Sustainable Solutions for 5G’s Influx of IoT: Low Power Operations in IoT Devices
With bandwidth demand for video conferencing and streaming video skyrocketing and the reality of more than 15 billion connected IoT devices worldwide doubling the number of mobile phones, the mobile industry is abuzz with encouragement to consumers to upgrade their devices and service subscriptions to 5G. With network speeds up to 100 times faster than 4G and the ability to support 500 times the number of connected devices per square kilometer compared to 4G, there are obvious benefits for both consumers and operators to transition to 5G.
However, various studies and analyses have raised concerns over the increased power consumption of 5G networks and devices compared to 4G, and the negative environmental impact this may have.
Understanding these dynamics, the engineers at InterDigital have been leading research and defining innovative solutions to mitigate and reduce power consumption of 5G networks and devices. From power optimizations that can be enabled in the lower-layer access technologies used to manage connectivity and transmit and receive data between devices and the network, to higher-layer services that can be deployed in 5G devices and distributed throughout cloud and edge compute functions in the 5G network, InterDigital’s research has focused on several aspects of the overall 5G system.
One major focus area of InterDigital’s research has involved connected IoT devices. When it comes to optimizing power consumption, connected IoT devices have a unique set of technical challenges that differ from smartphones. For this reason, along with the fact that connected IoT devices already outnumber smartphones 2 to 1 (a gap that is expected to widen to 4 to 1 by 2025), optimizing power consumption for this category of device has been a top priority for InterDigital.
Power Optimization for IoT
IoT devices differ from smart phones in a few key ways. IoT devices typically handle comparatively small amounts of data and are instead optimized to prioritize battery life. Battery-powered IoT devices are required to remain operational for long durations (e.g. a decade or more) without recharging or replacing their batteries. They often have a much smaller battery or no battery at all and instead rely on harvesting ambient energy such as solar or vibrations. As such, to conserve their batteries, many of these devices need to remain disconnected and completely powered off for long durations until they are awoken to perform an operation, for instance a sensor that wakes-up monthly to report a reading and then returns to sleep. Other devices must minimize the amount of circuitry they leave powered on so they are still able to be remotely triggered to fully power-up if/when needed, for example an actuator that must be sent a remote command to close a critical valve in a timely manner. Unlike smartphones which can rely on users for daily recharging, and/or transitioning in and out of low power modes when needed(e.g., a user enabling/disabling airplane mode to turn off/on cellular, Wi-Fi or Bluetooth radios to extend the life of the battery), IoT devices typically do not interact with users. Instead, these connected IoT devices must rely on more autonomous forms of management and control within the devices themselves or the networks to which they connect.
At InterDigital, our research teams have been working on several key technologies over the past decade to help optimize the power consumption of the 5G system with respect to IoT devices. The following are a few examples of foundational technologies developed by innovative InterDigital engineers in collaboration with our industry and academic partners and contributed to standards bodies such as 3GPP, oneM2M, and IETF so that the 5G ecosystem may benefit from them.
IoT Device Triggering
Device triggering technology enables IoT devices to “sleep” for long periods of time when not in use, and intelligently “wake up” when needed. Specifically, IoT devices register to a service in the network and provide instructions for triggering the device. The instructions can include a device’s contact information as well as its sleep schedule. Once registered, the IoT device powers down to sleep to conserve power. Based on its sleep schedule, the device can periodically wake up for a short duration of time to listen for triggers by optimally enabling only a small, select portion of its circuitry required to receive a trigger, but otherwise, the device consumes a relatively small amount of power. If the service in the network needs to communicate with the device (e.g., send it a command), it schedules a trigger to be sent to the device during a scheduled wake up time. If the device receives a trigger, it fully powers-up, connects to the network and communicates with, or receives a command from, the service. Afterwards, the device powers back down and goes back to sleep to continue conserving its power. To provide even further power savings and the ability to trigger an IoT device in a more on-demand manner without having to wait for a scheduled wake-up time, ultra-low power wake-up receiver technology can also be used. This technology is embedded within an IoT device and can receive a low power wake-up signal. The ultra-low power wake-up receiver circuitry does not draw power from the battery of the IoT device. Instead, the receiver is passive in nature and is powered by the energy captured from the radio waves used to transmit the wake-up signal to the IoT device. This solution is analogous to other forms of passive receiver technologies like radio frequency identification (RFID) used in applications like electronic toll collection, however, this ultra-low power wake-up receiver technology operates in longer-range wide-area cellular networks.
IoT Device Twins
Device twin technology enables physical IoT devices deployed in the field to be represented by digital counterparts that reside within services hosted in the cloud or on network edge compute nodes. These digital counterparts, or twins, are then used to exchange information between the devices themselves and the applications and services that wish to interact with them. For example, a sensor can publish readings to its digital device twin. Use of device twin technology can provide a huge power consumption benefit for IoT devices because device twins allow IoT devices to disconnect, power down and sleep for longer durations of time without impacting the functionality of applications, services and other devices that require data from these IoT devices. Even while IoT devices are sleeping, their data can be accessed from their device twin.
Device twin technology also reduces the number of requests IoT devices must send and receive, since they only need to interact directly with their device twins instead of other applications, services, or devices. An IoT device is only required to wake-up long enough to send an update to its device twin, and then can immediately go back to sleep. Even an actuator-type device, which requires receiving commands, can benefit from device twins because actuator commands can be stored within the digital twin. The IoT device can then wake-up based on its own schedule, check its device twin for any new commands, perform these commands, update its device twin with the results, and then return to sleep.
IoT Message Profiles
Many IoT devices generate a stream of sensor readings on a periodic schedule or event basis. The number of bytes of data included within each message the device sends or receives can have a major impact on the battery life of the device. The larger the message sent or received over the network, the longer that device must remain powered to send the message. Over time, this can add up and dramatically reduce the lifetime of the device battery.
In many IoT use cases, there is a certain amount of information contained within these messages that is static in nature and does not change – for example, certain fields within upper layer protocol headers of the message or in the data payload of the message. This static information is critical to processing the message and provides critical context to the services and applications, like analytics, consuming and processing the message. However, this static information can result in significant overhead on the device and the network if included in every message the device sends. To alleviate this, IoT message profiles can be used. Within a profile, static information elements can be defined along with criteria defining which device(s) and/or message type(s) a profile is applicable to. IoT message profiles can be configured within the network services to which an IoT device sends it messages, like a cloud or edge data service. As the service receives messages, it compares the messages against the profiles and if a match is found, the messages are enriched with the static information defined in the profiles and further processed by the service or applications (e.g., analytics operations are performed). This allows devices to send only the bare minimum amount of information that has dynamically changed per message (e.g., a sensor reading) and streamline the size of the message. IoT message profiles can also be configured on IoT devices themselves and used in a similar manner by the devices to enrich the messages they receive with static information. Therefore, the use of IoT message profile technology helps minimize the size of messages flowing between IoT devices and services in the network, which in turn translates into less energy consumed by the IoT devices as well as reduced load and energy consumption by the network infrastructure used by the IoT devices.
Offloading IoT Event Processing
Offloading event detection and processing from user applications to services deployed in the cloud and on edge compute nodes in the network reduces the overall number of messages exchanged over the network. Rather than an application retrieving every sensor reading published by the device(s) it is interested in and checking whether sensor readings have crossed a certain threshold value of interest, the application instead offloads an event detection rule to a service in the network. Within the rule, the application specifies the devices it is interested in, the data of interest from these devices, conditions of interest for this data, and actions that are to be performed if/when these conditions are detected. For example, if pressure within a specified boiler device exceeds a specified threshold, then power to the boiler is turned off. Using the rule, the service can efficiently offload and perform this work on behalf of the application.
Offloading eliminates the need to send individual copies of device data to each application to process, which greatly reduces the amount of messaging in the system. It also enables the pooling and reuse of compute and storage resources of the service by the IoT devices and applications in the system. For example, data offloaded by IoT devices to a service in the network can be used both for the monitoring of a single condition defined in a single rule from a single application, and also simultaneously used by the service to monitor many conditions defined in many rules from many applications. As result, offloading IoT event processing can play a major role in helping optimize energy consumption in the overall 5G system.
Radio Frequency (RF) Energy Harvesting
Many IoT devices, given their minimal power requirements, are ideal candidates for leveraging energy harvesting technologies such as solar, thermal, wind, and vibrations. Use of these different technologies however is often very use case-specific and dependent upon the availability and suitability of the energy source (e.g., solar is best suited for outdoor applications). One promising technology that InterDigital engineers have been developing enables energy to be harvested from cellular radio signals. This technology enables IoT devices to harvest energy from the very same cellular network signals they use to communicate with cell towers. Given the widespread deployment and coverage of cellular networks, this form of energy harvesting could make a huge impact and further complement existing energy harvesting technologies to revolutionize how connected IoT devices are powered. This could not only help charge and extend the battery life of these devices, but one day it may eliminate the need for batteries in these devices altogether. This could allow devices to be deployed in places that are not yet practical today, like those implanted within infrastructure like concrete walls and bridges and eliminate the need to replace batteries every five to ten years.
As the mass scale out of connected IoT devices within the 5G ecosystem continues, the importance of using and responsibly deploying the technologies described above will become critical. This will help reduce the power consumption of 5G systems and make these deployments more sustainable. For this reason, InterDigital is committed and actively engaged in forward looking research to identify technologies that can further optimize the sustainability of 5G and future 6G deployments.
InterDigital’s sustainability microsite includes the latest research and white papers exploring the impact of energy sustainability in the evolution of advanced wireless, video, IoT proliferation, and more. Learn more about InterDigital’s sustainability thought leadership and efforts to identify innovative solutions to address energy consumption and environmental sustainability across all sectors of the technology ecosystem, here.