Construction Corner: Algorithms could provide a new way to scan walls

The world is full of things we use every day without giving them a second thought.

Computer algorithms are one of the best modern examples. They’re everywhere.

They’re what drives search engines like Google.

You probably have a smartphone in your pocket or handbag that is loaded with apps. Those apps are simply algorithms.

Want a weather forecast? A couple of taps on your smartphone and an algorithm will provide it. Want to check your email? Again, a couple of taps.

They’re useful on the job, too.

Building an electrical transmission line? An app will tell you how much sag you should allow between towers so the line doesn’t contract in cold weather to the point where it snaps.

But what is an algorithm?

Basically, it’s just a procedure or formula for solving a problem by conducting a sequence of specified actions. That usually means that it’s a small procedure that solves a recurrent problem.

Now a research team at Duke University, in Durham, N.C., has developed an algorithm for a scanner that uses microwaves to allow people to peer through or into a wall.

I suppose that might sound silly, but you’d be surprised how often people need to know what’s inside a wall.

A construction worker might want to locate a stud, electrical conduits, wire pipes or junction boxes.

Daniel Marks, a research professor of electrical and computer engineering at Duke, says most technologies that can see through walls use a broad range of frequencies, which makes them expensive.

They also don’t have a very good resolution, “so they might be fine for seeing a person moving on the other side of a wall, they’re terrible for finding thin conduits or wires.”

Scanning systems now in use generally rely upon knowing what material the wall is made of. This allows the software to predict how the wall will affect the scanning waves so that it can separate the echoes and distortions from the solid objects the user is looking for.

So Marks and his colleagues, David Smith and Okan Yurduseven, have come up with a system that takes advantage of the wall’s symmetry instead. Since walls are usually flat and uniform in all directions, they distort waves in a symmetrical fashion.

“We wrote an algorithm that separates the data into parts — one that shows circular symmetry and another that doesn’t,” says Yurduseven. “The data that don’t have any symmetry is what we’re trying to see.”

The technique uses only a single frequency to scan because that cuts down on the number of interference patterns created by the wall. And single-frequency emitters are much cheaper than broadband emitters.

Sticking to a narrow frequency range should also simplify licensing with the U.S. Federal Communications Commission, since it would be easy to avoid interfering with microwave frequencies dedicated to other technologies, such as Wi-Fi, cellular phone service and Bluetooth.

The researchers have built a prototype device to test their work.

In their lab, they built a couple of different kinds of walls and then placed objects behind them that a worker might want to find, like studs or junction boxes.

After scanning a section of wall through gypsum board, the raw data they obtained was cluttered enough that it was difficult to make out anything other than a metal junction box.

But after analysing the data and removing the symmetrical pattern, the pictures became much clearer and each individual component was easily recognized.

Marks says the next step will be to combine the technique the team has devised with a machine vision system that someone could move over a wall to see what’s inside.

When this system finally comes to market, it will be one more tool for skilled trades to use on the job site.

All because of algorithms.

Korky Koroluk is an Ottawa-based freelance writer. Send comments to editor@journalofcommerce.com.