Why LwM2M enables simpler approaches than MQTT
Written by Charles Mitchell, UK & Ireland Regional Sales Director, Quectel
Message queueing telemetry transport (MQTT) has been a fundamental enabler of integrating IoT devices to the cloud. However, with large-scale IoT projects now in deployment and the increased usage of low power wide area networks (LPWAN) for IoT integration into hyperscale clouds, there are a series of market challenges that mean MQTT is no longer suitable.
MQTT remains an excellent technology for fixed position devices that have wired connectivity and are wired into the power grid but for smaller, cheaper devices that utilise wireless connectivity and are powered by batteries, it demands too many resources. Lightweight M2M (LwM2M) in contrast has been designed to minimise data consumption and power consumption.
Massive connectivity is here, and huge growth is ongoing. Predictions are that there will be more than five billion devices connected by cellular IoT connectivity in 2025 and 60% of these will be LPWA connections. These will provide advantages in terms of better coverage, low cost, low power consumption and mobility. In addition, they will enable voice over LTE, bringing voice to the low power wireless market.
These attributes are set to ignite adoption for use cases in public utilities, industry, agriculture, smart living, and smart home sectors. With the arrival of digital twins between the intelligent edge and intelligent cloud, the role of LPWA connectivity has been highlighted because it helps enable the construction of twins. The challenge is that to model the physical world, you need to sense the physical world and model it as a digital twin. However, to achieve that you must go out and see what is happening in the world. From orchards to car parks to power grids and meters or in multi-dwelling units and office spaces with many endpoints that are battery-driven, LPWANs are the best-suited connectivity.
Technology providers are working to ensure that high-quantity IoT data of this type can effectively reach the hyperscale cloud utilising LPWAN as effectively as it can with 5G or Wi-Fi. Connectivity is complementary but the low power of NB-IoT and LoRa enables low-cost devices to get to the IoT platform in which applications can be fuelled by IoT.
Platforms have been developed to provide templates for various applications, these make it simple to develop for IoT use cases that rely on LPWAN. Devices are of no value until there are apps and outcomes so the opportunity is to merge LPWAN with hyperscale cloud services and to create an easy to deploy LPWAN Cloud Gateway Service. MQTT is pervasive but it is not easy to deploy and develop for at IoT hyperscale volumes, so the question is how to make it easy?
The number of developers that can take device data from a module and move it across an environment across various IoT technology domains globally is in the hundreds of thousands, not the millions of more mainstream developers. Therefore, development must be made as easy and as seamless as possible for app developers. Now is the time to make sure every developer can have easy access to the data that’s being kicked off LPWAN devices so it can be utilised and turned into value.
Hyperscalers rely on MQTT today because it is very capable, but it is simply not efficient enough for modern LPWAN IoT solutions which often use less data and have power constraints. With lightweight M2M (LwM2M), organisations are trying to create a substantially smaller use of data. LwM2M offers 72% less data transfer at initial connection, 31% less data transfer for steady-state device connection and power consumption can be 33% lower.
Qualitatively, MQTT has been easier to implement and procure than LwM2M solutions – but that is changing as these solutions are more widely used to simplify implementation of LPWAN devices into the hyperscale cloud. In addition, LwM2M provides more long-term flexibility in constrained networks and is therefore seeing greater adoption.