How Amsterdam crowdsourced environmental monitoring with the IoT

Amsterdam launched an IoT project to crowdsource local air quality measurement. While it faced a lot of hurdles in the project, it is implementing it on a wider scale based on the lessons learnt

It is almost impossible not to have heard about the Internet of things (IoT). Some describe it as a grand vision of billions of sensors measuring and communicating every detail of our lives. Others think about it as something that will happen in the future.

Both are correct in the sense that inexpensive sensors can measure or register aspects of our lives and, over time, these sensors will be an integral part of many devices: from appliances to doors and windows.

The rise of the IoT

Research firm Gartner describes it as - the Internet of Things (IoT) is the network of physical objects that contain embedded technology to communicate and sense or interact with their internal states or the external environment.

It is a network of technology that is able to measure something (e.g. temperature, heart rate or location and speed), communicate that measurement to another systems and/or change their own state or that of an external system. 

This is not something new, the technology has been available for many years but because of prohibitive cost and complexity, its deployment was restricted to just high-end solutions or devices.

But as the price of IoT devices goes down its mass deployment becomes a reality as people see value in making their homes smart. Furthermore, communication technology like Wi-Fi, Bluetooth and Zigbee can be embedded on the chip so devices can be smaller and operate with less power, adding to IoT’s appeal.

Today, the internet of things is closer than you might think. It is gaining momentum with projects taking place all over the world and people experimenting with it to implement pretty nifty ideas and reap benefits of a connected, smart world.

Gartner estimates that the IoT will have 26 billion units installed by 2020 and, by that time, IoT product and service suppliers will generate incremental revenue exceeding $300bn, mostly in services.

Its analysts listed IoT alongside, 3D printing, software-defined architecture and cloud computing, as one of the top ten technologies to have a significant impact on enterprises in the next three years.

To propel its entry into mainstream IT, several top hardware firms have joined forces to establish secure and reliable connectivity standards for the billions of devices expected to make up the internet of things (IoT).

The Open Interconnect Consortium (OIC) comprises Dell, Intel, Samsung, Amtel, Broadcom and Wind River, and is aimed at accelerating the development of the IoT.

Intel’s IoT lab in Swindon showcases various use cases of the IoT -  such as real-time Twitter analytics on a specific topic, and another system that displays customised, targeted adverts based on who is watching the screen without identifying personal information.

But where and how are IoT projects implemented?

Actor Michael J. Fox's charity for Parkinson's disease is seeking to improve the treatment for Parkinson's through a research project that involves IoT, wearable technologies, big data analytics and public cloud.

As part of the study, smart-watches will gather and transmit data in real time, all the time. Such data will allow researchers to look at even minute data points and analyse hundreds of readings per second from thousands of patients and attain a critical mass of data to detect patterns and make new discoveries.

Meanwhile, UK company Intouch, plans on using the Intel lab to develop IoT services that manage drainage in gullies in UK villages and avoid them from becoming flooded. 

“Floods are caused by changes in weather patterns and, as population density of UK increases, we cannot grow our infrastructure, so we must manage the existing infrastructure smartly,” said Simon Dawson from Intouch. Using devices and sensors to monitor the weather changes and the state of gullies can help local authorities avoid flood hazards, he said.

Measuring local air quality in Amsterdam – an IoT project

This year, Waag Society (an institute for art, technology and science) ran the Smart Citizen project in Amsterdam in conjunction with Amsterdam Smart City and Fablab Barcelona.

The objective of the project was to see if it would be possible for citizens to measure the air quality in their own environment using open source tools and technologies – crowdsourcing environmental monitoring.

Many people are worried about air and noise pollution. What if you could measure it yourself and compare it to other people in real time? Would that data change your behavior with regards to the environment? Can you go to your local government with the data and demand that something needs to be done?

Just a couple of questions that the project aimed at answering through the use of IoT technologies and sensor tools. Not all of them have been answered yet and actually a couple of new questions arose.

Open source hardware and software

The project used the Smart Citizen Kit developed by the Fablab Barcelona. It measures two toxic gases in the air (carbon monoxide (CO) and nitrogen dioxide (NO2)), temperature, relative humidity, light intensity, ambient noise and the number of available Wi-Fi networks. The average price of a Smart Citizen Kit is about £125 and also includes a website for data gathering and sharing, an online API (application programming interface) and a mobile app.

Image1 (Caption: Smart Citizen kit)

For the use outdoors, Waag selected housing for the sensors and modified it for the purpose of the project. 

The Smart Citizen Kit is in its early stages of development and is therefore not a turnkey solution – yet. It needs serious technological knowledge to get –and keep – it online. This is not uncommon for this kind of projects, however it required communication and education of the participants who expected a more mature product rather than a product that is being developed and improved over time.

The project started in September 2013 with a debate on the possibilities of measuring air quality by citizens themselves. An article was published in a local paper with the aim of recruiting 100 volunteers for the project. The response was bigger - about 200 people responded. 

Image2 (Caption: Smart Citizen Arduino board)

Most participants said their motivation to participate in the project was because of their concern about the air quality. Interest in seeing how technology and IoT can help in the project was the second reason cited by them.

Waag organised several so-called ‘Install Parties’ in March this year where people were instructed how to install the kit and use it. They were given a manual and had access to the helpdesk for support and by mid-April most of the participants installed the kit and data gathering would start.

Technology hurdles and sensor inadequacies

In general the technology was a stumbling block for some; especially Windows users had a hard time connecting the kit and making it work (Mac users had it easier. It was even necessary for the helpdesk to visit people in person and some of the more technology savvy participants volunteered to help others who had problems with the installation and the operation of the sensors. A number of boards (13%) broke or malfunctioned and had to be replaced.  

The Smart Citizen Kit (1.1) supported the older Wi-Fi standards (802.11b/g) rather than the newer one (802.11/n) and the Wi-Fi module had quite a short range of operation.

During the project, it was found that the sensors that measure air quality (CO and NO2) were not suited for measuring out-door air quality accurately. The problem was the accuracy and the range of the sensors. Sensors were good at measuring extreme pollution (100-1000 times the normal pollution) but could not measure the normal values in detail. Light and noise sensors worked well, but most of the participants were interested in air quality than rather than noise levels. 

Next generation sensors from IoT service providers are likely to be better and more suitable for measuring air quality by citizens. 

Another learning was that the placement of the kit made a huge difference in how accurately these sensors measured the air quality. In hindsight it does make sense: if you do one measurement at one location in the shadow and at another location in the sun you’re going to get different results. The sunlight meant that the heat was trapped in the houses, influencing both measurement and operation of the sensor. 

It also taught that the calibration of sensors is important, even high-end measurement equipment needs to be calibrated on a regular basis. If this is not done then data is hard to trust and it is therefore more or less worthless because you cannot draw conclusions from it. 

Data analytics

After the sensors collected the data, one of the challenges of the project was to use the data to get insights and improve the air quality in the city. For this it was important that the participants had a platform to share and compare data they had gathered. But the website was not able to accommodate this sense of community - everyone was just a dot on the map rather than a person doing measurements. But the meetings and lectures during the project allowed users to share findings and concerns.  

Image3 (Caption: Smart Citizen Website)

Data was also shared with expert institutions that could measure and analyse data. Combining data from different sources helped in understanding data and painted a better picture. 

The project and the data insights had an immediate impact - participants changed their routes in the city to avoid heavily polluted areas. Some participants used the data they gathered to start a dialogue with the city about noise pollution while others decided to use the bike more often. 

Lessons learnt from the IoT project

The Smart Citizen project was an experiment to see whether citizens can measure pollution in their environment. 

Here are some of the lessons learnt:

  • Select the right technology that is fit for purpose;
  • Technology should be foolproof and easy to install and operate;
  • Make Social interaction a key element in the project;
  • Combine citizen data with measurements / data from organizations;
  • Create a community. share information and experiences;
  • Make sure data is valid and trustworthy;
  • Educate participants on the meaning of data;
  • Do research into how measurement and knowledge changes behavior.

Next steps

As the pilot was successful and taught useful lessons on how best to use IoT services, The Amsterdam environmental monitoring project is going to be continued on a larger scale and with better, and heterogeneous technology. Waag and Amsterdam Smart City are preparing a new project by implementing the learning to improve the city’s air quality. 

Rob Blaauboer is a senior business consultant and a blogger. A regular contributor to Computer Weekly, he has a passion for innovation, gadgets, technology and new business models. He can be followed on Twitter at @robbla

Read more on Internet infrastructure