Almost 80% of industrial and municipal wastewater is discharged untreated into freshwater sources. As fertilizer usage has increased to 22 million tons a year, pollution from agricultural runoffs has worsened as well. It is important that we are able to monitor the quality of our freshwater supplies. Our proposed solution will provide effective monitoring of major waterways and report on their water quality at low cost and power.

Sensor Network System

The system consists of two main parts: a sensor node network that is responsible for monitoring the water quality, and a host that communicates with the sensor network and is capable of long-range communication to send the data to the user.

Sensor Nodes

The individual sensor nodes consist of:

1. ESP8266 Module
2. Battery
3. pH Sensor
4. Turbidity Sensor
5. Conductivity Sensor

Different or additional sensors can be added as well depending on the use case. The various sensors will allow the sensor node to capture the acidity, turbidity, and contaminant level of the surrounding water. The sensors will be attached to the ESP8266 through a shield. The ESP8266 will poll the sensors at regular time intervals and publish the data to the host. Between data collection and publishing, the ESP8266 will go into deep sleep to reduce power usage.


The host, a Raspberry Pi, will communicate with the individual sensor nodes over WiFi, and with the user through a long-range communication method. As it consumes higher power due to the need for long range communication, it can be fitted with a solar panel. As the ESP8266 has a normal working range of ~300m, the host can easily communicate with any sensor nodes within the 300m range, allowing it to monitor a wide area of water. Any sensor nodes outside the range will pass its data to the nearest sensor node, creating an information cascade to a sensor node that is within range, which will pass on data from all the out of range nodes to the host. As the monitored waterway may not have WiFi coverage, we propose the following communication methods:

1. GSM/3G/4G/LTE
2. Radio
3. LoRaWAN

As the Raspberry Pi is capable of interfacing with modules to allow it to communicate using any of the above 3 methods, depending on the location of the water body, the most cost-effective method will be used.

The hosts will then update a remote server with the data collected from their sensor node networks. The software on the remote server will then collate the data and present it to the user. Data analytics and deep learning can also be used to predict future water quality and determine the best action to be taken.


Our system is innovative in the host and network system, as well as the modular design. Most remote monitoring systems consist of one module that is capable of measuring water quality and sending the data over long distance to the user, at a higher cost. Our system offloads the need for long-range communication from each sensor node to the host, allowing for the individual sensor nodes to be kept as low cost as possible. With this method, the overall network can monitor an entire stretch of river with just one host and many low cost sensor nodes, compared to conventional monitoring systems which will need multiple modules with a cost equivalent to the host of our system.

A modular design for both the host and sensor nodes allows user to choose which sensors and communication methods they prefer. The host software can easily be configured to add new sensor nodes, allowing for simple deployment of new sensor nodes if the need arises.


A shield will be used to interface the various sensors with the ESP8266. As the various sensors are easily obtained off the shelf, the shield will be made to interface with the most common models. The sensors simply need to be plugged into the shield, which will be attached to the ESP8266. This simple and modular design keeps the cost low, and any faulty sensors or ESP8266 modules can easily be swapped out. By using commercial off the shelf products and developing ways to interface them with our system, this keeps the cost low as we do not need to develop in house methods to manufacture these parts. Weatherproof casings for the host and individual sensor nodes can easily be 3D printed.


Our sensor network system is designed to be low cost yet capable of monitoring a wide area of water. With the low cost of the sensor nodes, many can be deployed to pinpoint pollution sources, and track them closely over time. The customizable long-range communication methods give greater flexibility allowing the sensor network system to be deployed anywhere around the world. With this, we can provide close monitoring of the world’s major waterways at low cost, allowing us to design better systems and policies to maintain water quality, and for all to enjoy safer water consumption.



Singapore is a water scarce country, and it is important to us that we are able to monitor our own freshwater supplies as well. The smart water challenge gives us the opportunity to create a system that can impact lives around the world including our own. Also, we love Keysight and Electroboom. We hope that we can meet Electroboom in person, as well as win awesome Keysight equipment for our new school building to foster Makers in our school


Voting is closed!