Keeping the IoT green by reducing e-waste

Environmental caveat: e-waste

New technologies are fighting an uphill battle to reduce the environmental decay from century-old processes that are both inefficient and wasteful. The “internet of things” has the potential to reduce emissions from those processes and create a truly green earth. But while green use cases like water monitoring, smart grid and metering are all well known within enterprise IoT, the concern for the e-waste impact of IoT itself is less of a focus.

We have all read the projections of tens of billions of connected devices in the next few years, but how do we account for the production of those devices that require connectivity and energy? What happens when a device fails, or needs an update that cannot be done over the air?

Source: Tom Mills/Youtube

IoT can do a lot for our environment, but at the expense of a potentially significant increase in e-waste and energy usage.

Defining green IoT

This is why “green IoT” is not just how the internet of things is helping reduce the greenhouse effect within other industries, but also reducing the effect that IoT itself could have on the environment.

According to an Institute of Electrical and Electronics Engineers report titled Green Internet of Things for Smart World, green IoT can be defined as:

‘‘The energy efficient procedures (hardware or software) adopted by IoT either to facilitate reducing the greenhouse effect of existing applications and services or to reduce the impact of greenhouse effect of IoT itself. In the earlier case, the use of IoT will help reduce the greenhouse effect, whereas in the later case further optimization of IoT greenhouse footprint will be taken care. The entire life cycle of green IoT should focus on green design, green production, green utilization and finally green disposal/recycling to have no or very small impact on the environment.’’

How to make the IoT green

Here are some ways to reduce IoT’s own footprint, in order to minimize waste and power consumption, according to the IEEE:

• RFID
  • Reducing the sizes of [radio-frequency identification] tags should be considered to decrease the amount of nondegradable material used in their manufacturing (e.g., biodegradable RFID tags, printable RFID tags, paper-based RFID tags), because the tags themselves are difficult to recycle generally; and
  • Energy-efficient algorithms and protocols should be used to optimize tag estimation, adjust transmission power level dynamically, avoid tag collision, avoid overhearing, etc.

• Wireless Sensor Network
  • Make sensor nodes only work when necessary, while spending the rest of their lifetime in a sleep mode to save energy consumption;
  • Energy depletion (e.g., wireless charging, utilizing energy harvesting mechanisms which generate power from the environment (e.g., sun, kinetic energy, vibration, temperature differentials, etc.));
  • Radio optimization techniques (e.g., transmission power control, modulation optimization, cooperative communication, directional antennas, energy-efficient cognitive radio);
  • Data reduction mechanisms (e.g., aggregation, adaptive sampling, compression, network coding); and
  • Energy-efficient routing techniques (e.g., cluster architectures, energy as a routing metric, multipath routing, relay node placement, node mobility).

Source: Temboo

• Cloud computing
  • Adoption of hardware and software that decrease energy consumption. In this regard, hardware solutions should target at designing and manufacturing devices that consume less energy. Software solutions should try to offer efficient software designs consuming less energy with minimum resource utilization;
  • Power-saving virtual machine techniques;
  • Various energy-efficient resource allocation mechanisms like auction-based resource allocation, gossip-based resource allocation and related task scheduling mechanisms;
  • Effective and accurate models and evaluation approaches regarding energy-saving policies; and
  • Green CC schemes based on cloud supporting technologies (networks, communications, etc.).

• Machine to machine
  • Intelligently adjust the transmission power to the minimal necessary level;
  • Design efficient communication protocols with the application of algorithmic and distributed computing techniques; and
  • Activity scheduling in which the objective is to switch some nodes to low-power operation “sleeping” mode so that only a subset of connected nodes remain active while keeping the functionality (e.g., data gathering) of the original network.

The IEEE provides guidelines on keeping IoT communications technologies green:

1. Turn off facilities that are not needed. If the facilities are always working, it will consume much energy. However, if the facilities are only turned on when necessary, the energy consumption will be reduced. For example, sleep scheduling is one of the widely used techniques for saving the energy consumption in wireless sensor networks, by making sensor nodes dynamically awake and asleep.
2. Send only data that are needed. Data (e.g., large-sized multimedia data) transmission consumes quite a lot of energy. Sending only the data that are needed by users, can significantly reduce the energy consumption. Predictive data delivery based on user behavior analysis, is one possible method to provide only required data to users.
3. Minimize length of data path. This also is a straightforward method to reduce energy consumption. Routing schemes considering the length of chosen data path could be energy efficient. In addition, network working mechanisms, which cater to the routing requirement, are also potential ways to achieve a much shorter data path.
4. Minimize length of wireless data path. Regarding minimizing length of wireless data path, energy-efficient architectural designs for wireless communication systems could be considered. Moreover, cooperative relaying for wireless communications also is promising in energy efficiency, by using relay nodes to overhear the transmission and relay the signal to the destination node, resulting in significant diversity gains.
5. Trade off processing for communications. Combining data from multiple sources, data fusion decreases the transmissions of similar data values, while transmitting more accurate data. Thus the energy efficiency is improved. As a new way of sensing the signal with a much lower number of linear measurements provided that the underlying signal is sparse, compressive sensing also is able to enhance energy efficiency.
6. Advanced communication techniques. Toward green communications, advanced communication techniques are emerging. For example, employed at both the transmitter and receiver, multiple-input multiple-output communication techniques demonstrate improved spectral efficiencies in multipath fading environments, relative to their single-input single-output counterparts. In addition, a cognitive-radio system, which is aware of its environment and can change its modes of operation (operating frequency, modulation scheme, waveform, transmitting power, etc.) via software and hardware manipulation, is able to improve spectrum-usage efficiency and minimize the problem of spectrum overcrowdedness.
7. Renewable green power sources. Different from traditional resources, a renewable resource (e.g., oxygen, fresh water, solar energy, timber and biomass) is a resource that is replaced naturally and can be utilized again. Therefore, utilizing renewable green power sources will have a fundamental impact on minimizing the dependence on oil and the emission of carbon dioxide.