Why Is the Upstream Path So Ugly?

It's well known that cable's upstream signal path, from the home to the headend, is rambunctious. It's also no secret that every existing and future two-way cable service needs safe passage along that decidedly intemperate lane.

Those serve as a good pair of reasons to describe why the skinny, 5- to 40-megahertz upstream band is so disagreeable.

The interlopers that lurk in the upstream cable path aren't, for the most part, deliberate offenders. This isn't like walking down a dark alley and running into a cloaked villain with a dagger specifically intended for you.

Traversing the upstream path is like being told that the only way to get to where you're going is via Beale Street, in Memphis. At midnight. On Saturday night. Sober, presumably, since upstream signals don't have mouths.

And every path from every home to the node is a Beale Street.

With no offense to the gloriously gritty city of Memphis, the name of the game on Beale Street is the copious ingestion of potent liquid refreshments. (A friend who lives there calls it "a theme park for alcoholics.") Picture a few blocks of pedestrian-oriented street, with police and barricades on either end. A cacophonous blend of blues and jazz spills from every waypoint in between. A wristband gets you access to libations at participating pubs; every pub participates.

Thus, it's not at all uncommon to be unintentionally knocked down by an outrageously happy drunk. He doesn't mean to mow you down. He just does.

In RF terms, the drunk (who hopefully doesn't retch on your shoes) is analogous to a type of noise known as "impulse noise." There are two cardinal types of noise that can ruffle the upstream path. Impulse noise, also known as "electrical transient noise," is one. "Ingress noise" is the other. Both are sometimes referred to as "spurious noise," or undesirably intrusive noise.

Impulse noise is a form of interference caused by quick jolts of unwanted electrical energy. It briefly slams into the cable plant, crushing any information traveling along the network. Then, as quickly as it careened in, it's gone. The trampled bits can't exactly get up and dust themselves off. They're gone, too, and need to be resent by the cable modem, set-top, phone unit or other device that had been trying to transmit.

Impulse noise occurs when electric-powered devices, from furnaces to hair dryers, kick on. Dimmer switches can be a big contributor, too. As is the case when you shuffle across the carpet in socks, then touch something-Zap! Ow!-electrical arcs occur, usually when you switch on a device.

Electrical arcs can temporarily obliterate huge portions of the 5- to 40-MHz band. (At a trade show a few years back, a cable-modem vendor actually used a hair dryer in its demo, to show how rugged its upstream modulation was against impulse noise.)

Ingress noise happens when unwanted RF signals leak into a section of coaxial cable, then travel along with the intended signal as it traverses the upstream path. Where the intended signal is amplified, so is the signal that leaked in.

Engineers estimate that as much as 70 percent of the noise in the upstream leaks in from individual subscriber homes, for a variety of reasons: Improperly installed (translation: consumer-installed) F-connectors; cracked or improperly shielded (translation: cheap, consumer- installed) coaxial cable; even bad shielding around the TV's tuner can do it.

The unwanted RF signals that leak in can come from ham-radio enthusiasts, short-wave radio transmissions, and-oddly enough-the sunspot cycle. Which, by the way, is at its peak right now-which doesn't just mean it's a good time to go see the Northern Lights. Sunspots are upheavals on the surface of the sun, which produce enormous eruptions of electromagnetic radiation. That, in turn, can ionize the Earth's atmosphere, which can cause erratic and disruptive radio-frequency interference.

The sunspot-generated interference alone isn't really the culprit. What can cause problems is the fact that short-wave radio operators hitch rides on it, especially in the 20- to 30-MHz zone-the good zone, in upstream terms.

Back to Beale Street. Recall that every upstream path from every subscribing, two-way home is a rowdy Beale Street, teeming with drunks (impulse noise) and shrews (ingress). Spectrally speaking, it's quite a crazy scene.

Remember the police and barricades at the ends of the strip? In RF terms, those are analogous to the end points of the network: The headend, on the one end, and the subscribing home on the other. They do the best they can to keep the worst offenders at bay; in RF terms, they do this with modulation, which are the techniques used to imprint information onto a carrier signal.

That's why the upstream is so disagreeable. Next time, the technologies and techniques to make it better.