Yue’s Yappings

RFID has been one of the most hyped technologies over the past decade. Study after study has breathlessly praised its benefits and predicted an industry worth billions upon billions of dollars. And on the contrarian side, we’ve heard ghastly visions of living in an Orwellian surveillance society.

RFID is a vibrant, steadily growing market. But its not exploding, as you would expect from all the articles in the main stream press. And that has left many people disappointed. Thousands of companies jumped into the market with dreams of riches. Today, many are struggling to survive, including tag suppliers, reader suppliers and even system integrators.

So let’s take a look at what’s holding the market back - I think it a good example of how hard it is for any new technology to establish itself.

What is RFID?

Let’s start with the basics - what is RFID? Its a technology that consists of two parts - a tag and a reader. Tags are attached to anything you want to track - clothing, boxes, animals and even humans. Each tag has a tiny chip that encodes information about what is being tracked. Usually product ids are stored on the chip - when the tag is detected the id is read from the chip and looked up in a database.

Which brings us to the second part of RFID - readers. Readers are devices that detect nearby chips and read their embedded information. Depending on various technical factors, nearby can be anywhere from a few inches to 60 feet.

Why is this Useful?

The basic idea of RFID is to replace barcodes. For all their utility, barcodes suffer from one big disadvantage, they have to be read by a laser. If you’ve ever used a self-check out line at the grocery store you know what a pain it is to search for the barcode and line it up just right with the laser. Wouldn’t it be nicer if you could just walk up to the check out machine, press a button, and have it recognize all the groceries in your bag without you having to remove them?

Although you won’t see RFID used at you local grocer (at least not yet), I bet you still use it almost every day. Do you have an E-ZPass for local toll roads? That’s RFID. Or maybe you have a security card to get into your building each morning? RFID again.

Or maybe your new Credit Card offers an “Express Pay” service? If you look closely at it, you’ll see an RFID chip embedded in it. All you have to do is touch the card against a reader, and voila, you’re done.

Or maybe you’ve just renewed your passport. If you look closely, it now also has a RFID tag.

In the business world, WallMart uses RFID to track its supplier’s deliveries, resulting in a more efficient supply chain. The US Military does the same. In China, employees in warehouses use handheld reader device to track parcels as they leave major Chinese ports on their way to the US or Europe.

With just a few seconds of thought, you can probably think of a dozen new ways of using RFID. Maybe you want to track your pets, or find your keys, or have your computer recognize you when you walk up to it.

A Slow Growing Market

If you think about the successful applications I mentioned above, they are all closed loop systems. One company controls both the tags and readers, and the communications between the tag and reader are kept simple. Once you start involving multiple products from multiple companies, or try to hook into backend systems, things start to get very complicated.

A Standards Mess

One reason is too many standards. There are multiple RFID standards, with ISO and EPC dominating. In addition, some countries, such as Japan, have their own standards, and others, like China, are developing their own standards.

The next problem is constant fights over frequencies. RFID can work at low frequency (LF), high frequency (HF) and Ultra High frequency (UHF). Frequencies vary from country to country, and are almost always politically contentious.

For example, in the United States, the Federal Communication Commission (FCC) specifies that UHF readers must communicate with tags between 902MHz and 928 MHz. In Europe, the regulatory agency is called ETSI, and is specifies that communication must happen between 865 to 868 MHz.

In additional to different operating bands in each country, there are also different rules on how readers should interact with tags and strict limits on spurious emissions outside of approved bands (some RFID bands are also used by cellphones).

To reliably interoperate, RFID tags and readers must be compliant with all of these standards! Needles to say, that a burdensome task, particularly for small companies. In addition, you can’t sell your product in a given country until its regulatory agency gives you its stamp of approval. And getting such a stamp is a long, expensive process.

Technical Problems

In addition to complex standards and regional regulatory constraints, RFID also suffers from a number of technical issues.

Tags, especially UHF tags, are extremely sensitive to the environment. They don’t work well around metal and water. This is problematic because most buildings contain lots of metal and people and animals are mostly made up of water. HF and LF tags work better with metal and water, but they have limited read range of only a few inches.

Security is also another big concern. Since anyone in range can talk to an RFID tag, how do you hide secret information and make sure it hasn’t been altered.

Patents
Patents are meant to encourage innovation. Ironically, in the RFID world, they do just the opposite. Patents have proved to be a huge impediment to the deployment of RFID systems. Just a few large corporations control most of the patents in the field. Which means if you are a small RFID startup, you have to pay a large royalty fee up-front.

Price
Finally, RFID is still too expensive. Barcodes are cheap. Really cheap. And in many industries, where profit margins are razor-thin, RFID tags simply aren’t yet affordable. Solving this problem is just a matter of time - although when you read some industry expert’s guess on how much longer it will take to reach price parity with barcodes make sure to multiply it by at least two and maybe five. Anyway, is certainly not happening in the next few years.

Future
So, what’s in store for the RFID market for the next couple of years? Many industry experts are predicating that 2008 and 2009 will finally be the breakout years for RFID. I’m skeptical. Without resolution of the issues I discussed above, its going to continue to be a long, slow growth for RFID.