Low-power diet for IoT devices

The number of IoT devices demonstrates steady, multiple annual growth, with IoT data turning into commodities. The IoT market is getting mature, so every competitive advantage matters. Traditionally,  the “smaller”, “cheaper”, “more versatile”, “less power-consuming” categories are among the candidates to win.

Taking into account that the majority of IoT devices are powered from a battery, every lost nAh results in shorter battery lifetime, uptime and additional on-site maintenance. In other words, it leads to an increasing cost of ownership and the situations where the cost of maintenance exceeds the cost of the device itself. Designing an IoT system without proper consideration of power-saving technologies is a risky and expensive game to play, especially in the context of IoT.

In this article, we cover:

• The most power-saving connectivity technologies
• Power consumption distribution within an IoT device
• Solutions prolonging battery life in IoT devices
• Common expectations towards battery longevity
• Prylada’s energy-efficient IoT solutions

The most power-saving connectivity technologies

Constantly expanding functionality of modern IoT devices designed to fit with a wider range of applications requires truly sophisticated power-saving solutions. The battery capacity factor becomes even more crucial when talking about the systems and devices that are designed to minimize maintenance costs.

Obviously, long battery lifetime is the result of optimized power consumption and the breeding ground for minimized device maintenance costs. With a huge variety of connectivity options, the new-generation Low Power Wide Area (LPWA) technologies, particularly NB-IoT and LTE-M, occupy the leading positions in terms of energy efficiency and low power consumption for IoT devices.

The key ingredients of this power-saving connectivity cuisine are Power Saving Mode (PSM) and Extended Discontinuous Reception (eDRX). So let’s fast forward to what this all means.

In a nutshell, PSM is a feature of a cellular modem that turns off the device radio and puts the device to sleep without the need of reconnecting itself to the network when it next wakes up. Reconnecting to a network increases energy consumption, so PSM helps reduce that by only connecting to the network when required.

eDRX is defined as an extension of the Discontinuous Reception (DRX) feature designed for the same purpose as PSM — reduce power consumption in IoT devices. But the general concept of DRX differs, which is well explained by everyrthingRF. A device goes into sleep mode for a certain period and then wakes up after a fixed interval to gather and transmit data, or receive signals. This means data may be delayed in getting to its destination due to this fixed wake-up periodicity, but this may not pose a problem for many IoT networks. With eDRX, the device can listen for pending data indications without having to establish a full network connection. By just listening for a pending data indication, eDRX uses less power than if it made a full network connection. The time needed for this listening process is also much shorter than the time it takes to make a full network connection.

Therefore, eDRX extends the DRX cycles to keep a device in a power-saving state for a longer period of time. When applied together, the eDRX and PSM features complement each other, which allow reaching even higher power efficiency.

While providing the same PSM and eDRX capabilities, the NB-IoT and LTE-M technologies are not one size fits all though. The selection of the right technology mostly relies on its availability in the region where it’s going to be deployed, the network operator and a specific use case. The primary question one should ask before making a decision on the connectivity technology: “What data, how much of it and how frequent is it expected to transfer across the system?” In our previous post about the leading LPWA technologies we presented a quick roundup of the differences between the features and applications of NB-IoT and LTE-M.

Still there are regions where the deployment of any LPWA technology is unavailable, and many devices are working on 2G. Even with the ongoing rollout of NB-IoT and LTE-M, some systems and applications may require 2G connectivity to fill the gaps, especially in terms of global coverage. This factor becomes a real power saving bottleneck for highly scaled networks.

Power consumption distribution within an IoT device

Implementing one of the LPWA technologies is only half a success in terms of low-power consumption. A high level of power efficiency can be only achieved by applying a complex of solutions that reduce a huge power appetite of some IoT device components. Let’s look at the picture of power consumption distribution between such components.

Cellular modem is definitely among the leaders of power budget consumers, but generally it depends on the IoT device architecture what share of that budget belongs to the modem. For instance, If we build a device with just a couple of simple digital inputs and some short-range wireless interface (e.g. Sub-1 GHz), then the NB-IoT/LTE-M modem is undoubtedly an energy guzzler. But as soon as the number of device interfaces grows, and some of them get galvanically isolated, the primacy goes to the interfaces.

The general concept underlined here is that the more interfaces a device is equipped with, the more power consumption accounts for the whole network (distribution between more network components). It means that the major percentage of power consumption shifts left from the modem itself.  All galvanic isolation is very expensive in terms of energy, and taking into account that a single isolated input/output (IO) can cost the power budget 35mA, the cellular modem easily loses its leadership within the network with just 4x (and more) industrial IOs.

Another winning team in the nomination of power expensive components are devices that have “always-listening” capabilities, for example speakers, video cameras and smartphones. The same relates to the “always-watching” devices, such as surveillance systems and cameras that recognize faces, gestures, body postures, and so on. Multiple fitness, tracking and running monitoring devices are also included into this category as “always-sensing”.

Smart power-saving solutions are even more crucial for the groups of devices listed above. And what is pretty attractive, such solutions already exist, including the power-save (sleep) modes that we already discussed in the previous paragraphs.

Solutions prolonging battery life in IoT devices

Before we proceed, it’s worth mentioning that all the suggested solutions are not carved-in-stone, since the definition of “best” is completely application-based.

One of the practices to start with is disconnecting the power source for the period when the IoT module is not in use. Usually, sensors that aggregate data do not need to operate continuously. Instead, they or their parts can be powered in a way so that they are active only during the periods when data are collected and transferred, and then shutted down again. A well-designed software can be the right solution to coordinate such active/inactive phases.

The way the sensors will reconnect to the network is also subject for optimization. Previously in the article, we have already covered the most power-saving connectivity technologies, such as PSM and eDRX.

Energy harvesting is another candidate for leadership among other power management best practices. The concept of this technique revolves around extracting unused and wasted power from the ambient energy sources surrounding a device.

• Thermal energy. Temperature differential is a source of electric potential that can be captured by thermoelectric materials used in many IoT devices.
• Solar energy. The IoT devices exposed to sunlight can benefit from this energy source enormously. One of the advantages of this technology is high-energy density.
• Wind and aeroelastics vibrations. This alternative energy source is gaining more attention due to global climate change. This is especially beneficial in the context of agricultural monitoring.    
• Mechanical vibrations. This ambient source is commonly converted into usable electrical energy and can work in favor of piezoelectric materials. Powering of sensors for damage detection in machinery is a common application of this energy harvesting technology.

In many environments where multiple sources of ambient energy are present it may be possible to develop a mixed energy harvesting strategy. To achieve the best power-saving results, the advantages of energy harvesting, IOT device power distribution architecture, and LPWA connectivity technologies can be combined.

Common expectations towards battery longevity

What is the optimal battery lifetime expected by the majority of IoT network stakeholders? Power consumption profile is getting more and more a requirement towards all modules integrated into an existing network. This is simply explained by the customer’s desire to make sure that everything embedded into their network is really energy-optimized.

Battery life in IoT devices is determined by multiple factors, such as the type of processor, software algorithms, firmware configurations, and communications technologies. All this makes the battery discharge patterns unique for every IoT application. While making predictions on battery lifetime, many engineers often ignore another (minor as it may seem, but critical in reality) factor referring to battery shelf-life and self-discharge.

One of the most common ways to predict the battery life of an IoT device is identifying its long-term consumption profile. Evaluation of power consumption of all integrated components is usually based on measuring by a special software, taking into account various operation states of the device. Each state is associated with a fixed power consumption, which can be derived from the datasheets of the component, or measured. The total battery discharge over the measured period can be then considered as the sum of all state combinations. It’s worth mentioning that this method is commonly recognized due to its practical approach.

The expectations of IoT network owners usually rely on measurement-based predictions. An ideal scenario is when an IoT device operates from one battery as long as intended during deployment, with potential failed states taken into account as well. But the ideality of this scenario gets easily broken due to the discharge variance of batteries of the same model. Looking at the measured average discharge time may not be sufficient.  Mistaken assumptions on remaining battery voltage may also be caused by inaccurate or outdated specifications provided in the component’s datasheets. Every time a new component is added to the entire device or the network in general, the energy consumption profile needs to be re-evaluated, even if this component is not actively used. Being aware of all the nuances and taking the relevant measures will help get closer to the expectations met.

Prylada’s energy-efficient IoT solutions

Prylada Gateway is a brand-new data acquisition module designed to integrate all devices into a single IoT network with centralized control. The module can be powered via dual-source prioritized power supply that ensures its reliability and enables battery backup functionality. This means that the backup-battery connection scheme enables the Prylada Gateway to power either from the mains or the built-in battery and allows switching between them automatically.

The implemented reverse polarity protection ensures that the module is not damaged if the power supply polarity is reversed on one hand, and is built in the effective way, eliminating power loss, inherent the diode-based solutions on the other.

Another component of the energy efficiency is the capability of the Prylada IoT Gateway blocks to be switched on and off independently. For example, if you switch off the block that integrates all interfaces, thus only the data transfer modem remains operating. Every Pryalada’s power supply block was selected to meet low-power consumption requirements.

High energy efficiency of the Prylada Gateway is also achieved due to its capability to get connected via the NB-IoT or LTE-M technologies. Moreover, the data transfer modem supports the power save mode and goes to sleep when it’s inactive.

The firmware is also designed with the power saving mode in mind. It’s configured to keep only the needed parts of the device awake, while shutting off the other parts that are not in use.  

Prylada provides far more solutions that optimize power consumption, for example, a smart bistable relay. It is designed to be activated by voltage pulses, which not only allows to save the controlled device battery, but also to fix the device in the actual position once the current is removed.

If you are interested in how our solutions can optimize your existing network, drop us a line at contact@prylada.com.

‍