IoT and M2M Connectivity Explained
What differentiates IoT and M2M devices from traditional sensors is their ability to broadcast data to a dedicated network (IoT) or directly to other devices (M2M). Reliable connectivity is the fundamental requirement for any IoT environment. While private smartphone users mostly have the same needs and expectations (which can be accommodated for, using 4G and 5G), the network requirements of IoT environments vary wildly for each use-case. As such, there are many different connectivity standards to accommodate different applications. These standards are either being developed, actively rolled out, or already widely available across the globe. Some are proprietary and require cost-intensive gateways to allow for maximum customization, while others have taken every sacrifice possible in order to reduce deployment and maintenance costs.
There is no “perfect” connectivity solution
Choosing the right connectivity standard is crucial for enterprises and device manufacturers. It should be one of the first and important decisions to make, as switching connectivity standard later one would mean having to revamp most of the network’s architecture.
There are a plethora of different network standards to choose from, some of which may seem very similar on paper. There is also plenty of biased information, which has been published by a company with obvious commercial incentives to push for a certain standard. Whether they are the proprietary owner of a connectivity standard, manufacture devices or gateways dedicated to a specific standard or a SIM card manufacturer that supports a certain standard.
Understanding the key factors of different connectivity standards will help enterprises looking to deploy IoT networks to understand where their priorities lie. For instance, if you are looking to manage a fleet of cargo ships, connectivity range is obviously far more important than data speed (unless the fleet manager needs data gathered by high-resolution IoT-enabled onboard CCTV). When a connectivity standard provides a high range, they will likely have to sacrifice other factors, such as data bandwidth or latency. Ultimately, there is no perfect connectivity standard, however, there is an ideal standard for your specific situation. Let’s take a look at some of the key factors at play:
The range is the distance over which a stable connection between various IoT devices can be maintained. While range is likely important for most IoT projects involving IoT devices scattered across international waters, deep underground, or behind thick concrete walls, there are also use cases where short-range standards can be a useful security asset. Limiting the communication range to the vicinity of the building makes it physically impossible to infiltrate the network from afar.
Bandwidth represents the volume of data that is transferred within a given timeframe. E.g. Kb/s, Mb/s or Gb/s. Every connectivity type has unique predefined packet sizes used to transfer data. The connectivity protocol of your choice should be able to match the maximum data volume that is expected to occur on your future IoT system. Protocols with packet size larger than needed make it inefficient to use (unnecessary power consumption) while utilizing a protocol with packets not large enough will lead to data congestion.
IoT-devices often transmit sensitive corporate data, which needs to be protected throughout all the stages of transmission and storage. As data gets moved from one place to another, it is vulnerable. However, thanks to numerous technological milestones in the security of connectivity protocols, IoT networks, and their exchanged data can be protected by port protection, authentication, and encryption.
Power consumption of IoT- and M2M-enabled devices is a crucial factor. How much power these devices consume is directly related to the connectivity standard they utilize. Power consumption is a crucial factor for many IoT environments, as many IoT and M2M devices are battery-powered. Thus high power consumption typically equals to high operating costs, as these devices are often located in remote locations, which means manually replacing their batteries is cost-intensive labour.
Many IoT environments are large-scale projects involving hundreds or thousands of IoT and M2M devices. Scalability is another crucial factor. A connectivity standard needs to continue to perform as the network becomes increasingly larger, as most networks are likely to grow as enterprises realize the valuable data their devices provide.
The most notable connectivity standards for IoT and M2M
NB-IoT (also known as Narrowband IoT or LTE Cat NB1) is an LTE-derivative that was specified in 3GPP’s release 13. As the name suggests, it uses a narrow band of 180kHz. This reduces its data rate significantly; however, it also reduces power consumption and boasts with a greater range than other LTE-derivatives, such as LTE-M. NB-IoT does not offer handover support, so it is much more suitable for stationary devices and less than ideal for mobile IoT applications. NB-IoT can already be utilized in several countries and is currently being rolled in many others. NB-IoT has a massive amount of potential use-cases, as long as devices for a specific solution will be stationary.
LTE-M / CAT-M (LPWAN)
LTE-M (Also known as LTE Cat M1 or eMTC) is another standard by the 3GPP. LTE-M can be seen as a simplified LTE, with the priorities of less power consumption and more range. There is a sacrifice, though, LTE-M only achieves around one-tenth of the data rate of conventional data rate, and the bandwidth is only 1.4MHz compared to LTE’s 20MHz. LTE-M utilizes two features known as PSM (Power Saving Mode) and eDRX (extended Discontinuous Reception), which both aim at reducing power consumption, by allowing for nodes to be inactive and longer paging cycles. LTE-M stands out compared to NB-IoT, as it possesses handover capabilities, which makes it particularly useful for IoT-network with devices that frequently change location. The LTE-M protocol will likely be utilized by agricultural monitoring, logistics, fleet management, asset tracking, smart wearables, or smart meters.
Sigfox (Proprietary LPWAN)
SigFox is a proprietary connectivity solution. However, it has been around for much longer and has already deployed stations across the globe. In the U.S. though, Sigfox has relatively weak deployment performance and is financially struggling. Sigfox can be praised for its low deployment costs. Its radio modules cost around 5$, while LoRa’s cost 10$ and NB-IoT’s cost around 12$. However, these low costs have also made it difficult for the company to make sufficient profits. Sigfox boasts with its wide coverage area and low power consumption. However, it sacrifices downlink capability and is regarded as uplink only. Downlink is possible, but it is very restricted and less-than-ideal.
LoRaWAN (Proprietary LPWAN)
LoRaWAN (Long Range Wide Area Network) is based on the physical LoRa (Long Range), which is a proprietary connectivity standard released by Semtech. LoRaWAN is an incredibly energy efficient protocol, as its asynchronous nodes only communicate with the network when they have data ready to send. LoRa can be seen as a proprietary derivative of chirp spread spectrum modulation (CSS), as opposed to FSK (Frequency Key Shifting), which is more common with the traditional wireless systems. Both feature low power consumption, but CSS has a much better communication range. Military and space communication operations have relied on CSS for decades, valuing its low power consumption, high range, and resilience. LoRa achieved a record in 2017 when it sent and received a data packet over a distance of over 702km. While LoRa works well for devices that are in motion and provides a longer battery life to its devices, compared to NB-IoT. It has a lower data rate and a higher latency compared to IoT. LoRa requires a gateway in order to be operated, while this is more costly, it can also be seen as an advantage as one can set up and manage their network.
4G (Cellular Network Protocol)
4G’s (fourth-generation communications system) architecture is quite similar to 3G, but it is enhanced with LTE (Long-Term Evolution). The standard was first launched in 2007 and back then, was praised for its improved data rate and capacity (network congestion was a major issue in the advent of 3G). While 4G is far from ideal for most IoT use-cases is more practical than 5G. 5G does not have a far reach, and thus many cells need to be installed in order to achieve coverage similar to 4G. Few know that 5G cells have a directional signal, while 4G can broadcast its signal in all directions, further improving its range of coverage. However, 4G does not have 5Gs unique slicing capabilities and thus cannot accommodate for different use-cases with different network requirements. While we likely won’t see widespread adoption of 4G in IoT or M2M networks, 4G development can take credit for two other, much more promising protocols. Two distinct, widely popular LPWAN standards were derived from 4G/LTE-technology, namely the aforementioned: LTE-M and NB-IoT.
5G (Cellular Network Protocol)
5G is the newest network mobile technology, which is actively being rolled out in many countries. While most people think of it as a connectivity standard that will revolutionize mobile communication, some of its core features also make it an attractive protocol for more than a handful of data-intensive IoT use-cases.
Amongst others: augmented or virtual reality, automated manufacturing, telemedicine, autonomous vehicles, smart cities, and any IoT or M2M device that needs data-intensive transmission (high definition video streams being the most prominent example).
5G’s speed is uncontested compared to other standards in this article. It can reduce latency to 1ms, achieve a data-rate to 20Gbps, all at a capacity of a million devices per square kilometer. However, such impressive feats always come with a sacrifice.
5G operates on a very high frequency, at very short wavelengths. These short wavelengths do not have a good range, so many 5G towers will need to be installed, and it will out of the question for enterprises looking to deploy remote M2M or IoT sensors.
One of 5Gs most unique features is network slicing. This allows users to slice different types of traffic, enabling them to accommodate for different uses. One slice can be used for less crucial data, while others can be used for first responders who need a reliable, low-latency connection, while another slice can be used for IoT networks, which require a high communication range.