Orthogonal Frequency Division Multiplexing

It seems like I run into this term with just about every new digital transmission technology. What the heck does “orthogonal” mean when applied to frequencies? What are the advantages of this modulation scheme?

It was this paper written back in 1994, by Phil Karn, an amateur radio operator, that made it make sense for me. He explains the concept in great detail along with the math behind it and also proposes one particular coding scheme (but there are many possible).

After gaining an understanding, I am curious if this isn’t what the proprietary Telebit protocol PEP is. They didn’t have the high end digital signal processors to work with but they had the Motorola 68000 CPU.

Now that I have a reasonable understanding of the protocol and the math behind it, I understand the reason it is being so widely implemented and why folks are pursing ultra-wideband transmission schemes. The potential benefits of this protocol are so great that I wonder if any other modulation schemes will still be in use in a couple of decades.

My question with respect to how “Orthogonal”, a geometrical term, applies to “frequency”. Imagine an old linear radio dial, the type with the numbers from one end to the other with a little pointer that moves when you twist a knob to select a station. Now imagine that dial is linear. Now on this linear dial you have lines marking off frequency divisions equally spaced apart. You’ve got a bunch of parallel equally spaced lines on the dial, they are orthogonal. Now that you can visualize that you can see how orthogonal can apply to frequencies.

Orthogonal frequency division multiplexing divides up bandwidth into a bunch of narrow channels using equally spaced carriers each modulated at a slow symbol rate. This is realized by using fast Fourier transforms rather than attempting to create hardware to modulate and demodulate each carrier. It is the advent of fast digital signal processors that has made it practical to encode high bit rate streams suitable for digital video or local area network applications.

There are a number of very interesting ramifications. If you use a large number of channels, this modulation technique can approach the Shannon limit which defines the theoretical maximum data transmission rate possible with a given bandwidth and signal to noise ratio. Another way of putting it is that OFDM utilizes the available spectrum with near perfect efficiency.

OFDM can transmit data over a channel in which the noise is much higher than the signal. That makes it possible for multiple devices to share the same spectrum without interfering with each other.

Each channel operates at a very low rate relative to the total speed. Each symbol transmitted is smeared out over a long time frame and in the receiver integrated over a long time frame. Ordinary modulation would suffer severe interference with even brief interruptions, but because of the way OFDM smears the transmitted symbols out over time, a brief interruption in the received signal will not cause a loss of data. Those of you who’ve been annoying by the swish-swish on FM while driving down the road can appreciate this.

By itself this scheme is still sensitive to selective channel interference. Additional coding can be done to reduce this. When this is done, OFDM becomes CODFM. Again the article references above gives one example.

In addition to making efficient use of bandwidth OFDM also makes efficient use of transmitted power since it can allow the maximum transmission rate for a given signal to noise ratio.

As the cost of digital signal processors continue to drop and the power of digital signal processors continues to increase, I expect that it’s a matter of time before CODFM pretty much displaces conventional modulation schemes altogether.

Technical Trends in Broadcasting

Digital is the world today, analog is on it’s way out. NTSC (Not The Same Color) television is being replaced by HDTV, and now AM and FM radio stations are being digitized. There are big advantages. Digital transmission allows for error correction and redundancy eliminating many of the annoying reception problems of analog service.

We’ve all experienced the swish-swish-swish of a marginal FM station signal or the buzz on an AM station as we drive under power lines. Digital promises to eliminate these things.

Digital also promises to improve sound quality giving us digital AM signals that approximate the quality of good FM reception, and FM signals that approximate CD quality.

On FM, digital will make additional channels available allowing the targeting of niche broadcasting markets.

There is a dark side. In the United States, AM channels are spaced 10 KHz apart allowing for 5 Khz upper and lower side bands without splattering into adjacent channels. However, current FCC regulations actually allow modulation up to 10 KHz, but sidebands past 10.2 KHz must be attenuated at least 25 DB from the main carrier. The reality is that there isn’t much energy content above that in most program material anyway.

Digital transmission is allowed to go out to 15 KHz. The analog audio occupies 0-5 KHz, and the digital portion of the signal occupies 5 KHz – 15 KHz. The ramification of this is that adjacent channels are going to be totally clobbered. Under the old rules, the sideband distant from the adjacent channel was reasonably protected, so with a narrow receiver you could still receive a station next to a stronger adjacent station. With digital, both sidebands will be clobbered and the adjacent channel will no longer be receivable.

Analog signals decode gracefully, as the signal gets weaker, the signal to noise ratio gets worse, but the signal may be receivable over thousands of miles at night. With digital either there is enough signal to decode, or there isn’t. This means that signals will either be receivable cleanly or not at all. The effect of night time fading may be to have signals come and go entirely.

Ah, but you still have the analog signal! Yep, except it will probably be buried by the digital transmission of a nearby digital station on an adjacent channel. I’m afraid they’re trying to kill one of my hobbies (DXing). It’s hard to say how it will play out. Surprisingly, TV DXers have had a fair degree of success with digital television.

To prevent adjacent channel interference, AM stations are presently only allowed to transmit digital during the day. I’m sure that this restriction will be lifted if digital AM proves to be popular.

Digital FM is less problematic because FM sidebands are wide enough to allow the digital signal to fit within the existing spectral footprint. With frequency modulation, sidebands are created not only at the sum and difference of the modulating frequency but also at harmonics of the modulating frequency. However, it is possible to choose a modulation index that limits significant sideband energy content to a desired bandwidth. Digital transmission may require discontinuing the use of certain subcarrier frequencies that would overlap the digital signal but the digital signal provides additional channels that can replace the functionality of the subcarriers.

The amount energy that goes into the digital subcarriers is very small so digital FM should not substantially detract from the analog signal and therefore should not negatively impact long distance reception.

A trend that seems to be accompanying digital conversion is power increases. I feel the FCC’s rules regarding FM broadcast really need to be revised to accommodate the realities of modern receivers. The FCC generally considers the 1mv contour an FM stations service area and does not protect beyond that contour. But many modern receivers have sensitivities below 5uV (mv = millivolt 1/1000th of a volt, uV = microvolt, 1/1,000,000 of a volt). I found one receiver with an advertised sensitivity of .25 uV. Even the cheapies tend to be better than 50uv these days.

In this region, what tends to be the reception limiting factor is not signal strength but multi path interference. The frequency of an FM station is being continuously modulated by the program content. This means that a 100.1 Mhz FM station might be 100.099 Mhz one moment and then a millisecond later 100.101 Mhz. The first signal arrives directly at the antenna, and then a signal reflected off a conductive object arrives at the antenna a millisecond later, and the two signals create a “beat” signal of 2 Khz. The end result of this is that you here this static distortion of the original signal that renders it very unpleasant to listen to.

More power doesn’t help with this problem, if often seems to make it worse. Modern FM signals which have a stereo pilot, stereo sidebands, and a variety of SCA subcarriers make the problem worse because all of those signals are modulating the frequency as well. Digital should improve this situation.

It is my belief that current FM stations are in large part overpowered. If you aren’t able to receive a distant station it is only rarely because that station is underpowered. More often it’s because a closer adjacent station is overpowered and overmodulated.

Given the reality of energy supplies being tight, this would seem to be an area where a lot of energy could be saved. I believe the FCC should update the signal level requirement to reflect modern receiver sensitivities, replace 1mv with 50uv or so, reducing transmitter power accordingly. There are 9000 FM stations in the United States, their total power output approximates that of a medium sized nuclear reactor. If contours were reduced from 1mv to 50uv, that would reduce

While increasing power generally does not help with multi-path interference, but does create more interference problems for people listening to other services, I believe the rules regarding antenna height above terrain verses power should be re-worked to favor higher towers over higher power. If the rules were re-worked to give higher towers an advantage over higher power then more broadcasters would opt for big towers and lower power.

This would be good for everyone concerned. It would be good for the broadcasters because they would get increased functional coverage area for only a capital expense while their operational expenses would be reduced. It would be good for the consumer because they would get a cleaner more multipath free signal. It would be good for the planet because less energy would be wasted.

In areas where the land is flat, not mountainous and hilly, and the antenna is not located at an extreme height relative to the terrain, a combination of more bays providing a higher gain and vertical directivity, that is to say power is concentrated in the horizontal plane, and lower transmitter power (to compensate for the higher antenna gain) can result in equivalent signal levels at a distance but lower signal levels near the transmitter resulting in less interference and more economical operation.

On the AM bands I am also seeing power increases left and right, KJR from 5 Kw to 50 Kw, same for KOL, KRKO also is getting a power increase later this year. With AM, if we still had clear channels, these power levels would be useful. Under present circumstances I’m not sure it really accomplishes anything except to better overload receivers in the local vicinity and cause more co-channel interference.

Classic Radio

I miss radio that involves live human beings and the audience. I liked the excitement that many of the older stations exuded. Radio now is so incredibly terribly bland. I don’t understand how it pays for the electricity to run the transmitters.

When KJR, KING, and KOL were all “top-40” in Seattle, I could also hear CKLG and CFUN in Vancouver BC, which were also top-40 as well as KTAC in Tacoma.

I liked the people on KJR in the 1960’s and early 1970’s. I enjoyed all the crazy stuff they came up with.

I remember some crazy ass thing Lan Roberts did where he was going to try to talk to space aliens or some such with the stations transmitter. Something about getting special permission from the F.C.C. to turn it off momentarily to listen for a reply. (never mind the fact that AM signals are absorbed by the ionosphere in the day and refracted back to the surface by the ionosphere at night thus never making it out into space).

And.. Mr. Science Mr. Science It’s Me! Jimmieeeeeee!

I was unaware until now that Lan Roberts had passed away from cancer in 2005. I knew he had been diagnosed with cancer way back when he was in Texas and I remember him saying in a post way back that he was thankful because he’d beat it, it had been six years or something like that and there were no signs.

They used to have that woman in the morning that screamed “Wake Up!” at the top of her lungs.

CFUN in Vancouver BC Canada, I took a liking to when KJR started to die. CFUN was high energy. Dead air didn’t happen, not even for a millisecond, music never stopped. If music wasn’t playing a musical station jingle was, commercials were done to music, cues were perfect, air personalities were excellent at talk overs right up to but never stomping on the lyrics.

I also liked CFUN’s audio processing, it was much like XERB and XEPRC when Wolfman Jack was on, they used heavy fast compression combined with reverb so the audio density was high, but it wasn’t flat bland, it was high energy exciting. It also wasn’t grossly clipped or distorted which was a problem with AM stations here in the Seattle area.

An engineering firm here in Seattle, allowed me to accompany them once when they did some maintenance work at KISW several decades ago. The transmitter site for KISW used to be on 92nd and Roosevelt, a couple of blocks from where I lived at the time. I’m not going to name the firm because of what I am about to describe.

I cringed when I saw how they adjusted the limiters / clippers at KISW. They had regular programming feeding, turned the clipper off and adjusted the limiter so limiting heavily, maybe 20db gain reduction, the output would drive the transmitter to about 120% modulation, then they turned on the clippers and clamped it down to 100% with the clippers. Thus the audio was essentially continuously clipped.

Now in fairness to them, many stations wanted to sound “loud” at the time and that was a way to achieve that (though I think there were much less audibly destructive methods of achieving loud high density sound). This particular firm was top-notch otherwise doing many complex tasks that others would not or could not take on.

That method of setting up audio processing, all I can say is Ick! Other stations took a different approach, using multi-band compressor / limiters like the Orban Optimod units which could achieve a quite loud and high audio density, especially if you added a small amount of reverb up front.

Arguably any of these methods would make an audiophile gag, but I particularly disliked the heavy use of clipping because of the intermodulation distortion and high order harmonics that it would introduce. In my view given AM radios limited S/N ratio, heavy processing was justified, but not so FM.

KING used to do a lot of promotions, a lot of music give-a-way’s. I got real good at manual speed dialing (this was back when you could only have AT&T Bell telephones) and they had no rule regarding the same person winning more than once, so they contributed to my album collection.

I met Bill Wolfenbarger, the chief engineer at KOL AM 1300, after contacting many radio stations in the area looking for used broadcast equipment we could use in our pirate station. Bill used to allow us to use a production studio to record things. I ruined that by exploding a tape reel in the studio.

They used to have these great big reel-to-reel floor standing tape recorders. I had recorded a reel of tape and was going to rewind. Now these machines had a fast-forward and fast-fast-forward, they also had a rewind and fast-rewind. The “fast” versions had red lettering that said, “Do not use with plastic reels”. Being the stupid teenager I was then, (as opposed to the stupid adult I am now), I had to know why so I hit the fast-rewind. The machine spun up and then Boosh! There were pieces of tape reel embedded in the wall, my shirt and T-shirt were sliced, but somehow I and others with me all escaped injury. However, that did, understandably, ruin my welcome in the production studio.

I was working on getting into radio legitimately and working towards my 1st class radio telephone operators license (which I did get in my junior year of high school) and Bill was one of the engineers willing to let me pick his brain and see a real station.

I saw quite a few actually, but KOL was uniquely impressive and Bill was uniquely willing to talk about radio and share his knowledge. It wasn’t new and sterile, but it was a class act. It was real organic radio with real human beings at the helm. The station was saturated with life energy. I was so sad to see it go when it was bought out and became KMPS country.

If you have any old photos of KOL, especially with the lighted call letters, and would be willing to allow me to share them here on the blog I’d really appreciate it.

Adrianne Curry Show

I would describe the Adrianne Curry Show as like a female version of the Howard Stern show on the Internet. It’s definitely not for children or conservative adults. It’s more like a wild on-air party than a traditional radio show.

She prefers to describe her show as like MySpace on crack. The show is connected to a web-based chat so users can drop in and participate that way.

If you like Howard Stern you might like this show. If you dislike Howard Stern because of the sexual content, then you probably won’t like this show.

Sirius XM Merger

Sirius and XM to merge according to an article in CNN MoneyWatch. Both companies are disappointed in their stock value and believe there will be value in a merger.

Here is my prediction for the future of satellite radio. Satellite radio has a limited future. I believe there is money to be made in it during the next five to ten years. I believe it will grow for the next five to ten years and then begin a decline.

What I believe will be it’s replacement is a new form of terrestrial radio which will be the result of IP radio combined with WiMax wireless broadband. The reason for this is one of simple economics. Broadcasting via satellite is tremendously expensive, terrestrial radio is dirt cheap. Conventional terrestrial radio lacks variety, particularly recently with a handful of giant corporations owing the majority of broadcast stations. Conventional terrestrial radio also lacks broad coverage and high quality audio for the most part.

The marriage of WiMAX and IP radio will address both these issues. It will become possible to listen to any IP radio station anywhere there is WiMAX coverage which I predict will eventually be essentially everywhere.

Presently there is a technical issue preventing this from being a reality and that is that presently IP radio is sent using UDP which has no delivery guarantee mechanism and the protocols used to send it themselves lack delivery guarantee and retransmission capability.

WiMax, presently is much like WiFi with longer range, will eventually be adapted to something resembling IP cellular service. This may be a completely de-facto adaptation with receivers adapting a strategy of seeking multiple connections and using the best at any given location.

The switch-over though will always involve interruption however momentary with packet loss, and a resulting interruption in audio. But as more people start using it that way, someone will adapt or create a protocol that includes error detection and retransmission for IP radio, and as soon as that happens commercial receivers incorporating WiMAX and IP radio in a user friendly box will emerge.

As soon as that happens, any incentive to switch to satellite radio will be largely lost owing to the much larger variety of programming material an lack of fees that WiMAX / IP radio will provide.

I expect this will occur within five years, but because people with investments in existing hardware or who enjoy programming on satellite radio that they’ve become accustomed to will continue listening, I don’t expect the satellite market to die for at least another five years after that but I do expect it will be on the decline.

PropNet – Realtime Mapping of Propagation

PropNet asks,

“If the band is open and nobody is transmitting, can anybody here it?”

Amateur radio operators have been exploring radio propagation for decades. They have come to utilize reflection from the ionosphere, tropospheric ducting, tropospheric scatter, reflection off the moon, and various other modes in which radio waves are transmitted beyond line of site to achieve communications.

Some DX hobbyists like myself enjoy receiving distant stations beyond the distances normal conditions and methods allow. This can involve optimizing equipment, developing techniques for detecting weak signals buried in noise, knowledge of various propagation modes and when they are likely to occur, and just plain patience and luck.

Amateur radio operators have come up with a new way to exploit digital technology and the Internet to allow mapping of radio propagation at various wavelengths in real time. The technology consists simply of beacons that broadcast digitally encoded messages on fixed frequencies at regular frequent intervals and receiver stations that listen for these beacons and when they receive one report back to a central internet server that then records that reception and maps that propagation path in real time making it possible to develop maps, updated in real time, of radio propagation at various frequencies.

Propnet.org is a the result. Presently the mapping is pretty sparse, however, as more people become aware of this project and join it will become more complete. You need an amateur radio license to participate as a transmitting beacon but anyone with the proper receiver equipment can participate as a receiving station.

In addition to this mapping project there are also links to five day tropospheric ducting forecasts and radio and television DX’ing.

Robert Van Dyke Vintage Audio

The majority of the equipment we used in our pirate radio station was tube equipment. I think about the only thing solid state was a couple of Radio Shack phono pre-amplifiers that amplified the output of the magnetic cartridges on the turn tables enough to drive the input of the all-tube Gates Yard mixing board that we used.

One of the tube pieces of equipment we used was a Langevin compressor / limiter which did a fairly decent job of crunching the audio down, at least as far as a single band compressor limiter can. We modified it somewhat so that the time constants were as short as possible without severe intermodulation by bass, and we modified the limiting stage so that it only limited on negative peaks and allowed the positive peaks to go as high as they will.

I did a Google search curious to see if any of these old beasts were still around and I ran across, Robert Van Dyke Vintage Audio. He sells all sorts of old tube equipment.

The Langevin compressor / limiter wasn’t among the stuff he currently has in stock but there is still a lot of cool stuff on his list. The equipment he offers for sale is serviceable and might just have a place in your radio station so I’ve added a link to the sidebar.

For those not familiar with tube equipment, there are desirable and undesirable aspects to tube verses solid state.

The desirable aspects is that tubes are very linear compared to their solid state equivalents, and so much less negative feedback is required, and as a result the phenomena common to solid state equipment known as transient intermodulation distortion is almost non-existent in tube equipment. Tubes clip “soft” instead of saturating hard like transistors. As a consequence distortion produced by clipping tends to be lower order harmonics which are audibly much less objectionable than higher order harmonics. In power amplifier applications, tubes amplifiers tend to have lower damping factors and some people (myself included) find this desirable, others who are more concerned with technical perfection prefer an amplifier with a high damping factor to make speakers behave. Tube amplifiers and other devices are less susceptible to radio frequency interference than solid state equipment and they handle overloads better.

The disadvantages to tubes are hiss, hum, poorer frequency response, higher harmonic distortion figures, and they are high maintenance requiring tube replacement, hum balance adjustment, and bias adjustments, among other things. Tubes are less efficient and generate more heat as well as producing less output power for the amount of electricity they consume. Tube equipment tends to be heavy and bulky.

If you’re like me and something that sounds aesthetically pleasing is more important than something with good technical specifications, then you might want to consider vintage tube equipment. I’ve given it up simply because I can’t afford it.

Rec Networks

REC Networks started as a record-a-call telephone entertainment company. They have since become involved in low power broadcasting and offer free services helpful to the unlicensed low power broadcaster such as free engineering information and channel searches.

I find their history interesting as exploring the telephone network was also something I found to be of interest as a teenager when operating our bootleg radio station.

We discovered something called “loops” back then. These were test numbers used by telephone technicians for trunk (trunks are the circuits that connect telephone central offices to each other) testing but we found them useful as way to allow listeners of our pirate radio station to call us without the number given being traceable to us.

Loops have two telephone numbers associated with them, back in the days they almost always used -0018 and -0019 for the loop in a given central office. If you had a trunks in a trunk group you needed to test, first you would call the -0018 side of the loop from a test panel on one of the trunks to be tested. It would send a 1 Khz tone from the far in at 0 dbm (dbm means decibels referenced to a 1 milliwatt signal level). You would receive the tone on your test panel at say -3dbm, meaning their was 3db loss in the trunk in the receive direction (from the distant central office to you). Then on another trunk circuit you would call the 0019 side of the loop. The loop would connect both trunk circuits together so now if you sent a tone at 0db on the second trunk and you received it back at -6db on the first you would know that the transmit loss on the second trunk was 3db (6db total loss -3db known receive loss of the first circuit). Then you would repeat the procedure for both circuits in the reverse order to determine the transmit loss for the first circuit and the receive loss for the second.

The important thing to know here is that if you called say 555-0018 you would hear a 1 Khz tone, and then if someone else called 555-0019 you would be connected together and could talk.

So there we were operating a pirate radio station and we wanted to take callers online or take requests but we didn’t want to give out our telephone number thus identifying ourselves to the FCC. So instead we’d call 555-0018, and give out 555-0019 over the air and wait for the tone to stop.

If you are thinking ah it was tough back then, today we’d just use a cell phone! Be aware that cell phones have GPS receivers built into them now and will give law enforcement your location within about six feet. Call forwarding is also trivial to trace.

That’s ok though I think the pirate stations that are operating in defiance of FCC orders to shut down on the grounds that it is their first amendment rights are doing the right thing. Surely if the framers of the constitution had anticipated broadcast, it would have been included, and logically should be included as part of “the press”.