Anders Andrae was recently quoted in the Guardian as, “We have a tsunami of data approaching. Everything which can be digitalised. It is a perfect storm. 5G is coming, IP traffic is much higher than estimated, and all cars and machines, robots, and artificial intelligence are being digitised, producing huge amounts of data which is stored in data centres. ‘Tsunami of data’ could consume one fifth of global electricity by 2025.” This statement summarizes the purpose of this blog.
Fig. 1: A Representational Image Showing Array of IoT Devices
No doubt, IoT and smart grid will provide countless energy saving opportunities. The astronomical data yielded by IoT industry will play a key role in designing and implementation of new energy saving technologies and bring down peak load off the grids. But what is being overlooked is the huge energy footprint that will be left by the IoT itself. Cisco predicts that the total number of IoT devices will sum up to 50 billion by next year and the energy consumed by these devices will impact the global energy demands adversely.
How IoT is becoming the energy culprit?
While some of the major applications like smart home use mainline power, there are several others that need rechargeable battery support. Designers as well as engineers need to formulate energy budgets that keep account of the major energy consuming IoT centers. For now, there are three major ways through which energy is being devoured by IoT devices:
- Data Centers: Data collection in IoT world takes place through sensors, surveillance cameras, machines and appliances that are present all around us. It can either be chips, driverless cars, enhanced humans, semiconductors, augmented reality, virtual reality, and so forth. All the data collected by these mediums is stored and analyzed in data centers that are already culprits of consuming mega amounts of energy. Increased demand for IoT will only increase this consumption.
Fig. 2: Picture Showing a Typical Data Center
- Embodied Energy: Digital technologies need much more power in manufacturing processes as their manual counterparts. New IoT technologies will put up a roaring demand with billions of sensor nodes, microchips and countless semiconductors. Since, every person is expected to have 5-6 wireless devices, it will amount to inordinate amount of power requirement just for manufacturing IoT products.
- M2M Communications: Machine to machine communications include data back ups, remote software updates, media backups, and data transmissions from IoT devices. For now, most M2M communications consume very less energy, but data intensive products like wearable medical devices and driverless cars will devour large chunks of energy.
Fig. 3: Image Showing Standard M2M Communication
What Is The Solution?
Every mass movement begins from small innovative ideas. So, we will need to start working from the base. Smart thinkers and innovators need to work on energy efficient IoT devices that can sustain on renewable forms of energy and have a extended lifetime. On an individual level, you will need to work on design, component, and connectivity level to solve the problem. Here are a few pointers you may want to take note of:
- MCU Selection: Whenever you are working with small devices, ensure that the MCU you select is highly energy efficient. Your computational requirements may compel you to go for a 8-bit or a 32-bit MCU but make sure you keep energy requirement as a major parameter for selection. If the device is able to quickly switch from passive to full-speed functional mode, it will conserve battery power to quite an extent. See that the MCU is able to do as much as possible without leveraging the CPU core.
Fig. 4: Image Representing MCU Selection for IoT Devices
- Sensor Interface: Considerable portions of energy can be saved via autonomous functionality of sensor and other peripheral interfaces.Battery life can be extended to maximum if the system is able to produce stimulus signal for the sensor from MCU, get back the results, analyze and disturb the MCU only when useful information is obtained.
- Network Topology: Network topology and protocol choice will have a significant part in power savings with respect to wireless connections. Try to use the combination of proprietary sub-GHz protocol and point-to-point link as far as possible to keep power requirements in control. But it may not work all the time since it restricts the scope of sensor deployment.
Fig. 5: Network Topologies Recommended for IoT Systems
- Configuration: An important elements, you can always go for either 2.4 GHZ or su-GHz technologies to enhance the system ability for multiple sensor deployment. You may need a complex protocol for this purpose but will be worth it. Mesh configuration is another good option, it pairs up perfectly with ZigBee protocol. Apart from great flexibility, it also offers a stable deployment alternative with a self-healing network.
- Self-Empowerment: One of the biggest and best design considerations, this one resolves the power consumption problem to quite an extent. Based on the kind of battery being used in application, choose the boost-converter regulator carefully. It will have a big impact on overall power consumption for IoT devices falling in range of 1uA – 7 uA power consumption.
- PMIC: A power management integrated circuit will always work in your favour. Not only will it help in power management but will also provide you with sharp control on complete system. Here a single power source will help in generation of multiple voltage rails to execute various aspects of embedded systems optimizing every voltage rail to offer sufficient power for a given application.
- Radio Selection: While most creators will make a radio based on application requirements, it is better to go for BLE radios for low-powered, small sized devices. Wi-Fi devices will need Wi-FI network with router and cellular devices will be free from any such constraints. But as you move from Bluetooth to Wi-Fi and cellular remember the energy requirement will go high exponentially.
- Bluetooth: BLE (Bluetooth Low Energy) is one of the least power consuming wireless technologies that will do wonders for IoT devices. It can be kept in a passive mode most of the time and will wake up the device at the time of usage and send a message. Best part, it can be used for both smartphone as well as PC.
Fig. 6: BLE Enabled Development Board by ARM
Irrespective of the design and connectivity modifications, power management wll be a big reason to worry for the IoT world in coming years. But as is the human nature, we will get through it and create a more self-reliable and sustainable world with smarter and more power-optimized devices. Radios are one of the keys to optimize power requirements of IoT systems. Combine this with more logical component and design selection and you will be able to navigate through tough phase of product development.