In 2003, there were just over 500 million internet-connected devices in the world. This represented a mere 0.08 devices per person, or just 0.02 devices for every individual with access to the internet.
By 2010 this number had risen to 12.5 billion connected objects, almost six devices for every one individual with access to the internet. Now, as we move into 2015, the number of connected ‘things’ is expected to reach 25 billion, ultimately edging towards 50 billion by the end of the decade.
This trend isn’t expected to stop there. Through the development of new connected nanotechnology, or so-called “smartdust”, this number could increase by hundreds of billions over night.
As the number of these connected objects increases, the intelligence of communication between each device is also expected to grow. As an example, a smart fire safety system is rendered more useful if it can connect with a smart thermostat. Furthermore, if this thermostat can then be connected to a user’s smart phone, an entirely new layer of intelligence is added to the system. As a result, the more objects that are added to the network, the easier it is for those objects to convert vast quantities of meaningless ‘data’ into genuinely functional information.
This is the fundamental goal of the Internet of Things: to develop a self-sustaining network of everyday objects that provides a higher collective value than the individual objects ever could on their own.
It is this idea that is currently being implemented by governments, corporate organisations, academic groups and hobbyists alike. From international IoT standards through to bespoke smart-sensor developments, the Internet of Things is expanding across all levels of the electronics community.
But despite this widespread commitment to IoT, there remain a number of complex challenges that the industry must overcome before universal adoption can be achieved. These challenges include everything from technical considerations such as universal protocols and standards, through to less tangible concepts such as the impact on personal privacy.
Until the industry addresses these challenges, adoption of IoT will remain forever on the horizon.
1. Adoption of a universal standard
While a lot of progress has been made towards a standard for IoT, far more is needed in terms of security, privacy and most importantly, architecture. According to a recent report in the Financial Times, an “intense battle” is developing between technology and telecoms groups for market domination of the Internet of Things. As the various market leaders attempt to flesh out their own space within this market, several competing standards have emerged. These include Google’s “Physical Web”, the IIC (Industrial Internet Consortium), the Open Interconnected Consortium, and “Thread”, a new IP-based wireless networking protocol pulling together support from Google, Samsung, ARM and Freescale Semiconductor.
While the work of these bodies is helping to stimulating IoT adoption, there still remains a worrying lack of standardisation across the industry. As a result, until universal standards are agreed, IoT risks becoming yet another asynchronous technology sector that relies on ad-hoc governance by the big IT companies.
2. Adoption of IPv6
Every device that connects to the internet requires its own unique numerical label – an IP address. As it stands, the vast majority of these IP addresses run on a fourth generation version of the Internet Protocol known as IPv4.
Using a 32-bit system, IPv4 offers a total of around 4.3 billion unique addresses, a figure that was originally reached in February 2010. With new devices being added to the internet every day, IPv4 is now massively overcapacity. As a result, adding new devices to the network can only be achieved through Network Address Translation (NAT) – a method of remapping one IP address into another in order to provide the illusion of additional space.
While this has proved a popular method for avoiding IP exhaustion, as the number of connected devices increases, NAT is unlikely to meet the demands of IoT.
Instead, network operators are now promoting the adoption of IPv6, a new 128-bit protocol that can offer up to 340 undecillion (340,000,000,000,000,000,000,000,000,000,000,000,000) device addresses. This new protocol will not only help to provide enough addresses to support the Internet of Things (including ‘smart dust’ applications), but will also offer improved security and network management capabilities.
Despite the various benefits of IPv6, a report from Google in December 2014 suggests that more than 94% of worldwide internet traffic is still being carried by IPv4. If IoT technology is to succeed, the rate of IPv6 adoption has to to increase.
3. Wireless protocol selection
In addition to the lack of infrastructure standardisation, design engineers are also faced with an overwhelming variety of connection technologies to include within any IoT device.
Typically these technologies include Wifi, Bluetooth, Bluetooth low energy, ANT, ZigBee and RF4CE – to name a few. In selecting the appropriate wireless technology, designers must consider numerous factors, including whether a proprietary solution is required, if an industry standard should be adopted, what the frequency band should be, and how much power the device will need. This selection will also impact how the device can communicate with other objects and could ultimately limit its adoption within the Internet of Things.
The decision to develop a proprietary solution versus leveraging a standard technology such as Bluetooth often boils down to the basic functionality required. Highly specialised functions may necessitate a proprietary implementation, which immediately limits the options available from OEM suppliers. The upside to this is that a proprietary wireless platform may be customised specifically for the application, helping to decrease the software footprint.
As is to be expected, Wi-fi currently dominates the world of wireless connectivity. However, given the power hungry nature of Wi-fi technology, it can be unsuitable for Internet of Things applications. Instead, many developers are now looking to use Low Energy Bluetooth and the Dynastream ‘ANT’ protocol. Both of these offer low-cost and ultra-low power solutions for short-range wireless communication, making them ideal for wearable and IoT devices.
4. Providing a power source
The decision to move away from Wi-Fi also raises a separate question for the adoption of IoT – how can we provide adequate power to so many portable internet-connected devices?
In order for the Internet of Things to reach its full potential, the various devices included on the network will need to become increasingly self-sustaining. As the IoT expands to include millions – if not billions – of embedded sensors, the idea of regularly changing the batteries in each device is hardly realistic. Instead, what is needed is for each individual sensor to generate electricity from its own environment, harnessing local elements such as vibrations, light and airflow.
For the most part, this is being achieved through the adoption of nanogenerator technologies. These flexible self-powered energy harvesters can be used to convert kinetic energy (created from vibrational and mechanical sources) into electrical power, removing the need of external circuits and batteries for electronic devices.
While the initial development of nanogenerators was slow, recent advancements have seen the energy efficiency of such devices increase by as much as 40 times. Now, through the work of companies such as EnOcean and Perpetuum, nanogeneration technology is being used to convert everything from subtle vibrations through to human footsteps into energy.
Despite these advances there is a long way to go before the issue of energy consumption is truly solved. As the world marches towards 50 billion connected devices, the need for flexible renewable power sources is going to become an increasingly common problem.
5. Privacy
One of the most common concerns surrounding IoT is the impact on individuals’ rights and personal privacy. Fears of invasive marketing and corporate surveillance have helped to develop a powerful lobby of misunderstanding and mistrust surrounding the Internet of Things. As such, if the industry is to encourage widespread consumer adoption, it falls to the designers to help ensure that personal privacy – and ultimately the user’s trust – is not abused.
From a design perspective, the best way to achieve this is to make sure that IoT products are genuinely delivering tangible value to people’s lives. By designing applications that are truly beneficial and not just novelty gimmicks, the IoT industry can provide end users with a trade off that genuinely works for them.
For every new technology that is introduced, the issue of privacy is always the first to be voiced by the media and ultimately the public. This has been true of everything from smartphones to satellite navigation systems. At the end of the day however, the reasons these technologies thrived is because they offered a tangible benefit while attempting to maximise security and privacy for the end user.
As a result, the data-collected from IoT devices should be made as transparent as possible, guarding people’s data with sensible controls and security measures. The basis for these controls was originally outlined in a 2009 European Commission report, which invited member states to provide guidance on the design and operation of internet connected objects. These guidelines highlighted the need for IoT devices to: be resilient to attack, use authenticated data, implement sensible access controls on the data collected, and offer a strong degree of client privacy. These are all considerations that both professional IoT designers and hobbyists must take into account when looking to develop a new connected device.
This list represents just a few of the on-going issues that will need to be overcome if IoT is to be truly adopted within 2015. While some of these obstacles fall to governments and industry bodies to address, many of them are already being faced head-on by designers and hobbyists from all around the world. This is what makes the Internet of Things such an exciting topic for those within the design community – not only anticipating the benefits, but overcoming the obstacles.
Article supplied by Farnell element14.
