Still Ruled by the Tuning Fork
It's time to give up our 19th century thinking. The tuning fork was the height of 19th century signal processing!01-Sep-2004
Updated: 21-Oct-2004

This is a copy of my September/October 2004 VON Magazine Column

It's time to give up our 19th century thinking.

In the 1800s telegraph companies looked to the hypothetical “harmonic telegraph” as a way to increase the capacity of their infrastructure–pairs of tuning forks would be used to create distinct signals on a single wire. But was the mechanical tuning fork up to the task? In any case, while attempting to build such a harmonic telegraph, a frustrated Alexander Graham Bell succumbed to the distraction of developing a talking telegraph.

A telephone circuit required dedicating expensive resources for the duration of a call. Data networks didn’t have this burden–they tolerate congestion and failure. This made them relatively inexpensive and it drove a cycle of innovation and increasing capacity. VoIP “happened” when the capacity removed the need to dedicate resources. Today even the PSTN is becoming a privately branded VoIP network. Demands to regulate VoIP are just as pointless as trying to require postage on email.

But landline telephony is so 20th century. Today the cell phone is a necessity–a landline phone is redundant. Cellular phones live within a walled garden fiercely defended by companies with billions of dollars at stake and politicians acting on behalf of constituents who share the mythology that radio waves are very fragile and require special protection.

The concept of the harmonic telegraph resurfaced as spectrum allocation. The electronic version worked well enough to allow us to assign a different frequency to each transmitter. It may have been the best idea at the time but it is fundamentally flawed.

If you look at two boats near each other notice that their wakes (signals) pass right through each other. But if you are only allowed to measure the depth at single place in the water you can’t separate out the contribution from each wake because we’ve discarded the context. We call this artificial phenomenon “interference.” In the radio space, listening to a single frequency produces this effect.

Science subjects ideas to testing but some ideas are too obvious and escape scrutiny. The principle of frequency allocation was so obvious that research presumed the tuning fork model and focused on how to make it work better rather than whether it was a good idea. In reality, it was a terrible idea–a very fragile approach that requires a world-wide system to police the ownership of transmitters lest they step on each other. This artificial scarcity results in giving “important” applications priority. We have a regulatory regimen which is very judgmental–innovation requires prior approval.

This is a neat tautology. Research presumes the bad idea and then reinforces the collateral damage in order to sustain the idea. The franchise owners (the carriers) join in judging content. CellCos can charge thousands of times as much for an SMS (Short Messaging Service) byte as for a voice byte even though SMS costs them far less!

During World War II Hedy Lamarr realized that if you spread a signal across the spectrum the enemy wouldn’t be able to understand the signal and would also be unable to jam it. Military shortsightedness spared Hollywood the embarrassment of a brainy actress. In the 1950s Claude Shannon showed us that “meaning” (communications) was independent of “transport” (bits)–tele/com not telecom!

These ideas found a home in CDMA (Code Division Multiple Access) which uses digital packets and frequency hopping. But the cellular franchise is protected by the mythology of the harmonic telegraph and CDMA is confined to emulating circuits.

Microwave ovens operate at 2.4 GHz because water molecules absorb the energy of such waves and heat up. 2.4 GHz was considered useless for communications so it was set aside as an untamed frontier. Without the sheriff patrolling the town you had to watch out for yourself. The only way to survive in this extreme environment was to operate by Hedy’s rules–you had to use frequency hopping and other innovations to avoid being stomped on and to find your own signals out of the chaos. Not only did it work–cordless phones and 802.11 (Wi-Fi) attest to that–but it also brought wireless communications onto the Internet’s cost/benefit path! Even better, once we can connect to the Internet we can reach any distance with a simple short range radio.

The cellular industry is a product of the 19th century notion of frequency allocation. The packet connectivity to the Internet is simple and now obvious. Freed of wires, we shouldn’t find ourselves having to preserve a dysfunctional fiction.

Those who are mired in 19th century concepts of circuit-based telephony and the harmonic telegraph live in fear and demand protection. We can’t afford to coddle ignorance nor let fear justify oppression.