Is it possible to determine whether a phone call is local or long distance by analyzing the audio?

Nice. The only way I could detect a working echo canceller was to inject a 2100 Hz tone (or phase reversal) to disable the echo canceller. Then, it was obvious. ITU-T G.165 and G.164 have the specs for tone disable somewhere. I think 2225 Hz is used in some countries. I got lazy and just used a fax machine to initiate an echo canceller disable tone.

I'm not so sure. Even local calls had some echo from acoustic coupling in the handset. Central office delays were all over the map in the 1960's (I don't know about the 1970's) as Ma Bell converted from analog to digital muxes. Local calls were also not always contained within a single CO. Local area code include multiple CO's any of which could introduce echos and delays. I suspect he really wants to know if the call was terminated in the same CO as it originated.

About all I can be certain of is that if the voice traffic switched to half duplex (talk one direction at a time), it was long distance between two different overloaded carriers. No echo analysis required for that. Half duplex was common in the 1950's but less so by the

1970's. Also, I might be able to look at the inter-syllable background noise, and try to determine if an AGC (automatic gain control) or MUX was involved. From that, I might be able to guess the degree of processing, which might point to the equipment type at the CO. More processing might mean it goes through more analog electronics, which means a long distance call. It might be a long distance call, but might also be a call between local CO's inside an area code, which is deemed "local".

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

The flaw in the *practice* of most phreaking was: who are you going to *call*? A stranger? Call a relative and you risk TPC later contacting them and inquiring as to who they *received* the call from. What *stranger* are you likely to call and chat for a significant amount of time? ($$)

In certain exchanges, TPC would install traps on all the lines just for that reason.

And, getting caught with a "box" was sort of hard to explain ("Oh, it's a miniature *organ*! I use it to play music while riding the bus... Yeah, I know it sounds REALLY CRAPPY...")

I'm not sure it was ever considered /prima facie/ evidence -- but anyone aware of that sort of crime (campus!) knew what it meant! E.g., in many jurisdictions, possessions of lock picks (without a license) is evidence of a crime.

Another prevalent hack was the annual publication (in _TAP_) of the "eagerly awaited" credit card code. This has the advantage that you could keep it in your head and achieve the same results as carrying a box on your person (effectively, a "red box in your head").

If anything is proof of the futility of "security by obscurity"...!

This is some work I did dealing with USR modems and the local POTS lines in 1999: With a modem, I would guess(tm) that the upper end is at 3KHz on the worst lines, and 3.5KHz when fairly close to the CO. With a proper line tester, maybe a little higher in frequency.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

On a sunny day (Thu, 12 Mar 2015 16:06:15 GMT) it happened Jan Panteltje wrote in :

But but but.... panteltje12: ~ # telnet news.mit.edu 119 Trying 18.181.0.25... Connected to news.mit.edu. Escape character is '^]'.

502 senator-bedfellow.mit.edu: Access denied to your node - snipped-for-privacy@mit.edu Connection closed by foreign host.

panteltje12: ~ # telnet news.aioe.org 119 Trying 94.75.214.39... Connected to news.aioe.org. Escape character is '^]'.

200 news.aioe.org InterNetNews NNRP server INN 2.5.2 ready (posting ok)

So what good is it? ;-)

It was never open to the general public. They are understandably very protective of their bandwidth, which at some point was unbelievable.

There used to be mailing lists called LISTSERV based on IBM mainframes, ours at Yale. At the time a student friend of mine got his PhD in Computer Science, had to return to Venezuela and bequeathed me the administration of the list, which grew considerably. Recall that in the early 90s there were coups and tons of rumors down there. Friendly employees of the two main Venezuelan newspapers sent me the news nightly, plus a cousin who worked for IBM in Caracas (in theory it was only for IBM customers). The 3 were violating rules. We had a mailing list in which a magazine was distributed in *Postscript* with pictures, fonts, etc.

Some presidential candidate (later, president) came to Harvard and his staff (later, cabinet members) were patronizing us:

"Oh, I brought a copy of 'El Nacional' and 'El Universal' of today. I am sure you youngsters would like to read it!"

I told him:

"Abdon: We receive both newspapers, and distribute them all over the world, the *previous* night, before it gets printed."

and showed him the Postscript magazine. His jaw dropped.

We were using about 10% of MITnet traffic. One day Jeff Schiller (MIT main network guy) told me and my buddy Hector:

"Guys, why don't you make your lists distributed? For instance, the Brazilians have a server for the West of the US and another, here at MIT, for the East of the US".

We ignored the kind advice, and in fact I decided to do the opposite: The subscribers in Europe, Russia, etc. were attended by a sub-list in France. I told the girl in charge "Hand over the list, the whole thing is going to be based at MIT".

Fun times...

Oh, speaking of unbelievable bandwidth. One day my boss was notified by his (Cecilia D'Oliveira):

"Jeff, pack your bags, you are going to DC, to declare before Congress".

It turns out that our benefactor and hero, Senator Al Gore, had gotten us the best possible Christmas gift: a national network of T1s. In the airplane, Jeff, had no idea what to say, how to justify such expense to laymen, ignoramuses congresspeople. When asked, he simply replied:

"Well, we are go "Makes sense. Approved".

Good times...

-Ramon

I forgot the must fun part:

"Abdon: This is the magazine that you will read *next* month in Caracas".

HP had recently moved from 300 dpi to 600 dpi and MIT has the laser printers before they were available to the general customers. It took us a while to get the double side working properly.

Then Digital Equipment (founded at MIT, circa 1957) gave us advance hardware (RISC DECstations). Until then, silly me, I had no idea that hardware also has versions (1.0, 1.5, etc).

-RFH

Practically nothing going past the next county (if that) was done with copper wire by 1970. AT&T Long Lines had moved pretty much the whole country to microwave carrier by the late 1950s. Blocks of carrier channels were converted down from microwave FM and then the whole carrier block was modulated back onto another microwave link for the next ~30 mile hop. So, the individual FM subchannels were not demodulated until the end of some large chunk of distance at a switching station.

The links were truly speed of light, and I don't know how much delay was added at each repeater station. (Probably not much at all.) All the filters when the subchannels were demodulated probably did add a bit of delay.

Jon

I posted this question in the Stack Exchange Forum:

and was asked for an "Audio Corpus".

It is here:

In one of the audio clips, I removed some of the background noise, most likely caused by Fonzi's tape recorder.

TIA (again),

-Ramon

On a sunny day (Thu, 12 Mar 2015 12:37:49 -0500) it happened Ramon F Herrera wrote in :

Nice stories, yes Usenet was, and still is, a powerful tool. I guess I could set up a node, and get / send news via mit.edu, or some other big one, that was the common procedure. But I no longer (at least today mmm) run servers here (Netherlands), even my website is hosted in the US with Wild West Domains. And my ISP would complain about the bandwidth use (cable). For news I have satellite and see can Europe, China, Russia, Middle East, and we actually could get some South America relayed via Spain but have not tested that in years. Only the US NASA TV is below the horizon, but I can see that on internet,

One thing, these days with glass fiber being installed, the 'huge bandwidth' from the past would be tiny these days, and Usenet usage is dropping, many servers do not support binaries anymore, so maybe that bandwidth is now in the shadows. It is all youtube, you'd be surprised what you find there! But I hope Usenet stays, keep up the good work.

We wouldn't dare insulting you or professional dedication.

:-)

-Ramon

Bad idea. Every time you process and re-record that audio, you both introduce and lose important artifacts. For example, by chopping off the high end noise, you lose the tape recorders 100 or 150 KHz bias signal and its low level mixing products from the recording. The bias leakage is one of the easy ways I can tell if the recording was really made on a tape recorder, or if it was digitally faked on a PC. Also, if I see more than one tape bias carrier, I know that it was re-recorded on two different machines, which implies that it has been edited or tweaked.

For the S.E.D. readers, the 100 KHz bias leak usually shows up at around -60dB below maximum audio level, which is difficult to see under the tape hiss. The better tape machines use 150 KHz, which is almost impossible to see. Both can be found with filters if the exact frequency is known. Fortunately, junk tape recorders leak much worse, and are easier to see.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

Thanks again, Jeff.

Check out Bob Primeau [kinda] colleague of yours:

He is getting lots of new clients, nobody has dropped a horse's head in his bed, etc.

Incidentally, I was privileged to work with *the* company that created the Internet, Bolt, Beranek & Newman.

They happen to be the ones who analyzed the "Dictabelt Recording": again, they gained a lot of professional respect and are in excellent health.

Regards,

-Ramon

Jeff, please bear with me. I am honestly trying to see your point of view.

Suppose there is a lawsuit and a court calls you as an *expert witness*.

We experts (you in your field, I in mine) are supposed to have high ethical standards, we are supposed to be above the yelling, accusations, etc.

How about jury duty? That is your citizen's obligation. That is the price we pay for our system, the greatest in the world.

Well, an expert is more important for Democracy than some uneducated member of the jury.

-Ramon

I have no point of view.

Expert witnesses are rarely subpoenaed. They are hired by the attorneys or by the court. "...courts hold to the traditional rule that an expert cannot be subpoenaed to serve as a witness against his or her will." The few times I've played expert witness, I testified by deposition and never appeared in court. The one time I was scheduled to testify, I sat waiting outside for days, and was never called (or paid).

I presume you haven't read much of Usenet, where yelling, accusations, character assassinations, bluster, conjecture, lying, and juvenile pissing matches are the norm. Oh yeah, welcome to Usenet.

I've been on 3 juries. Two were a waste of everyone's time. I wanted to be on a real jury so that I could experience the American legal system. What I got was a voluntary manslaughter trial, where literally everyone including the police were lying and seriously incompetent. I was the foreman. It was an interesting experience in the reality of trial by a collection of people not smart enough to get out of jury duty. If that was the greatest in the world, I would hate to see the lesser legal systems.

Cool. I've always wanted to setup a technocracy, where the government is run by the experts. However, I suspect that it's easier to obtain a decision from a jury selected from the GUM (great unwashed masses), than from a panel of university scholars, who derive so much enjoyment from endless debate and hair splitting.

In 1963, I believe that the Dallas PD was using Motorola Dispatcher radios, probably D33AAT: These were part germanium and part tube radios. Over the air, they had some interesting and obvious characteristics. The first second or more (depending on battery voltage) of each transmission was cut off as the quick heat tube filaments (1AD4 6397 2E24) warmed up. Officers using these radios had to push HARD on the PTT button, wait about 1 second, and then talk. There was no way to accidentally jam one of these radios into transmit using the PTT button. Most of the radio transmissions I heard on various recordings found on YouTube appeared to involve motorcycle police.

The D33AAT produced about 10 watts of RF at VHF at about 10% efficiency from a vibrator power supply. One could usually hear the vibrator hash on the transmissions. Despite that claim by someone on the recording that it was a motorcycle with a stuck transmitter, I doubt it because there was no vibrator hash, and the PTT on the microphone intentionally required a death grip to depress. It's been

50 years, but I'm fairly familiar with what those old radios sounded like on the air, having used them in commercial and ham service in the late 1960's. The motorcycle radio mics of the day were not noise canceling mics. Instead, to minimize wind and road noise, they simply reduced the mic gain requiring the office to yell into the microphone to be heard. An open mic wouldn't hear much beyond mechanically coupled noise from whatever it's sitting upon (i.e. engine noise).

All the recordings I heard sound like noise and distortion had been added to what would normally be a very clear wide-band (+/-30Khz deviation) audio system. I've heard some crappy radios in the 1960's but nothing that awful sounding. Yet, nobody asked for a repeat transmission, which suggests that it was perfectly clear to the officers involved. Hmmm...

Anyway, have fun with your conspiracy. I'll pass.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

On a sunny day (Fri, 13 Mar 2015 14:52:23 -0700) it happened Jeff Liebermann wrote in :

It is d*mn good to read somebody who knows what he is talking about. Yes, the bias issue is a good tool. There is a catch however: Some old cheap tape recorders used DC bias, I once had one like that, There is also such a thing as bias symmetry, you get the lowest tape noise for a purely symmetrical bias (so bias harmonics should or could exist too). The profi recorders in the studio had a balanced bias generator for symmetry, the cheap consumers ones just a single transistor oscillator, the latter ones achieved usually no better than -45 or at the most -50 dB noise level. When the VHS video recorders came on the market, the FM modulated video signal, so a FM carrier, was used as BIAS for an analog color chroma signal. So the bias was about 1.5 MHz (sweeping), the modulating chroma about 560 kHz IIRC. Rotating heads to get that (bias = FM video) of the tape again.

Old tape days.... For forensic investigation to know details like what you describe is of high value. Few more years an nobody remembers, Maybe google will store it a while.

Posted from s.e.d BTW :-)

The story of the forensic audio analysis of the stuck mic is fasinating. A young man named Steve Barber heard something on the recording that all the experts missed which invalidated the experts analysis.

"Hold everything secure" are the famous words.

Regardless of what you think about the assasination, This is an interesting story.

Mark

And this

Enjoy

Mark

During the early 1960's, when I was starting college, I worked for a repair shop that did warranty repair for various importers, mostly on tape recorders. The bottom of the line recorders had a magnet that swung in front of the tape on record. The result was a tape that was magnetically polarized perpendicular to the tape path. Play such a tape on any tape recorder often enough, and you get the thoroughly magnetized tape head, when then had to be demagnetized. I saw various schemes to self-demagnetize the heads, none of which really worked.

In the 1960's, I had a Roberts something 1/4" reel to reel tape recorder with cross field heads. I'm not certain of the timing, but I think they were introduced in that late 1960's, a few years after the assassination.

The tape bias leak is important because the recording had to be transferred from the original Dictaphone plastic belt to reproducible media, which in its day, was 1/4" reel to reel tape. It would be amusing to obtain one of these early recordings and see how many bias leaks could be found, possibly indicating how many times the audio was re-recorded. From this web page, it's been through many hands: (3 pages). My guess(tm) is that it would take several days of DSP "sweeps" across the frequencies of interest to extract coherent bias signals and probably prove nothing more than inept handling of evidence.

I skimmed through a few of these recordings and suspect this one might be good: Not much to be seen on the compressed audio using Spectrum Lab software. I did find some noise compression, which makes me suspect that it may have been "enhanced" for clarity. I found another recording claiming to be the original, which sounded far worse.

The commentary mumbles something about crosstalk between Ch1 and Ch2 in the recorder. This is the AT2C recorder used: When one 15 min belt becomes full, it switches to the other recorder so that the first belt on the first recorder can be changed without losing anything. Therefore, the alleged Ch2 -> Ch 1 crosstalk did not happen in the recorder. The alleged crosstalk probably came from the dispatch console, which likely had separate speakers for each channel, which would partially retransmit anything heard on Ch2 onto the Ch1 recording.

Enough conspiracy theory for one day. Gone to do somethine useful.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

On a sunny day (Sat, 14 Mar 2015 10:43:39 -0700) it happened Jeff Liebermann wrote in :

I never had one, was popular with audiophiles I think, or at least marketed to that group. Akai?

Not sure about the timing, now we are in 1978 or so, but companding-expanding was used by some on international satellite telephone links that used FM modulation.

There is also tape press through? that creates a copy of the signal on the same reel where the turns overlap: (0)Enough conspiracy theory for one day. Gone to do somethine useful.

Yea, I am not going to listen to all that stuff, I remember Kennedy for wanting the moon, and it got done, his speeches gave me the creeps though.

I used to stay up to listen to Reagan...

That old recorder with DC bias I had, had a permanent magnet erase head :-) Did not even have a capstan, just drove the reels with a DC motor, variable tape speed, sort of a dictation machine, cheap though :-)