A design team in the U.K. has showed off a model of a bridge that uses a box roughly the size of a pack of cigarettes (pictured) and a series of sensors to communicate the stresses generated by trains passing overhead. The idea is that this model setup could offer governments and structural engineers a way to see weak spots in a bridge before it fails. It’s both an example of a new wireless protocol and a model for the internet of things.
The demo works via a machine-to-machine-style network using the white spaces spectrum in between TV channels. The technology is called Weightless, and was devised by chip company Neul and is supported by Google, ARM, CSR and others. The idea behind the Weightless standard is to create a machine-to-machine network analogous to Wi-Fi or Bluetooth for the internet of things.
The wireless signals would be able to travel longer distances than both Bluetooth or Wi-Fi making it ideal for large-scale deployments of sensors. It would also use unlicensed spectrum so people using the technology don’t have to pay data fees to carriers — although someone would have to pay for backhaul to the broader internet somehow. The bridge demonstration ,which used a model bridge, was the first public demonstration of the weightless technology in action.
From the Weightless site:
In the demonstration, a model bridge equipped with strain gauges sensing displacement in the bridge structure are connected to a pre-qualified Weightless terminal device measuring 115mm x 60mm x 30mm built into the bridge structure. This communicates over a Weightless air interface to a base station that could be sited several kilometres from the bridge. In this demonstration a user application processes the data to determine when the parameters measured by the sensors integrated into the bridge structure fall outside of expected values to warn of potential structural problems such as metal fatigue and degradation.
This is a perfect example of how people want the internet of things to work. Connected sensors relay data to a base station that then send the data someplace where it can be analyzed and then offer an alarm if something is wrong. The idea is simple, but in practice it’s hard. For example, Weightless wants to be a wireless standard for shipping data over longer distances, but that means the radios have to be cheap. Cheap radios tend to come when something is popular. But not every country has implemented white spaces standards. The U.K. is ahead of the U.S., and in the U.S. it’s unclear how widespread the spectrum will be.
Thus, people aren’t sure they want to invest in chips for the standard. That’s only the wireless aspect. There are also questions about the sensors in terms of the characteristics they should have a and how they record data. Will that data be in a proprietary format? When it comes to analyzing the data it’s unclear where it should happen. Will every bridge or tunnel or street have a nearby base station that can provide the compute? That’s cheaper and maybe faster than sending all of the data up to the cloud, but then the raw data isn’t available for other applications to use it in real time for an entirely different purpose.
That’s an architecture decision that engineers and municipal planners will have to think about. But there’s also a larger social contract written when we add intelligence to our infrastructure. If the bridge shows a stress, does the government shut it down? We’re going to suddenly start collecting a lot of new data and try to draw conclusions about it without actually knowing how or when a problem could arise. Maybe you can say that given the data, the bridge should shut down on days when wind speeds hit 50 mph or more.
So while this model bridge is the first demonstration of a specific type of wireless protocol in action, it’s also a good time to start thinking about how we’re going to build self-aware bridges, and how that knowledge will change our approach to safety and even allocating transportation spending. As my colleague Om points out, as we bring more devices, infrastructure and data online, we’re going to cause massive shifts in society.