with HTTP-based IoT systems. CoAP relies on User Datagram Protocol to establish secure communications and enable data transmission between multiple points. Often used for machine-to-machine (M2M) applications, CoAP enables constrained devices to join an IoT environment, even with the presence of low bandwidth, low availability and/or low-energy devices.
Object Management Group (OMG) developed Data Distribution Service for real-time systems. OMG describes DDS as "a middleware protocol and API standard for data-centric connectivity," explaining that "it integrates the components of a system together, providing low-latency data connectivity, extreme reliability and a scalable architecture that business and mission-critical IoT applications need." This M2M standard enables high-performance and highly scalable real-time data exchange using a publish-subscribe pattern.
6. LoRa and LoRaWAN
LoRa, for long range, is a noncellular wireless technology that, as its name describes, offers long-range communication capabilities. It's low power with secure data transmission for M2M applications and IoT deployments. A proprietary technology, it's now part of Semtech's radio frequency platform. The LoRa Alliance, of which Semtech was a founding member, is now the governing body of LoRa technology. The LoRa Alliance also designed and now maintains LoRaWAN, an open cloud-based protocol that enables IoT devices to communicate LoRa.
OMA SpecWorks describes its Lightweight M2M (LWM2M) as "a device management protocol designed for sensor networks and the demands of an M2M environment." This communication protocol was designed specifically for remote device management and telemetry in IoT environments and other M2M applications; as such, it's a good option for low-power devices with limited processing and storage capabilities.
Developed in 1999 and first known as Message Queuing Telemetry Transport, it's now just MQTT. There is no longer any message queueing in this protocol. MQTT uses a publish-subscribe architecture to enable M2M communication. Its simple messaging protocol works with constrained devices and enables communication between multiple devices. It was designed to work in low-bandwidth situations, such as for sensors and mobile devices on unreliable networks. That capability makes it a commonly preferred option for connecting devices with a small code footprint, as well as for wireless networks with varying levels of latency stemming from bandwidth constraints or unreliable connections. MQTT, which started as a proprietary protocol, is now the leading open source protocol for connecting IoT and industrial IoT devices.
Given its pervasiveness in home, commercial and industrial buildings, Wi-Fi is a frequently used IoT protocol. It offers fast data transfer and is capable of processing large amounts of data. Wi-Fi is particularly well suited within LAN environments, with short- to medium-range distances. Moreover, Wi-Fi's multiple standards -- the most common in homes and some businesses being 802.11n -- give technologists options for deployment. However, many Wi-Fi standards, including the one commonly used in homes, is too power-consuming for some IoT use cases, particularly low-power/battery-powered devices. That limits Wi-Fi as an option for some deployments. Additionally, Wi-Fi's low range and low scalability also limit its feasibility for use in many IoT deployments.
Dating back to the early 2000s when the Jabber open source community first designed its Extensible Messaging and Presence Protocol for real-time human-to-human communication, XMPP is now used for M2M communication in lightweight middleware and for routing XML data. XMPP supports the real-time exchange of structured but extensible data between multiple entities on a network, and it's most often used for consumer-oriented IoT deployments, such as smart appliances. It's an open source protocol supported by the XMPP Standards Foundation.
Zigbee is a mesh network protocol that was designed for building and home automation applications, and it's one of the most popular mesh protocols in IoT environments. A short-range and low-power protocol, Zigbee can be used to extend communication over multiple devices. It has a longer range than BLE, but it has a lower data rate than BLE. Overseen by the Zigbee Alliance, it offers a flexible, self-organizing mesh, ultralow power and a library of applications.
Another proprietary option, Z-Wave is a wireless mesh network communication protocol built on low-power radio frequency technology. Like Bluetooth and Wi-Fi, Z-Wave lets smart devices communicate with encryption, thereby providing a level of security to the IoT deployment. It's commonly used for home automation products and security systems, as well as in commercial applications, such as energy management technologies. It operates on 908.42 MHz radio frequency in the U.S.; although, its frequencies vary country by country. Z-Wave is supported by the Z-Wave Alliance, a member consortium focused on expanding the technology and interoperability of devices that use Z-Wave.
Choosing the right IoT protocol
No single IoT communications protocol is best, nor is any one right for every deployment.
Rather, enterprise technologists must determine which protocol will be best for their organizations based on the unique circumstances of their planned IoT deployments, said Scott Laliberte, managing director and global leader of the Emerging Technology Group with the consulting firm Protiviti. Those determinations should weigh a range of factors, from the power needs of the connected devices and the location of those devices, to the geographic size and features where the deployment will be situated, to the deployment's security requirements.