Now that's a sound idea

The technology powering Chirp spun out from a longstanding research project carried out by some of the country's leading acoustic and computer scientists from University College London.

The project was originally conceived to overcome shortcomings in today’s networking technologies. As the IoT gathers pace, an increasing number of devices are trying to connect to the Internet and communicate with each other. However the networking technologies in common use today are not compatible with vast swathes of older devices that are still in use. Plus, there are multiple platforms and technologies used to bridge the gap between individual devices and the Internet, and they are often not interoperable or compatible with one another. So, older devices simply aren't up to the job of connecting, and newer devices struggle to talk to each other.

A good example of the interoperability problems can be seen in the payments industry. Many ATMs and till systems still run on Windows XP, sometimes with even older systems connected behind them. They simply cannot talk to growing number of modern technologies to enable seamless mobile transactions. It's a situation that is holding back the industry.

The technologies designed to overcome some of these shortcomings, like Bluetooth, work up to a point. But they require installed hardware to operate ruling out many older devices. They also highlight another challenge: friction. Devices that are trying to communicate over Bluetooth need to be connected to each other first. While this need can be overcome by other technologies, like RFID (Radio Frequency Identification) and NFC (Near Field Communication), both have relatively short range capabilities and are not widely available.

And so the Chirp team set about developing a solution to connect disparate platforms and networks seamlessly by pairing them securely. Explaining the original concept behind Chirp, Moran said: "The Chirp team realised that sound is everywhere, it's the original form of human communication. There are a plethora of devices with speakers and microphones built-in, and they all work on the same, universal, protocol; sound."

There was clearly a gap for a widely compatible, low-friction method for devices to connect to each other and to networks. Drawing on this insight, Chirp was designed as a new protocol that can share data between devices that have speakers and microphones, using just sound. While its conceptual origins do lie in the legacy modem technologies, a key difference is that the Chirp technology transfers data over the air rather than through cables.

While on the surface it may look simple, Chirp went through years of research and four years of intensive market testing with a consumer application for transferring photos, links, music and other media content using just sound.

How to Chirp
Any device with a speaker can emit a Chirp and any device with a microphone, and a relatively small amount of processing power, can decode one. Chirps can be audible or ultrasonic, depending on the equipment or application they are designed for. There are situations where the sound indicating a transfer is an advantage and others where the sound is unwanted, while some speakers and microphones are simply unable to handle the ultrasonic frequencies involved in the inaudible Chirps.

The standard Chirp technology draws on a network application programming interface (API) which essentially converts data to a ten character shortcode and an audio protocol which turns a set of characters into a series of tones. This method gives a Chirp a data capacity of 50 bits. That may not sound like much, but when you consider switching on and off a light takes just one bit, you can see there is a lot you can do with 50.

For transmission, the shortcode is embedded into a string of 20 characters, which has what's known as a 'front door' element at the beginning, to tell the receiving device that the sound is a Chirp. At the end of the string are eight 'ReedSolomon' characters to enable state-of-the-art error correction. This is a technique that is widely used by data technologies including DVDs and satellite communications.

This error correction was one of the most challenging and critical aspects of the Chirp project. The Chirps have to work in a wide variety of real-world situations and surroundings and need to be able to cope with background noise and other disruptions. The Chirp team are pretty tight-lipped when pressed about how they developed the robustness, but they do say it took a lot of work and tackling this challenge was a major part of the four-year development programme that went into Chirp. Once perfected, the error correction meant that a Chirp could be reconstructed with anything up to 25% of its content missing or misheard.

There’s an app available in the Android and iOS app stores—Chirp Share—that demonstrates the technology, it’s simple but does a good job of showing how the technology works. Pop your phone into aeroplane mode, and you can marvel as you transfer symbols between your phone and a computer—or indeed another phone running the app. Try it in a room with background noise, and you will see just how robust the technology can be.

The effort that went into developing Chirp was not all under the hood, however. The development team realised how unpleasant noises produced by the old modem technology could be. So they engineered their sounds to be more pleasing to the ear by using semitone intervals which made a Chirp sound more melodic. There were no engineering benefits to this, but the focus on aesthetics made a significant difference to the experience of the audible Chirps.

The benefits of Chirp-ing
The technology isn’t looking to replace networks; rather it has been designed to act as a universal interface - the last, or first, three feet of the connection between a device and its destination, which could be another device, system, network or the Internet.

What it does offer are many other advantages. Universal access is a significant benefit and helps to overcome the issues of enabling a broad range of new and legacy devices to communicate. There is also a lack of friction in the Chirp solution. Other technologies, like Bluetooth for example, need to be joined before they can communicate; Chirp doesn't. The sound-based approach doesn't suffer from contention issues either. These are problems experienced when too many devices are trying to access a network.

And data transfer through a Chirp can give users a far greater level of anonymity, as the lack of requirement for devices to formally join, means that no identifying information needs to be transferred for a Chirp to work.

Initially, the technology was used to launch a consumer-focused app—one that enabled people to share anything from pictures to Mp3s, across a range of devices. However, when Moran joined the team, he realised that commercialising the technology as a business-to-consumer tool would be tough. Many had tried before, and only a few survived. So the direction of the business changed to focus on business-tobusiness applications, where Chirp technology is embedded directly into customer’s products, regardless of the platform used.

In the business-to-business space the applications are endless, but, the team are approaching it along two primary paths. They work with partners on bespoke projects and implementations of the technology. A practical example of this approach has been work with the digital games company Activision. Skylanders, one of Activision's flagship games, is a character-based console game aimed at the under-tens. Within the game, players create their own unique characters, and as part of the game's ecosystem, they are encouraged to share them.

The game has around 200 million children playing it worldwide. However, due to their young average age, Activision established that a significant majority of players wouldn't be able to access data networks on their mobile devices. This barrier to communication constrained the ability to share characters and limited this function and its potential to support the game’s growth.

Sound was the perfect solution. It enabled players to transfer characters from the screen to their mobile devices without the need for any dedicated connection, complicated user accounts or data requirements. Once on the child’s mobile device—a tablet perhaps—they could then easily use another Chirp to share their character with a friend.

However, in the Activision example, 50 bits of data wasn't enough, so the Chirp team adapted the technology to be able to transfer enough data to represent the character using a custom built 1.2Kbit protocol. As a result, the technology is being used by millions of kids to enhance their enjoyment of Skylanders.

There was also an added benefit that the team didn't initially anticipate. The kids quickly realised they could share screencasts of themselves sharing their characters via Youtube, and then other users, anywhere in the world, could access the Chirp, and the character, by watching the video and capturing the Chirp. This unintended consequence took character sharing from something that was done between friends to a craze with global reach.

Moran and the team also want Chirp to be accessible, so the other path the company is following has seen it release a series of software development kits (SDKs), which enable the Chirp technology to be used with a range of software, devices and platforms that might otherwise be interoperable with the other. Technology and software developers can access the SDKs and integrate Chirp technology into their projects. As an example, the team have used a Raspberry Pi—a widely available, low-cost, programmable board—to receive Chirps and use them to control a series of colour adjustable lights.

It is a simple but effective way of showing the technology in action. But some of the other realworld applications the team have worked on have far more significant benefits.

For example, in India Chirp has worked with a bus ticketing company, Shuttl, to enable a payment and ticketing solution. Mobiles are prolific in India, but the majority are old with basic technical capabilities. The ticketing system in the region requires tickets to be purchased in advance of travel and then authorised by the bus driver. To enable electronic ticket sales a method for them to be authorised electronically was needed too. However, connecting to the network on a bus is simply too slow and not possible for every passenger. QR codes were tried but also suffered from issues with device compatibility. So, Shuttl was looking for an off-line solution, with broad compatibility, and Chirp fitted the brief. Now, millions of customers board the bus and play a Chirp from their phone; the sound tells the driver they have a ticket for travel and the ticket details are recorded.

As well as enabling the move to electronic ticket sales, the solution has significantly reduced the time it takes to get passengers on board. Alongside applications in consumer technologies, the fast growing Industrial Internet of Things (IIoT) space is seen as having a significant scope for the technology too.

There is currently an exponential growth in IIoT and related services, which means that the increase in complexity of these systems—as they have to be continuously available and highly faulttolerant—is hyper-exponential. Using sound as a complementary technology offers advantages through its simplicity, as other technologies become increasingly more complex.

If a device that may or may not otherwise be connected to a network, is broadcasting data via Chirp technology, it doesn't matter what else is going on. The network could be down, the data centre could even be on fire, but if a technician is within earshot, they can still access the data being broadcast.

There are other advantages in IIoT applications. Chirp technology is platform-independent, so the cost of joining discrete networks and systems with Chirp can be far less than other options. And sound does not cause, or suffer, from issues that can affect radio-based networks.

This last advantage is why Chirp was asked if it could help with data capture in a nuclear facility. These sites are strictly regulated on safety grounds and introducing wifi technology is a not an option. Sound, however, is inert. So Chirp was able to implement its system, ultrasound in this instance, and allow engineers to access data from sensors right across the site. As a result, an enormous amount of time has been saved on tasks that, traditionally, need to be carried out manually.

aditionally, need to be carried out manually. The team is also working with several major manufacturers of industrial robotics. Embedded Chirp technology enables sound to be used as a method to give commands to the robots. The robots can also communicate with each other or share diagnostic information, and this can be done quickly and safely in radio-sensitive environments.

The more you consider the possible applications, the more you realise that a technology that was all but written off at the dawn of the broadband era may never have been more relevant. Where else could this be the case? Is anyone up for resurrecting the fax or Telex?

Chirp has worked with Breakthrough funding, a company that helps UK SMEs achieve R&D tax credits - a government scheme created to enhance and reward innovation amongst UK businesses. Could you be eligible? Click here to learn more.

For more information, click here.