and where you've been going. And what you've been wearing...
[NY Daily News]
The NYPD is amassing a database of cell phone users, instructing cops to log serial numbers from suspects' phones in hopes of connecting them to past or future crimes.
In the era of disposable, anonymous cell phones, the file could be a treasure-trove for detectives investigating drug rings and other criminal enterprises, police sources say.
Read more:
formatting link
----------
Question to our knowledgable folk here: is enough of the phone's ID transmitted in the clear when it does the periodic "here I am" ping that people could track it? (Aside from the cellco, of course).
In other words, could the NYPD, now that it's got this database, use its own receivers to keep maps of everyone's travel?
_____________________________________________________ Knowledge may be power, but communications is the key snipped-for-privacy@panix.com [to foil spammers, my address has been double rot-13 encoded]
If the cell phone system can track the mere existence of a cell phone and it's nearest tower and if the software exists in the cell command and control ssytem then I see no reason why the "volume of calls and callers would be too high to be tracked." Just throw some more hardware at the problem.
It isn't just governments, anyone who can get their hands on this information can effectively track someone for any sort of nefarious purpose - I wonder how much the industrial espionage sector has benefited by having access to the effective movement patterns of individuals courtesy of the equipment the individual themselves use?
The argument that this data - which is in the hands of private companies - is secure may not hold much water with those who have seen supposedly confidential data leak out of these places on a regular basis - and we only ever hear about a fraction of the actual number of security breaches!
-- Regards, David.
David Clayton Melbourne, Victoria, Australia. Knowledge is a measure of how many answers you have, intelligence is a measure of how many questions you have.
Yup. Telcos have staff people that do nothing -but- handle those requests from L.E. It takes a court order, but in any big city, there are judges spend almost all their time approving those, and other similar, routine subpoena requests. Total turn-around time is typically 'same day', from starting to prepare the request for the subpoena to 'data in hand'.
Yup. Retained "on-line" for at least 6 months. Accessible from "off-line" storage for considerably longer.
Wrong. A fully-populated exchange (10,000 lines), with an -average- of 100 calls/day/line (laughably high), generates 1 million call records/day. assume an average of 80 bytes/record (which is _really_ high, since you can get source number in 32 bits, called number in another 32, start time and end time in 32 bits each, and have 32 bits of flags in only 20 bytes) . That's 80 megabytes/day/exchange. A
-year- of that data will fit on a 16gb thumb drive. At a _retail_ cost of $30 or less. In perspective, that is $0.003 per line, _per_year_. The _biggest_ part of the cost is indexing the physical storage, so that one knows _where_ to find the media with the desired records on it. This may amount to a few hundred dollars per C.O., per year. An actual retrieval from that stale data will likely cost hundreds of dollars -- most of it in the manual labor needed to get the stuff out of 'dead storage' and to somewhere it can be machine-processed.
Well, if you mean the 1950's or so, that may have been correct. _ANYTHING_ with 'Centrex'-type capabilities could generate SMDR-type records for every call, incoming or outgoing. If a switch can do it for Centrex service, it can do it for all users as well. And, historically, _did_. And still does.
Only a small minority of land-line phone service is 'flat rate' based, even today -- business service is all 'metered'. and to 'audit' such a bill for accuracy, you have to show when, and _to_where_, each and every call was made.
There is an entire industry out there based on doing this kind of double- checking, for the purpose of keeping the telco 'honest' in their charges.
Doing this for 'local' calling is as important for a business with a large local customer base as it is for one with a national base.
It would take a truly _freak_ set of circumstances for that kind of gross 'location error' to happen.
The base-stations engage in constant inter-communication, with regard to who 'hears' which phones with what strength. And the 'best tower' wins. *AND* when 'who hears best' _changes_ -- as when the phone moves -- the phone will be 'handed off' to the new 'best tower'.
In order for a phone to lock up with a 'distant' tower, for basic 'housekeeping' purposes, it would require that *every* 'closer' tower be getting a poorer signal.
Without knowing the 'actual facts' in the case, "across town" would imply that the alibi location was _several_ cells removed from the location of the tower that was communicating with his phone. IF that is the case, the odds of his _phone_ being where he said _he_ was are vanishingly small.
Now, it _is_ possible for a *call* connection to be handled by something other than the 'closest' tower -- for instance when all the voice slots on the close tower are already in use. But there is a
-clear- trail showing anything of that sort in the cell-system internal logs. (This kind of stuff is "critical" info for planning purposes, with regard to determining when a cell needs to be 'split'.)
***** Moderator's Note *****
Freak circumstances abound in the radio world, and it's entirely possible that a cell site that's not physically "closest" to a phone might be the one with the best signal strengh.
As far as the signal from the phone to the cell tower, something as trivial as a reflector in the near field could dramatically raise the signal strengh at a distant tower while lowering the strength at "close" towers. in addition, various "ducting" effects from the buildings that line a street, aircraft reflections, race conditions in the data network and control logic, and channel loading factors all affect which tower is the "best" choice for a given call at any given instant. It's certainly possible.
As far as the signal from the tower to the cell phone, I don't know. They're definitely _NOT_ the same thing: spread-spectrum receiver design is where the big boys play, and their sandboxes are at the labs of the major equipment manufacturers. I'm not an expert in the field, but it stands to reason that there will be substantial variations between the receiver quality from one manufacturer to another, and that means that the handsetmust play a role in the tower selection process, because otherwise it's possible to get a problem called "Hidden Transmitter Syndrome", where an "Aligator" (Large mouth, no ears) handset can interfere with a channel that it can no longer receive.
Disregarding a bunch of 'practical' constraints, "yes".
It'd take a sh*tload of receivers -- ideally one that is *co-located* with each cell base station, with an antenna of equal (or nearly so) quality.
If the various wireless carriers (the actual license holders) are co-located, you can get away with only one receiver in that location. OTOH, if they're at _different_ locations, you need one set-up for each carrier's base station.
And then there are the legal issues.
Again, as a _practical_ matter, it's easier to have the cell carrier do it for for you. *IF* law-enforcement has a legitimate need to track a particular phone, one that they can demonstrate to a judge, then CALEA access should provide the necessary data.
I have seen many freak propagation events at what should be line-of-sight frequencies.
I live in Pahrump, NV, separated from Las Vegas by a mountain range whose peak elevation is about 8000 feet above local terrain. One night I had a 2-way 10-minute contact on 2 meters (146 MHz) ham radio with a station on the other side of the mountains, direct from his car to my car, no repeaters. We tried it again for several nights following, but never could repeat it. We must have gotten a lucky bounce off something in the atmosphere.
In the middle 1970's, I lived in southern New Hampshire, 20 miles north of Boston. One night, while watching TV Channel 2 from Boston of the air, the picture got strong interference. When the interference cleared, I was watching channel 2 from New York City, 200 miles away. The New York signal completely overrode the Boston signal. This lasted a couple of hours. I never saw the phenomenon again.
......... The whole point was that the signal from this handset was acquired by one of the many directional antennas in the opposite direction that it was supposed to work because the radiation envelope still had a significant node that would work in the direction the handset was claimed to be located by the defence.
Anyone that has worked with microwave equipment knows that unidirectional antennas are not 100% directional, and there are significant usable nodes outside of the direction most of the signal is supposed to go.
IIRC in this case the handset was (claimed to be) reasonably close to the base station on the opposite side of the direction the prosecution claimed as part of the evidence that resulted in a conviction, but post-trial other experts showed that it could well have been located where the defence actually claimed with all the factors of base station operation taken into account.
The main point was that the initial evidence showed an over-simplistic description of "perfect" antenna radiation patterns of all the microwave antennas on this tower (which was probably initially done with good intentions to simplify things for a non-technical audience - the jury) and told that the signal could *only* have come from one direction, which was patently incorrect.
-- Regards, David.
David Clayton Melbourne, Victoria, Australia. Knowledge is a measure of how many answers you have, intelligence is a measure of how many questions you have.
At the frequencies, ranges, and power levels used by cell phones, such freak circumstances are _very_ remote.
A planar reflector could increase the signal a maximum of 3db at said 'distant' tower. Any other -- higher gain -- shape would require that the reflector keep a constant position relative to the handset, an accuracy requirement measured in inches.
Near-range 'ducting' effects are, indeed, possible. BUT, like 'focused' reflective effects, they tend to be _very_ localized. Like a 'whispering chamber', move a _fraction_ of a wavelength, and the effect disappears.
'Aircraft reflections' are a _momentary_ effect, at most. Any such reflection path is guaranteed to be significantly longer than the direct line-of-sight path. To make up for those simple distance losses, the reflector must have a gain of better than unity -- usually significantly better. This requires maintaining a consistent bearing in relation to _both_ the transmitter and the receiver. If _any_ of the reflector, the transmitter, or the receiver, are in motion, maintaining that relationship gets very *VERY* difficult, for anything more than the most fleeting interval.
'Channel loading' is irrelevant to the common control channel that the phone uses for registering and keep-alives. It has a _lot_ to do with which tower the handset uses for an actual _call_. Of course I already covered that. :)
Yup. it is a bi-directional communication. A tower responds to an initial call from the phone with an "I hear you, how do you hear me?" interrogatory. Only one tower respond at a time. which base responds is based on receive signal strength, as reported to 'master control'. The set is expected to respond with an "I hear you (xmtr id) this well (number)." With no response received in a 'timely' manner, control has the 'next best' receiving point transmit an interrogatory. Repeat down the line until at least one response is received. If the set reports marginal reception, continue down the line. After 'master control' decides which tower is best at the moment, it has that station initiate the downward negotiation of the phone transmit power.
Aside: Cell base stations use higher power than handsets, thus the set generally has a -much- easier time hearing than talking.
Alligators are, for practical purposes, not a problem in cell service. If a phone is 'near-deaf', it attempts to communicate only at relatively infrequent intervals -- a few milliseconds per attempt, at intervals in the tens to hundreds of seconds, is relatively immaterial. 10,000 alligators, say, in close proximity does make for a 'swamp' that one would have problems draining. :)
***** Moderator's Note *****
All your points are correct: ducting, airplane flutter, etc., *ARE* momentary phenomina.
The question is if one of those phenomina could have lasted long enough to cause the cell phone in question to be "associated" with a cell tower far away from its actual location.
As for channel loading, I was referring to the _system's_ need to balance the number of handsets which are associated with any given tower, so as to assure the greatest likelihood of having channels available for sets approaching each site, sets likely to originate, etc. My point was that the system may have chosen a less-than-perfect tower for that cellphone if the towers which were 'nearby' in electronic terms all had too many calls anticipated or in progress.
Long story short: even if the tower in question did not have a minor lobe on its antenna facing in an unwanted direction, I'd be suspicious of any 'location' information derived from what might have been a transitory event. GPS I might believe, but not this.
A nicely charged E layer no doubt. We're entering solar max again which means that the layers are going to get a good dose of charge.
***** Moderator's Note *****
I hear this sunspot cycle is turning out to be a dud. That's really too bad, since high sunspot numbers mean lots of unusual radio propagation, which makes ham operators happy. Of course, it also affects commercial, police, fire, and other public safety users in certain bands, such as the 30 to 50 MHz band in the U.S., so there are lots of non-ham users who'd just as soon do without sunspots entirely.
Yes, the cost of storage is down _relative_ to just a few years ago. But there is still a cost to the hardware, power consumption, security, a/c, and floor space. It still adds up.
Further, these days we store a lot more information about a transaction as compared to the past since we have the room for it. We also have cross references and indexing which take up space and CPU cycles, not merely plain raw sequential files.
Bummer.
I often originate cell phone calls from the exact same static physical location. Yet on the bill three different tower locations (towns) are shown for the various calls. That is, the same location is handle by at least three different towers in different towns, and probably more.
On the road, I once made a call and it was shown as carried by a tower in a town 30 miles away.
What troubles me is that skilled scammers will forge someone's supposedly secure computer account ID so they get blamed for a computer crime.
The cost of a local call in my area has been and remains one message unit--7c. Back in the days when all they had were line meters, 7c was equivalent to 70c, and for a busy business, that added up. But back then businesses had to accept the meter value (per line available in Centrex, but still a meter). Suburban calls had multiple units, again, only the meter value. I believe for special audits they could add a _special_ register tracking device, but this was not the norm.
Today, 7c isn't that much. Companies freely allow guests to make local calls and even provide phones in reception areas for that purpose; in the old days there would be a pay phone.
I find it hard to imagine a company auditing a bunch of 7c calls to extract out the time and to whom called. They already know from either Centrex or modern PBX records calls from a given extension. (In the old days a PBX operator could log local calls if mgmt so desired.)
The phone bills I have seen still only list message units, not itemized local calls.
If one employee made 1,000 local calls a month, at $0.07 that would come out to $70.00, which doesn't seem to be worth the expense of an audit. Penny smart dollar foolish.
In snipped-for-privacy@bbs.cpcn.com writes: [snip]
Otoh, this can come in handy when, for example, you're in Canada near Niagara Falls and you "force" the phone to use the US based tower so you don't get hit with the international roaming rates.
(You can do this with the GSM system by telling your phone to register with the T-Mobile signal as opposed to it grabbing the closer Canadian "Rogers" tower. Don't know if you can work similar tricks with the others).
In the mid-1990's I was driving east on I-10 in Columbia County, Florida. There was a huge crash in the westbound lanes. I called
9-1-1 to report it, but had no signal. I kept trying and finally reached a 9-1-1 center in Jacksonville, some distance away from my location and quite a distance from the crash site. I explained to her about the crash and all she could do was give me the direct phone number of the Columbia County Sheriff's Office. She claimed to have no way to contact them herself nor could she contact the Florida Highway Patrol. I thought both of her excuses were rather thin. But I was amazed my cellular signal made it so far.
With my old analog phone the 'home' region was rather small (unlike today), and roaming was $1/minute. However, the phone had a distinct easy-to-see yellow blinking light that flashed if roaming was on. The literature warned that service boundaries were not exact. Sometimes the same location near the boundary would be roaming, other times not. At the time all I could do is drive closer to home to get away from the roaming territory if I wished to make a call. (Back then the phone plan was priced for urgent calls and used mostly as such but the monthly fee was very low.)
I don't know if modern phones display an icon when roaming, but for us old folk those icons are very tiny, hard to see and hard to distinguish. I miss the simplicity of my old analog phone but I do like the very light weight and extra features (like alarm clock and speakerphone) of the modern phone.
(I just got a new phone, an LG simple model, and I hope it works out better than the Motorola did).
As an aside, Centrex could be provided by either crossbar or step-by- step, though not panel. Wiring SxS to be Centrex was not hard to do (see the Bell Labs history Vol II), but of course features were limited. Our system had ONI to record the calling number, but only for toll calls, not local calls. The answering switchboard was a old style cord board, I think a 603.
One characteristic of the early Centrex was manual call transfer. If an extension flashed the switchhook, the attendant would come on and handle the transfer; later Centrex gave a dial tone and allowed the extension to dial it themselves. That may have required ESS for "Centrex II".
On the topic of freak radio transmission, I heard a story that I wonder is true or possible: (it happened about 20-30 years ago).
The Phila transit system had a radio system for supervisors, as did most carriers. One day the dispatcher received a call that went something like this:
"This is the J line car, we are derailed". "Huh? J is a bus route. Where are you" "Market Street" "Huh? What are you doing there?"
It turned out the Phila dispatcher was getting a car, through a freak radio disturbance, from a San Francisco streetcar which happened to use the same frequency.
Was it possible for the radio systems of 30 years ago to freakishly propagate coast-to-coast? Frankly it sounds a bit far fetched to me. Note in those days the radios on the supervisors' cars used those big whip antennas and perhaps more power than units of today, and the base antennas were large towers.
***** Moderator's Note *****
Vehicles that have large 'whip' antennas - the kind with a large spring at the bottom - are using the "VHF-Low" band, which spans ~30 to ~50 MHz. Since this band is subject to a variety of over-the-horizon propagation affects, and (according to Wikipedia)
1979 was near the high point in the sunspot cycle, it's very possible that someone in Pennsylvania was talking to someone in California.
One of the first uses of a Centrex-like service was "in-dialing" to military bases. At Fort Sill, Oklahoma, site of the Artillery School, they simply gave it a prefix that trunked right into the military-owned and -maintained SxS switch.
Fort Sam Houston in San Antonio had it a few months earlier.
Anyone who ever tried to reach someone at a military base before that service went into operation will appreciate what an advance it was.
Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.