Sensor networks

A sensor network is a group of sensors where each sensor monitors data in a different location and sends that data to a central location for storage, viewing, and analysis.

There are many applications for sensor networks, from monitoring a single home, to the surveillance of a large city, to earthquake detection for the whole world.

The primary goal of a home security sensor network is to detect an intruder. Many different types of sensors can help collect data towards that goal, such as magnetic open sensors on doors and windows, acoustic-based glass break sensors, security cameras, and motion detectors.

All of the sensors send their data to a central system, which typically stores the data for a period of time and gives the homeowners a way to view it. More sophisticated system can analyze the data and send an alert to the homeowners when it sees enough evidence of an intrusion.

Researchers, farmers, and governments need to monitor aspects of the natural environment such as air pollution, water quality, soil conditions, and weather metrics. The traditional approach to monitoring is to collect a sample, bring it back to a lab, analyze it, and record the results. Needless to say, that approach is slow and dependent on human labor, so traditional monitoring doesn't produce a lot of data.

A more automated and scalable approach is to use a sensor network. Sensors can be distributed across an area, collect the environmental data, and send it back to a central server for processing.

In the Great Lakes of the United States, dozens of buoys use sensors to collect data about wind speed, water temperature, air temperature, and wave height. Some of the buoys are privately owned while others are supported by research groups, but they all send their data back to the same server for collection.${}^1$‍

The sensor data is shared publicly on the the Web, so anyone can visit the Great Lakes Buoy Portal and see the reported conditions. For some of the buoys on Lake Erie, you can even send a text message to the buoy and it will reply back with its latest data.

In a city, crime can happen anywhere, and with modern transportation, criminals can come from anywhere to commit crime in a city. Police officers can't be everywhere at once, so how can they track down the criminals that commit crimes?

One approach is a network of surveillance cameras throughout the city, all linked together and sending video back to the police station.

The city of Chicago, Illinois operates a network of more than 30,000 security cameras, plus additional cameras in police cars, boats, and command vans.${}^{2,3}$‍

Some of the security cameras on the streets and highways of Chicago are also equipped with automated license plate readers, technology that snaps a photo of a car's license plate and converts the photo to a textual representation. Massive databases keep track of each license plate: where it's been sighted, whether it's associated with a stolen car, and if the driver has known warrants.${}^4$‍

To help respond to gun violence, Chicago also installed a network of gunshot detectors that use both acoustic and infrared sensors to detect gun shots and triangulate their location. The sound of gunfire can travel two miles at night, so those sensors can often provide a more precise location than reports in 911 calls.${}^5$‍

Surveillance cameras, license plate readers, and gun sensors give police departments new technology to catch criminals and even prevent crime from happening.

🤔 Sensor networks in public areas bring up questions of ethics and privacy, however. What are the risks of audio and video recordings? What level of surveillance is worth the risks? How could governments ensure ethical use of sensor data?

Ever woken up in the night from your bed shaking? If you live near a fault line, your first thought might be "Earthquake!" When that happens to me, I immediately look online for reported earthquakes in my area. One way to verify my suspicions is with a search of local Twitter posts, but the more scientific way is checking data from earthquake sensor networks.

The Global Seismographic Network is made up of over 150 seismic stations, each with seismological and geophysical sensors.

The seismic stations transmit their data in real-time over the Internet so that scientists can use the data immediately to understand what's happening in the earth's crust all over the world.

Here's a recording from a seismic station in the Philippines:

See the waveform between 6:00 and 7:00? That's around the time that a 5.5 earthquake hit the coast of Indonesia, a few islands over.

Scientists don't expect laypeople to read seismographs, of course. Instead, we can check websites like USGS Latest Earthquakes which reports the magnitude, locations, and times of recent earthquakes.

Those earthquake reports are based on data flowing in from many sensors: the global network of sensors, the US-only Advanced National Seismic System, and regional sensor networks in earthquake hot spots like Alaska, Hawaii, and California.

Regional sensor networks are often quite dense. The Southern California Seismic Network collects data from hundreds of stations.

The global sensor network will capture the high-magnitude earthquakes, the kind felt from multiple countries, while regional sensors will also capture the lower-magnitude earthquakes, the kind that your neighbor might sleep through.

With thousands of sensors contributing data, computers can pinpoint the epicenter of earthquakes, predict the intensity of ground shaking in the region, and send out warnings to people and infrastructure in harm's way.${}^6$‍ A few seconds of warning may not seem like a lot, but when it's used to automatically slow down trains and shut off gas valves, those few seconds could save lives.

🙋🏽🙋🏻‍♀️🙋🏿‍♂️Do you have any questions about this topic? We'd love to answer—just ask in the questions area below!