The Future of Wireless Networking

• By David Chisnall
• May 15, 2009
• Article is provided courtesy of Prentice Hall

Like this article? We recommend
Introduction to Wireless Systems

January 1, 2000—a historic day. Telecom companies around the world would simultaneously abolish long-distance calling charges, making every call a local call. At least, according to Arthur C. Clarke, writing around a decade earlier.

In the real world, this didn't happen. Telephone companies still charge different amounts based on your distance from the other endpoint of a call. It wasn't exactly unlike Sir Arthur's prediction, however. International calls I've made in the past few years have cost me absolutely nothing beyond the fixed network connection rate I pay for a broadband connection. But their endpoints have been other devices connected to the Internet, rather than the old telephone network. In practice, most of the cable between the two endpoints will be the same, irrespective of which logical network places the call.

The All-IP Network

The wired infrastructure in the UK (operated by BT) and the next-generation wireless telephone networks all have one thing in common: They use IP internally and run telephone calls as Voice over IP (VoIP). My current phone supports UMTS, and is assigned a 10/8 IP address whenever I use it. This means that it's behind a Network Address Translation (NAT) and therefore cannot receive inbound connections.

The 10/8 subnet is the largest private allocation. As a /8, it has 2^24—a little under 17 million—unique addresses. This isn't a very large number in comparison to the number of devices connected to the mobile phone network, which is why mobile operators are likely to be among the first major deployments of IPv6. With v6, a company (or even an individual) can easily get a /64, meaning that the first 64 bits identify the network and the last 64 bits identify the device. To put this into perspective, the company has enough address space for every valid IPv4 address to be a network the size of the Internet, or for every person to have three billion devices on his or her network. More to the point, it allows every device to have its own IP address, and the routing tables to be sufficiently sparse that routing is cheap.

One other advantage that IPv6 adds, of particular relevance in this arena, is Mobile IPv6, in which a device can change its location in the network and still be routable, without dropping existing connections. Moving between towers is generally handled lower down the protocol stack, but this new arrangement allows a phone to move between currently-separate networks while still maintaining connections, as long as both endpoints have IPv6 connectivity.

An all-IP network emphasizes the distinction between selling access and selling services—a distinction that mobile phone companies currently try to blur. When you make a telephone call, you're using their network, and you're also making use of the peer agreements they have with other phone networks, while enjoying the use of their routing system.

Mapping from telephone numbers is much less important than it used to be. I used to make telephone calls by a complicated process. First, I would look up a number in a paper directory or a local cache—either a personal address book or my brain—and then enter this number on a dial (later, on a keypad). By contrast, now I just select the person's name from my address book and press "call." A recent study claimed that peoples' memories are getting worse because they can't remember their friends' phone numbers. For me, it's not just that I don't remember my friends' numbers anymore—I never knew them. In many cases, a friend sent me a vCard via email or Bluetooth, and I never looked at the number field in the first place. Already, the idea of calling a phone (rather than a person) seems faintly archaic.