When devising an IoT architecture, the subject of unidirectional (one-way) vs bidirectional (two-way) communication is likely to come up at least once. In Low Power Wide Area Networks (LPWAN) for remote telemetry applications, bidirectional connectivity has aroused mixed feelings among many industrial adopters. Due to security concerns, there’s been a common preference for unidirectional communication to connect LPWAN sensor networks.

Yet, as the industrial IoT knocks on the door, two-way communication is revealing a new host of compelling use cases that were previously less known. For those looking to get a better grasp of this topic, this blog delves into how industrial users can reap the benefits of bidirectional connectivity for granular LPWAN telemetry sensors while clearing major concerns around it.

The Concept of Two-Way Communication in IoT

Way before the term “Internet of Things” exists, M2M two-way communication had been part of our daily life. Think of the last time you send and receive an SMS on your phone or trying to upload and download something from the Internet with your PC.

As IoT ushers in a new wave of connected devices, we continue to witness growing pervasiveness of two-way connectivity, particularly in the consumer space. For example, by setting up the connection between your wearable device and smartphone, you can have incoming calls displayed on your smartwatch and easily choose to answer or decline them from the wearable screen. Similarly, it is easy to view and adjust the settings of your smart thermostat via the phone while being away from home.

As in the first example, bidirectional communication can be established directly between two devices. But with the cloud, you more often see it associated with data exchange between devices and a cloud-based analytics platform via an IoT gateway – as in the second example. Leveraging machine learning and AI algorithms to analyze user preferences, the cloud then processes these inputs and issues a response to the IoT device.

How about the Industrial IoT Space?

Industrial users are usually hesitant in adopting IoT two-way connectivity for telemetry applications due to its security implications. Until today, bidirectional communication mostly pertains to time-sensitive, closed-loop communication between controllers and I/O modules that are contained on the shop floor. The idea of exposing critical operational data to the Internet via two-way wireless connectivity now stirs up concerns around data privacy and malicious attempts. Plus, the fact that many legacy systems are designed with minimal security features in mind further magnifies these concerns.

Another major reason why two-way communication takes little spotlight in industrial LPWAN is that it isn’t a prerequisite for many condition-based monitoring use cases. Quite often, companies only focus on fetching data from previously isolated and disconnected assets for fine-grained operational insights and integration into higher management systems like ERP. Decision-making is then executed through separate workflows such as dispatching technicians for asset maintenance or setting up supply orders – rather than actuating field sensors.

The Greatest Use Case for Two-Way Communication in LPWAN Sensor Networks

What industrial users often miss out when considering bidirectional communication is its enormous benefit for device configuration and management. With the emergence of industrial innovations that rely on extensive telemetry networks like the digital twin, the ability to provision and configure sensor networks at scale has become more imperative than ever.

In LPWAN deployments with a vast number of endpoints, manually registering each device to the network is a labor-intensive and daunting process. Leveraging two-way connectivity, you can automatically authenticate and provision remote endpoints over the air (OTA). Likewise, if a device is no longer needed, you can conveniently retire it from afar – without the need to travel to the site.

Beyond device on- and off-boarding, bidirectional communication is also a game-changer for seamless and hassle-free configuration of field devices throughout their lifecycle. This advantage is particularly pertinent in industrial operations where regular sensor calibration is key to accurate process and asset measurements.

Under changes in ambient conditions, mechanical wear and tear or a shift in the required operating range, industrial sensors are subject to measurement errors from time to time. For this reason, calibration which us relevant adjustment(s) in sensor configuration, must be performed periodically to minimize unwanted deviations in sensor outputs. The process can be very time consuming, especially if devices are hard to reach. With bidirectional communication, you can issue calibration metrics to remote IoT sensors in simple steps to save big on time and costs.

For the new generation of smart sensors purpose-built for functionality and power efficiency, over-the-air device configuration is even more important. Instead of capturing a single parameter, today’s smart sensors are self-contained and designed with multi-sensing capability in mind. In this context, two-way IoT connectivity enables simple (re-)configuration of different sensing functions, as well as message frequency along the line – to optimize device performance and battery life.

For example, you can switch off irrelevant sensing units to drastically save on energy and later switch them back on when the need arises – all conveniently from the control center. Likewise, if the sensor is transmitting too often, it’s easy to adjust the message frequency from afar.

Coming Back to the Security Question…

Contrary to popular belief, two-way wireless connectivity doesn’t always mean Internet-connected. (Third-party) cloud architecture might seem prevalent in IoT, but it’s by no means a must – especially in industrial deployments. With a privately managed and controlled network, bidirectional communication between connected devices and your backend/ management console can stay safe on-premises.

Two-Way Communication in LPWAN

While most legacy solutions offer bidirectional communication at the user’s disposal, their underlying mechanism is fraught with a major pitfall that risks message delay. Typically, in an LPWAN network, a downlink message is issued to end devices only if an uplink message has been sent. To minimize power consumption, the device is active just for a few seconds after the uplink transmission to listen for whether a response is coming back. However, it’s common that data processing at the backend for downlink communication can take longer than the provided time window. As such, sensor devices could go back to sleep before the downlink arrives, meaning it must be postponed to the next time an uplink is sent – which could be an hour or even a day later.

To mitigate this risk of obsolete messages, the MIOTY protocol used in MYTHINGS networks allows devices to go back to sleep immediately after the uplink transmission and wake up only some seconds later to listen for the response. By giving the backend sufficient time to schedule downlink messages, this mechanism enables a highly responsive, yet power-optimized LPWAN networks.

Wrapping Up

For industrial IoT applications that rely on vast, granular LPWAN telemetry networks, two-way communication is often less about actuating field sensors. Instead, its greatest use case lies in the ability to manage and (re-)configure sensor devices at scale and conveniently from afar. The larger the network, the greater the benefits. With the right architecture, security and data privacy concerns can also be ruled out to help you fully harness the value of your connected infrastructure.