Bad things happen when ideal IT concepts bump into the realities of imperfect hardware. This time, Im talking about the slow-motion train wreck of BPL (broadband over power lines), a basically bad idea thats now the subject of a newly launched IEEE standard process.
With lots of people wanting its benefits and few people understanding its drawbacks, BPL seems likely to gain too much momentum to be killed. A win for BPL, though, could be a loss for some valuable applications of the radio spectrum—but youd never know that theres a serious risk, or even a controversy, if all you saw was the IEEEs cheerful July 20 announcement of IEEE P1675, "Standard for Broadband over Power Line Hardware."
The IEEE announcement calls the BPL proposition "relatively straightforward," saying "A computer-router combination and a coupler take the signal from an optical-fiber cable as it enters a substation and imposes it on the electric current. The signal travels over the medium-voltage lines, with repeaters placed every 0.5 to 1 mile to keep the signal viable. A repeater/router near a residence or business extracts the signal off the medium voltage just before the transformer and injects it onto the low-voltage wiring on the other side of the transformer. The signal is now on all of the low-voltage wiring within the structure and can be accessed at any outlet by plugging in a modem."
The elephant in the living room prompts my follow-up question: "Where else is that signal accessible—whether its wanted or not?" Incredibly, the words "radio" and "interference" are not even mentioned in the IEEE announcement, even though the risk of radio interference from BPL is obvious and grave.
BPL proposals place data signals on carrier channels that span a broad swath of frequencies. Those carrier frequencies overlap those used by everything from international shortwave broadcasts to standard time signals to CB radios (remember those?) to baby monitors to the low end of the range of TV channels. Although not intentionally radiated, those BPL signals will be traveling on wires that cant help but behave to some degree as antennas.
We keep signals confined, on a small scale, by precisely tailoring signal paths in chips and on circuit boards. We control them on a larger scale by using shielded conductors, or twisted-pair lines that cancel stray radiation by combining equal and opposite components. And when we need to convey a complex signal at high power levels—for example, when feeding a moon-bounce radio antenna array—we dont use just an ordinary wire. We use a transmission line, a carefully tailored component that matches voltage and current ratios between the source and destination and that minimizes stray radiation of signals.
Electrical power lines are designed to carry power and are optimized for efficiency and safety—not for minimum radiation of high-bandwidth energy.
The IEEE P1675 announcement speaks about traditional power-system priorities and quotes Terrence Burns, chair of the IEEE BPL Standards Working Group, as saying, "Power companies face a number of issues ... for example, how to assess the performance and safety of repeaters/ routers, medium- and low-voltage coupling hardware and other equipment before buying. Other issues include how best to put this equipment in place and to keep the overall system operating well and prevent it from interfering with power delivery. The new standard will help them deal with these concerns."
Is radio interference someone elses problem?
To be fair, some BPL proposals do include active measures for detecting and avoiding communication interference. Im sorry to rain on proposers various parades, but shortwave communications are what we turn to when other things arent working, and I dont like that failures in a new and complex system could put an essential backup system at risk.
"Nearly all electrical utilities are exploring BPL because the potential benefits are so substantial," said Burns. Yes, and there would also be "potential benefits" in breaking the second law of thermodynamics, but no one expects to be taken seriously if he or she proposes to try. Radios realities deserve equal respect from the proponents of BPL.
Technology Editor Peter Coffee can be reached at firstname.lastname@example.org.
Check out eWEEK.coms Infrastructure Center at http://infrastructure.eweek.com for the latest news, views and analysis on servers, switches and networking protocols for the enterprise and small businesses.