Wireless IoT sensing devices can be placed on, in, or near people, equipment, infrastructure, and our environment. This gives us new tools to address the most urgent challenges of our 21st century world: from climate change, to ensuring clean energy, safe food, and foremost, caring for the health and well-being of an aging population. However, to achieve this, we need to address the ‘powering the IoT’ gap. That is, solutions need to run on batteries that outlive the IoT devices they power.
This article explores the critical contribution that energy harvesting (EH) powered solutions can provide for the IoT. Of the trillion sensors that could be deployed within the next several years, a significant majority will be of the ultra-low power (ULP) wireless variety. These are also the best candidates for EH, which can either supplement external power or serve as independent power sources.
The approach we take to powering the IoT is critical to enabling so many of the technologies changing our world every day. For example, connected and autonomous vehicles (CAV), will be dependent on reliable and ubiquitous sensing with both high- and low-bandwidth connectivity, all of which requires increased power density and weight-reduction — two things that self-powered, wireless sensor networks support.
A major added value of EH is to provide/supplement system energy at the point of consumption by capturing ambient energy in the operating environment. The justification and success of EH implementations, particularly in terms of total cost of ownership, are highly dependent on the method of calculating payback. For instance, adding $3 – $5 to a system bill of material for EH capabilities might seem crazy when comparing it to a disposable coin cell costing approximately $0.25 at volume. Even neglecting environmental and sustainability factors, there is a lot to consider in the financial analysis. If that battery ever needs to be replaced, then the labor/access-logistics costs alone can obliterate the coin cell savings by orders of magnitude — heaven forbid if that battery is in a harsh and/or inaccessible environment, such as a concrete wall, high ceiling, human body, or deep oil well.
EH entails using ambient energies that are available — heat, vibrations, light — for sources of power. There is a sweet spot, from around one microwatt to a few hundred microwatts, where there is the ‘double impact’ of significantly less drain on the existing power source and increased viability for using ambient energies from reasonably sized harvesters. This can significantly increase battery life, in some cases even leading to complete power autonomy. (This is discussed in a recent EU publication and is illustrated in Figure 1. i)
A key challenge driven by integrating EH into system design is dealing with energy sources that can be quite sporadic in nature. They need energy storage and power management devices/ circuits to capture the energy and make it available for later use. Not only are there unique engineering efforts that must be made to address power extraction from ambient scavenging, but many of those needs can be different for each method of EH. In other words, the capture of raw energy from the EH transducer and power conversion/management/regulation needs are different for photovoltaics (PV) than for thermoelectric generators (TEG) or vibrational harvesting. Even powering different flavors of PV cells, can vary greatly based on the technology. The general approach tends to be driven by the nature of the raw, harvested energy, be it DC (PV, TEG) or AC (vibrational, triboelectric, RF).
PV cells directly convert light energy from the sun and/or manmade sources, whereas a TEG extracts energy from a temperature differential to generate electrical energy. Vibrational (electrodynamic or piezoelectric) and triboelectric sources are derived from physical movement. RF capture typically involves the use of a rectifying antenna (rectenna) and balancing network, and then, as is common, feeding it into a DC/DC conversion block.
An optimal, EH-enabled system solution may require maximum power point tracking (MPPT) and/or carefully controlled impedance matching to fully realize its maximal energy potential. In addition, many ambient energies are at very low power and voltage levels. Most commercial-off-the-shelf (COTS) power management ICs (PMICs) are incapable of converting energies below 10 μW and 100 mV to usable electricity. An example of research community-driven efforts to resolve this is the MISCHIEF platform being developed by Tyndall National Institute (Cork, Ireland). MISCHIEF is an innovative high efficiency, low quiescent current PMIC capable of handling an unprecedented range of ambient energies particularly in the sub 10 μW domain that heretofore were unusable. It is modular and highly configurable so that it is easy to add new circuit blocks and/or adjust set point ranges. It also has a digital interface enabling it to interact with other components to dynamically adjust their operation mode (sleep, standby, sense, transmit, process). This minimizes their power consumption while meeting the application needs.
Energy storage is critical for intermittent energy sources, since it provides a buffer to handle peak demand, so the upstream power source only has to provide for the system steady state needs instead of worst-case, peak power demands.
Creating an EH Ecosystem
Contributors to the Power IoT and EH communities — developers, manufacturers of materials and devices, as well as installers, integrators, and end users — have tended to work in siloed environments. However, for EH to successfully have major penetration into mainstream applications, EH transducer folks will have to work closely with power management and energy storage folks, let alone with the many other low-power system component providers and end-users. This is particularly true for many of the sensor network, low power type applications this article focuses on.
In the Power Sources Manufacturers Association (PSMA) Energy Harvesting Committee (EHC), we are working to establish a coordinated ecosystem to address these gaps. This approach started with gathering stakeholders and providing educational resources such as webinars and industry events such as the Applied Power Electronics Conference (APEC), and now a dedicated biennial event, EnerHarv ii. Since the EHC is composed of many members of both the IEEE Power Electronics Society (PELS) and PSMA organizations it is ideally suited to tackle these multidimensional challenges. Led by Thomas Becker of Thobecore and Michalis Kiziroglou of Imperial College London, the PSMA EHC also has recently published informative white papers on their website iii.
Figure 4 is an illustration of the “ecosystem” needed to coordinate the separate elements of an EH powered IoT. A central core of scientific disciplines — energy harvesting, storage, micropower management and system integration — is linked to IoT applications through enabling technologies. Stakeholders such as suppliers, developers, users, and integrators, need to closely cooperate in component selection, device and system specification, redundancy, reliability, configurability, data processing needs, and various other application considerations, in order to optimize these systems. Standardization and interoperability with supporting simulation tools operating at device-level and system-level are also critical enablers underpinning this ecosystem.
IoT System Benefits of EH
The financial benefits of EH go far beyond the first-order analyses of comparing Bill of Material costs. Although, operating expenses are obviously reduced by using energy harvesting, savings in capital expenses might not be so obvious. If energy demand is reduced at the edge, where energy has the highest supply cost, then there is a rippling effect all the way upstream to the power plant. Every step in the power chain tends to add safety margins to designs, so after many system levels of critical energy storage, and utility distribution, there is an incredible amount of added “fat”. This unnecessarily increases both capital and operating expenses.
While mainly considered for high bandwidth, low latency applications, emerging 5G technology and related infrastructure, if harnessed correctly can play a major role in creating savings for low power, low data rate transfer applications iv. The greatly increased number of access points that will become available with the deployment of 5G makes for a unique opportunity to address energy efficiency optimization from the highest to the lowest levels. For instance, the standard governing what we call “5G” supports a feature called “discontinuous transmission (DTx)” for dropping data packets to save energy in the microsecond time frame. This will enable small-cell power management techniques and grid-scale energy storage to work in conjunction with energy management systems installed in all manner of sources, such as power plants and various distributed energy resources, and loads, such as local grid blocks, houses/buildings, to save energy in the seconds to minutes time frames and beyond.
Furthermore, one should consider the significant reliability benefits that come along with eliminating the need for an external power source. Electromechanical components, such as connectors, and non-rechargeable energy sources, such as primary batteries, and large energy storage elements, such as electrolytic capacitors, tend to present system reliability issues. The energy autonomy introduced by implementing EH can help minimize the use of these less-reliable components, which improves the overall reliability of the system.
The big takeaway message is that a tiny investment in system capital expenditures can lead to huge rewards in reduced capital and operating expenses — with improved system reliability as a bonus.
This article was written by Mike Hayes, Head of Group, ICT for Energy Efficiency, Tyndall National Institute (Cork, Ireland) and Brian Zahnstecher, Principal of PowerRox (San Jose, CA). For more information, contact Mr. Hayes at
iv. M. Hayes and B. Zahnstecher, "The Virtuous Circle of 5G, IoT and Energy Harvesting," in IEEE Power Electronics Magazine, vol. 8, no. 3, Sept. 2021