Developers of IoT applications such as remote sensing and control are being pressed to find the optimum solution for ubiquitous, long-range, low-power, low-cost wireless communications. Three more particularly thorny requirements include high reliability, low latency and minimal interference. While 5G promises to meet those criteria, designers can right now use cellular 4G LTE networks which will be in place for some time to come.
Applications for remote sensing and control include global asset monitoring and tracking, utility metering, industrial machine connectivity, and predictive maintenance in large outdoor plants. The latter includes refineries, chemical plants and mines, smart city infrastructure, wearable and at-home medical monitoring, and smart agriculture.
While there has been much talk of 5G as networks roll out in various regions led by various providers, 4G cellular communications will continue to dominate for the foreseeable future. As such, developers need to be pragmatic in their selection of IoT RF interfaces. This goes beyond the choice of RF module and extends to careful consideration of the ecosystem that is needed to make cellular IoT practical. Critical ecosystem elements include the software stacks, cellular infrastructure, and even the carrier data plans and billing systems needed to ensure comprehensive cellular coverage for IoT use.
This article will provide a broad look at cellular IoT applications and the embedded technology needed to realize these applications. It will also provide a detailed discussion on the use of the most current hardware and software technology and the resources available that can help engineers develop cellular IoT designs for the diverse applications listed above. Information on readily available data plans will also be included.
Why not 5G?
Despite much progress, 5G network and equipment standards are not yet fully finalized. Even when the standards are finalized, it will take several years for standardized 5G networks and equipment to be built and deployed. Meanwhile, 4G LTE networks have been in commercial operation since 2011 and are quite capable of delivering the performance and range required by most IoT applications.
By at least one estimate, 4G LTE networks account for about 40% of the current worldwide cellular market, while older 2G and 3G networks each represent about 30% of the market. Even by 2025, 5G market penetration is not expected to be more than about 15%. With that in mind, designers of IoT systems that need long range and low-power should leverage the existing cellular infrastructure and conform to the 4G LTE and earlier standards. These are not only in place now, but are even evolving as in the case of 4G LTE to meet the needs of the IoT.
LTE evolving for IoT
The Third Generation Partnership Project’s (3GPP’s) Release 13 of the LTE standard defined new LTE categories for IoT applications: Category M1 (Cat-M1), formerly known as eMTC (enhanced Machine Type Communication), and Category NB1 (Cat-NB1), formerly known as Narrowband-IoT (NB-IoT). These new categories extend LTE for IoT by enabling support for lower power, longer range, lower latency, and lower cost, as well as minimal interference by virtue of being in licensed bands.
Cat-M1 defines a 1.4 MHz channel width and a throughput of 375 kilobits per second (kbits/s) for the uplink and 300 kbits/s for the downlink. Cat-NB1 defines a much narrower channel width of 200 kilohertz (kHz) with a throughput measured in tens of kilobits per second. Cat-M1 latency is approximately 10 to 15 milliseconds, while Cat-NB1 latencies are measured in seconds and can be as much as much as 10 seconds in some deployment scenarios.
This performance is sufficient for many IoT sensing applications such as meter readers, health status monitors, and highly mobile fitness applications that can benefit from the long reach and ubiquitous presence of cellular communications. Currently, and for the foreseeable future, no other low-power, wide area, wireless technology offers the scalability, security, and longevity of the established 4G LTE networks.
Connecting to the cloud
Several vendors already offer modules that either operate as cellular data modems or integrate a cellular data modem into an embedded development platform. These modules connect IoT devices to the cloud via 4G LTE (or even earlier) cellular networks. However, a hardware module alone can’t connect an IoT device to the cloud. Appropriate software and a managed connection to a cellular provider are also required. Without all three, there’s no connectivity.
The choice between cellular IoT modules with and without application processors depends on whether the project’s hardware design is starting from scratch, or if it’s adding a cellular IoT connection to an existing embedded design. The following are brief overviews of some 4G LTE cellular modem modules and ICs with and without onboard application processors.
Sierra Wireless' AirPrime WP7702 low-power, wide area (LPWA) module integrates an application processing subsystem and a cellular data modem into a small package measuring 22 x 23 x 2.5 millimeters (mm). The module complies with the 3GPP’s Release 13 standard and implements the Cat-M1 and Cat-NB1 protocols. Its peak Cat-M1 data rates are 300 kbits/s for downloads and 375 kbits/s for uploads. Its Cat-NB1 data rates are 27 kbits/s for downloads and 65 kbits/s for uploads.
Figure 1: Sierra Wireless' AirPrime WP7702 RF module includes an application processor and support for both Cat-M1 and Cat-NB1 cellular protocols. (Image source: Sierra Wireless)
In tandem with its AirPrime RF modules, Sierra Wireless provides the Developer Studio integrated development environment (IDE). Built on the Eclipse Java IDE, it allows developers to create applications using the open source Legato application framework with an intuitive graphical user interface (GUI) running on Windows, Linux, and MacOS hosts. The tool includes utilities and functions useful during the various phases of the design cycle to support application software development for wireless data applications.
Legato combines a Linux-based OS distribution (which runs on the WP7702 module’s integrated, 1.3 gigahertz (GHz) Arm® Cortex®-A7 processor), a board support package (BSP), and customized development tools that run on a host PC. Sierra Wireless also offers the AirVantage IoT Platform, a self-service portal that provides connectivity and device management for a fleet of Sierra Wireless cellular modems connected to a number of different worldwide cellular carriers. These include AT&T, Verizon, NTT, Telstra, KT, and SKT. AirVantage also automates firmware updates for Sierra Wireless modems over the wireless connection.
Talon Communications, Inc. put the Sierra Wireless WP7702 module on a carrier card that doubles as a development platform. Together, the WP7702 module and carrier card comprise the mangOH Red™ evaluation board, which has an onboard socket for a cellular micro SIM card (required for obtaining carrier service). The mangOH Red platform breaks out many of the WP7702 module’s interface pins to connectors, including three antenna connectors, two micro USB connectors, a full-sized USB host port, a pin header with I2C, SPI, UART and GPIO I/O pins, and a 3.5 mm stereo audio output jack.
Figure 2: The mangOH Red development board supports application development for Sierra Wireless' WP7702 RF module. (Image source: Talon Communications)
The mangOH Red’s USB host port connects the development platform to a host computer for software development. Loading the appropriate windows driver and installing Sierra Wireless’ Legato Developer Studio completes the setup for wireless IoT application development using the WP7702 module.
u-blox's SARA-R410M-02B is an ultra-compact LTE Cat-M1 and Cat-NB1 RF transceiver module measuring 16 x 26 x 2.5 mm and comes in a 96 pin LGA package.
This transceiver module connects to a host processor over its USB or UART interface and is controlled by the host processor using a string-oriented AT command set defined by the 3GPP. The SARA-R410M-02B also has a SIM card interface for carrier service identification.
Figure 3: The u-blox SARA-R410M-02B RF transceiver module implements a complete Cat-M1 and Cat-NB1 radio and baseband that can be attached to a host processor. (Image source: u-blox)
The u-blox EVK-R4 evaluation kit incorporates and breaks out the u-blox SARA-R410M module’s I/O pins. It provides appropriate connectors for attaching the module to antennas, power, and a host processor. It also has an onboard SIM card holder and accepts a GNSS (Global Navigation Satellite System) daughter card. GNSS devices are often paired with cellular radios for tracking applications. (For more information about GNSS devices and modules, see, “Design Location Tracking Systems Quickly Using GNSS Modules” and “Add Rapid Acquisition and High Accuracy to Tracking Applications Using Cost-Effective GNSS Modules.”)
Figure 4: u-blox’s EVK-R4 development kit breaks out the u-blox SARA-R410M module’s I/O pins for easier development. (Image source: u-blox)
Hologram, Inc. took the u-blox SARA-R410M module and mounted it on a small USB board to create the HOL-NOVA-R410. This solution provides a quick way to add LTE Cat-M1 and Cat-NB1 RF transceiver capabilities to existing products with USB ports.
Figure 5: Hologram Inc’s NOVA-R410 places a u-blox SARA-R410M cellular RF modem on a small USB carrier to simplify adding long distance IoT RF communications to USB equipped systems. (Image source: u-blox)
Nordic Semiconductor's nRF9160 system-in-package (SiP) incorporates an application microcontroller, a full LTE modem, a transceiver front-end, and power management in a package measuring 10 x 16 x 1 mm. The module includes GPS support for asset tracking. Combining location data obtained from the cellular network with GPS satellite trilateration allows remote monitoring of the device’s position.
The nRF9160’s application processor is an Arm Cortex-M33 running at 64 megahertz, which is combined with 256 kilobytes (Kbytes) of static RAM and 1 megabyte (Mbyte) of flash memory. The module’s 4G LTE modem implements the 3GPP Release 13 Cat-M1 and Cat-NB1, and the Release 14 Cat-NB1 and Cat-NB2 protocols.
Nordic Semiconductor’s nRF9160-DK development kit for the nRF9160 module includes an nRF9160 module mounted on a carrier board.
Figure 6: Nordic Semiconductor’s nRF9160-DK development kit breaks out all of the pins of an nRF9160 cellular module for development work and comes with extensive software support. (Image source: Nordic Semiconductor)
The software development kit (SDK) includes:
- The Zephyr Project scalable real-time operating system (RTOS) for the nRF9160
- The MCUboot secure bootloader
- The nrfxlib RTOS independent libraries
A block diagram of the nRF9160-DK development kit shows the support components an nRF9160 might require.
Figure 7: The block diagram for Nordic Semiconductor’s NRF9160-DK development kit illustrates the support components an nRF9160 cellular IoT SiP might require. (Image source: Nordic Semiconductor)
Nordic recommends using the Embedded Studio IDE from Segger Microcontroller Systems for building nRF9160 applications. A specialized version of the Segger Embedded Studio is available free of charge for use with Nordic Semiconductor devices, including the nRF9160 SiP.
A word on data plans
Before deploying a device on an operator’s network, it must first go through a qualification process to make sure it complies with the operator’s requirements with respect to bands and interference. Before going through that process, the developer needs to choose a suitable data plan, and factor the cost of that data plan over the long term. To help with this, a list of available IoT cellular data plans is provided here as a useful resource.
The cellular IoT landscape is changing rapidly, especially with the advent and imminent introduction of 5G cellular technologies. RF modules for cellular IoT applications are available, but they require the support of an ecosystem to make cellular IoT practical. This ecosystem includes the software development tools, stacks, and libraries needed to turn the silicon and module solutions into a deployable product. Until such time as 5G becomes widely available, modules based on 4G LTE will be a viable solution for remote IoT sensing and control for many years to come.