1. Originally Posted by Wiredwrx
I think you may not understand how radio systems work. Basic explanation is this,

A repeater listens on one frequency (called the input frequency) and whatever it hears, it automatically transmits on a different frequency, called the output frequency. The officers radios actually transmit and receive on the exact opposite of the repeater. So, the repeater listens on "channel A" and transmits on "channel B", but the officers radios transmit on "channel A" so the repeater hears it, and listens on "channel B" so the officers can hear the repeater on "channel B"
Well, I'll admit I don't know much about the specifics of police radios, but the idea I'm trying to use would (I think) work on a more basic principle that would apply to most any system.

Originally Posted by Wiredwrx
Also, the PD has MANY repeaters, too cover the area as completely as possible. This means there is overlap of the signal. Also, it will be IMPOSSIBLE for a receiver to know which repeater it is hearing, so it can't tell if repeater A is weak, or repeater B is weak.
Well, I do understand what you're saying. You're just going to get a SUM intensity from many repeaters, and there's not going to be a nice easy way of identifying how much each repeater contributed to each signal. That's why I was saying you'd have to know the positions of the receivers w/ GPS (and your own position), so you could *calculate* what those repeaters *should* put out. Then add all of those together. If your real receiver measures intensity MORE than this by a non-negligible amount, that's the cruiser portion.

And I think you are correct, I'm assuming that the repeater is putting out the signal in the same band as the cruisers, but that actually complicates the problem. It's simpler if the cruisers are the *only* ones transmitting on "channel A". A couple questions that maybe you could explain, as I don't know much about police radio systems:

- If the cruisers transmit only on "channel A", and listen only to "channel B", is there an automatic failsafe mode if all the repeaters suddenly die, or a couple cruisers have to chase someone into some rural area with no repeaters? ie, a way for them to listen on channel A or transmit on channel B?

- Is there a provision for multiple-repeating? ie, cruiser sends a signal on "channel A", receiver boosts the signal and transmits on "channel B", then another receiver listens and re-boosts?

Originally Posted by Wiredwrx
While that is a problem of some sorts, you could always listen to officers signal, on "channel A", however, if the officer is using a protable radio, they are not very strong while mobile are much more powerfull.
If this is the case, as I hinted at above, the problem becomes infinitely easier, no GPS needed, no false alarms, etc. Yes, you have to differeniate between portable/mobile radios, but still! Let's say the mobile radios are 10 times more powerful than the portable radios. Now our unit picks up a signal. We have calibrated the system so that, if it was a portable radio, the signal looks like it is 2 miles away. If we now assume that the radio is a mobile radio, how far should we expect it to be:

Intensity = k_portable * 1 / (2 miles)^2
Intensity = k_mobile * 1 / (x miles)^2
= 10 k_portable * 1 / (x miles)^2

Solve for x.

k_portable / 4 = 10 * k_portable / x^2
x = Sqrt[40] = 6.3 miles

Not that big of a difference. We'd still want the detector to start beeping, probably.

Originally Posted by Wiredwrx
Now, remember, a mobile/portable transmits in ALL directions. ANd there will be multiple transmitters. You will never be able to accurately track the signals, and you will have no idea where they are cause they will all be moving, and you will be moving, Assuming a signal is 5 miles away (just assuming it can be determined to be 5 miles away), it is 5 miles in any direction. It just makes no sense to try.
Wait, it makes no sense to try what? Knowing that there is a cruiser in 5 miles (in any direction, unknown to us) is good enough for me. That's all I am told by my radar detector. I don't get told specifically *where* the cruiser is, just that he's somewhere close by.

Originally Posted by Wiredwrx
As you move, another signal will be 5 miles away, while the original might be 3 miles away. what about the original being 5 miles away, and then someone else tranmits 3 miles away, then the original transmits 4 miles away, so where are they now? ALso remember, that they can be 3 and 5 miles IN ANY DIRECTION. No amount of calculation power will be able to determine who is where when, and going where.
Well, first look at the simpler case where there is one cruiser, and he isn't moving. There I think we both agree on a method you could use to find the direction between you and him (drive around in your car and watch the cruiser signal intensity change). That is assuming he's transmitting the same thing over time (which is probably valid if you look at the data frequency).

Now look at the case with two cruisers, also not moving. There, you'd notice that you can't perform a k * 1/r^2 fit on the intensity data you're getting. You'd then try a [ (k1) * 1/(r1)^2 + (k2) * 1/(r2)^2 ] fit. With enough data, you can use minimization of error techniques to fit the data, and find his position.

Now if they are moving, one thing you can do is just take your sample points faster. This means less distance between readings, hence less intensity change, hence a little error matters a lot more, and you have a more vague idea about the direction they are in.

In reality, what I would do is setup about 10-20 "scenarios" (1 cruiser not moving, 2 cruisers not moving, multiple cruisers moving, etc) and for each case, try to use a statistical technique to *force* the data to conform to each pattern. You use RMS error minimizing to do this, varying the free parameters (angle of cruiser relative to front of your car, if you'd like to picture it that way) until you get the least RMS error between your intensity-position data and the intensity predictions of your model. You feed in a bunch of GPS positions and intensities, lots of data... If the simplest model has a very large RMS error, and the "next complex" model in the line is a better fit, you can let the user know that there are probably two cruisers, etc.

Anyway, like I said earlier, all this (everything since the last quote) is really not necessary, just knowing "there is a cruiser 3-5 miles in any direction" is enough for me. Anything more than that, and it sounds like you're getting the detector to plan a get-away route for a bank robbery. I just want something to tell me when to drive a little more conservatively.

2. Originally Posted by spmclaugh
Well, I'll admit I don't know much about the specifics of police radios, but the idea I'm trying to use would (I think) work on a more basic principle that would apply to most any system.

Well, I do understand what you're saying. You're just going to get a SUM intensity from many repeaters, and there's not going to be a nice easy way of identifying how much each repeater contributed to each signal. That's why I was saying you'd have to know the positions of the receivers w/ GPS (and your own position), so you could *calculate* what those repeaters *should* put out. Then add all of those together. If your real receiver measures intensity MORE than this by a non-negligible amount, that's the cruiser portion.
THat is incorrect. There is no sum intensity. All the repeaters transmit at the same time. They are all high power. They also overlap. Also, you must know that transmitter power does not sum in that way. In fact, they employ VERY sophisticated timing between the transmitters so that they do now cancel them selves out. THey must therefore be 180 degrees out of phase. This is what happens

Originally Posted by spmclaugh
And I think you are correct, I'm assuming that the repeater is putting out the signal in the same band as the cruisers, but that actually complicates the problem. It's simpler if the cruisers are the *only* ones transmitting on "channel A". A couple questions that maybe you could explain, as I don't know much about police radio systems:

- If the cruisers transmit only on "channel A", and listen only to "channel B", is there an automatic failsafe mode if all the repeaters suddenly die, or a couple cruisers have to chase someone into some rural area with no repeaters? ie, a way for them to listen on channel A or transmit on channel B?
There is always a "Talk a round" mode where the units can communicate with each other. This is on the repeater output frequency, so even if a unit is "using" the repeater, they can hear a person on the talk around frequency. However, the dispatcher CAN NOT hear that person, and neither will anyone who is far away fromt he unit be able to hear him.

Originally Posted by spmclaugh
- Is there a provision for multiple-repeating? ie, cruiser sends a signal on "channel A", receiver boosts the signal and transmits on "channel B", then another receiver listens and re-boosts?
Here is another thing. Repeaters are the boosters. They provide coverage thru improved hearing and transmitting. Most PDs link all the repeaters together. If you look above, each circle is a repeaters coverage, both hearing and transmitting coverage. They are all linked together, so that all of them transmit what one of them hears. There is no need, technically, to have a reboost of a handheld in a properly designed system. There are situations where PDs use an small in car repeater. That requires a 3rd channel. Basically, the officer carries a low power protable, transmits and receives on channel C, the car then repeates channel C to the repeater on channel B (the one the repeater listens to) and then the repeater acts normally. The car then hears on channel A, and repeats the message on channel C so the officer can hear it.

Originally Posted by spmclaugh
If this is the case, as I hinted at above, the problem becomes infinitely easier, no GPS needed, no false alarms, etc. Yes, you have to differeniate between portable/mobile radios, but still! Let's say the mobile radios are 10 times more powerful than the portable radios. Now our unit picks up a signal. We have calibrated the system so that, if it was a portable radio, the signal looks like it is 2 miles away. If we now assume that the radio is a mobile radio, how far should we expect it to be:

Intensity = k_portable * 1 / (2 miles)^2
Intensity = k_mobile * 1 / (x miles)^2
= 10 k_portable * 1 / (x miles)^2

Solve for x.

k_portable / 4 = 10 * k_portable / x^2
x = Sqrt[40] = 6.3 miles

Not that big of a difference. We'd still want the detector to start beeping, probably.
The problem being that a portable putting out 5 watts has a range of a couple of miles, lets say, while a mobile can have a range of 50 or 75. Unfortunately, range is dependent on MANY factors, transmitter power, antenna, feedline loss, and Altitude, of both the transmitting unit AND the reciever. Someone on the 100 floor of a building can hear a unit 10 miles away, while someone using the same unit to receive, int he same situation, might not be able to hear him from 10 blocks away. So while it might be calibrated to "beep" on receiving a portable, since it must be just 10 or 20 blocks away, a mobile could be on the other side of city, and a reciever will hear it. So, it will beep when the mobile is FAR FAR away. it will be worthless. You would have to listen to the portable/mobiles frequency, because as I mentioned, the repeaters will provide the same amount of power/range no matter where the mobile or portable is transmitting from.

3. Originally Posted by spmclaugh
Wait, it makes no sense to try what? Knowing that there is a cruiser in 5 miles (in any direction, unknown to us) is good enough for me. That's all I am told by my radar detector. I don't get told specifically *where* the cruiser is, just that he's somewhere close by.
Here is the thing. Besides what I mentioned above, that a mobile could "trip" the receiver and beep when across town, you have to realize that a radar gun transmitts in a single direction. It is a foruces beam, unlike a mobile or portable which transmits omnidirectionally. Next, radar guns have limited power, ont he order of less then one watt, and radars operate in the GHZ range. Thier propogation is very small. Basically, the signal won't travel very far. But PDs operate in the VHF, UHF and Ultra UHF bands. They have better propogation properties. Here is what propogation means. In the GHZ range, 1 watt of power will go 1/2 mile, while at VHF, 1 watt will go 5 miles, UHF 4 miles and Ultra UHF 3.5 miles.(these are just illustrative, like I said, there are many factors to consider, but this assumes all things are equal, and, the mileage is just part of the idea, I have no idea exactly how far a signal will go with out testing)

Originally Posted by spmclaugh
Well, first look at the simpler case where there is one cruiser, and he isn't moving. There I think we both agree on a method you could use to find the direction between you and him (drive around in your car and watch the cruiser signal intensity change). That is assuming he's transmitting the same thing over time (which is probably valid if you look at the data frequency).
Huh. Officers generally spend about 30 minutes a day transmitting, perhaps at most an hour in an 8 hour period. Most often, much less. There is no way to accurately "track" him because of this. I don't get what you mean by "which is probably valid if you look at the data frequency" What do you mean by data frequency. THough technically, if there was only one officer, and he continuoly transmitted, and you moved around, you might be able to get a relative area where here is, but it would take some time. And, if he moved, well, you would have to start all over.

Originally Posted by spmclaugh
Now look at the case of two cruisers not moving. There, you'd notice that you can't perform a k * 1/r^2 fit on the intensity data you're getting. You'd then try a [ (k1) * 1/(r1)^2 + (k2) * 1/(r2)^2 ] fit. With enough data, you can use minimization of error techniques to fit the data, and find his position.
Whos position. You have 2 signals, which are indistiguishable to the receiver and the calculations. And, they are not transmitting continuously, so the amount of data available per unit is very small.

Originally Posted by spmclaugh
Now if they are moving, one thing you can do is just take your sample points faster. This means less distance between readings, hence less intensity change, hence a little error matters a lot more, and you have a more vague idea about the direction they are in.
Again, the added problem of getting data is difficult because they don't transmit that often, they move ,A was here, but is now there, and b is halfway between A and B, then A and B are in the same place, and then they switch places, but to the receiver, they are all 5 miles away from you. And, they don't transmit in every location, and sometimes they transmit while they are moving.

Originally Posted by spmclaugh
In reality, what I would do is setup about 10-20 "scenarios" (1 cruiser not moving, 2 cruisers not moving, multiple cruisers moving, etc) and for each case, try to use a statistical technique to *force* the data to conform to each pattern.
There IS no pattern. Officers do not follow a pattern.

Originally Posted by spmclaugh
You use RMS error minimizing to do this, varying the free parameters (angle of cruiser relative to front of your car, if you'd like to picture it that way) until you get the least RMS error between your intensity-position data and the intensity predictions of your model. You feed in a bunch of GPS positions and intensities, lots of data... If the simplest model has a very large RMS error, and the "next complex" model in the line is a better fit, you can let the user know that there are probably two cruisers, etc.
HUH

Originally Posted by spmclaugh
Anyway, like I said earlier, all this (everything since the last quote) is really not necessary, just knowing "there is a cruiser 3-5 miles in any direction" is enough for me. Anything more than that, and it sounds like you're getting the detector to plan a get-away route for a bank robbery. I just want something to tell me when to drive a little more conservatively.
Back to the beginning, the problem is that there is no way to accurately tell when a transmitter is that close because of the difference in mobile, and portable power, though I guess technically, if ALL you want to know is if there is a transmitter near by, you can make the receiver near deaf, so that it won't pick up strang signals as easily, but then weak signals will have to be on top of you to receiver, or let the receiver be as is, and just get false alarms when the po po are on the other side of town.

Michael

4. Originally Posted by PandaInnovation
I haven't read this WHOLE thread, but here's something that ACTUALLY WORKS!

I have a VHF mobile radio in my truck. I have it set to receive 154.905MHz, which is the channel the the Cali Highway Patrol uses for their mobile extenders. These extenders are in the CHP cars and they allow for the officer's handpak to work with the lowband (42MHz) system. Here's a diagram:

Handpak ---> Moble Extender ---> Lowband Mobile Radio in CHP Car ---> Dispatach

These mobile repeaters (MR) are really low power, perhaps 100-200mW or so. We know the handpaks are low power too (so they only work so far from a CHP vehicle).

Basically, if you have a radio listening on 154.905 and there is a CHP unit close enough to you (less than a mile??), you'll hear whatever is coming across the radio system. This is not just limited to that particular officer's transmissions as everything is rebroadcasted out the MR.

Great tool for roadtrips down Hwy 5!!
The CHPs mobile repeater/extender is a very unique situation. Very few PDs use extenders.

Michael

5. Originally Posted by Wiredwrx
You idea about Jurisdiction difference is true, but really, you can;t get that far into the analyses, because there is no way for you to access the GPS data.

Remember that GPS units DO NOT TRANSMIT. They only recieve. The GPS information is sent on the services DATA radio network. Unlike the old motorola systems, in particular the MDT systems, RD LAP and data on new systems (such as the ones with newly installed GPS, which is a VERY new feature) are all encoded. Soo, breaking it is very difficult, if not near impossible, and it is also illegal to break the encryption.

Radio DF'ing is almost impossible for many reasons. First, Most PDs operate io-n a repeater system. So, the signals "comes" from many places. Even if you listen to the repeaters "input", sometimes the units transmit on a portable with very little power, and sometimes with a mobile, with much higher power. This will make DF'ing very diffuclt. The next thing has to do with traingulation. IN order to get a somewhat accurate location, you need to triangulate the signal. That means you will need at least 3 receivers that receive the signal at the same time. That means all three must be within receiving distance of the transmitter. This would either take 3 fixed locations, with antennas mounted High, like 20 or 30 stories high, and don't forget, for a city or town, you will have to have overlapping coverage of the antennas. In a large city, you will need many receivers. Or, if you want to make them mobile, while possible, the problem will be having at least three receivers hear the transmitter. The next issue will be, to triangulate, you have to have the exact location of the receivers that are doing the trinagulation, and they all have to be on the exact smae time. This is usually occomplished with a connection to the atomic clock, or from a GPS signal. That is not to hard to accomplish, but it must be done cause even a difference of a second will mean VERY innacurate results because the triangulation requires calculation in the nano and milliseconds range.

Then, once you overcome the trinagulation issues, the information MUST be sent somewhere, to accomplish the calculations to determine the transmitters location. Then, to top it off, police cars and officers are mobile, and they transmit often. In an hour, there will be so many transmissions from so many locations all over the city, that it will be impossible to know who is where and when. While most agencies do have som sort of ID on the radio signal, it is not easy to decode. There is no SW only decode of MDC ID, though there is some hardware out there that does allow serial output, and I have never seen a aftermarket decoder for MODAT.

Then, to top it off, the FED is requiring agencies to migrate to Apco 25. Apco 25 is almost impossible to monitor, and many of the agencies are adding encryption to the voice channels to make monitoring impossible.

At this point, it just aint worth it or possible.

Michael
Trasnsmission wasn't the idea...i woudl tell you what a friend proposed but that would be bad... Then gas prices went up to \$3 and it just wasn't cost effective to speed anymore.. Then again, they sped more than i did, so it doesn't bother me too much. And your last paragraph is the second reason why we didn't bother with my friend's suggestions

All in all, there are easier ways of avioding the polie...and sometimes the old tricks are the best ones (like making sure someone's going a lot faster than you per 12 mile stretch at night on the interstate)

6. Originally Posted by Genesisfactor
Trasnsmission wasn't the idea
What was the idea?

Michael

7. Don't take this the wrong way, but it would be pretty stupid of me to tell you, wouldn't it? Not everything i do or have access to should be shared. The comment was made as food for thought. Any of you have balls to figure it out and do it, do it. Maybe you'll figure out why else we kille dth eproject. Hypothetically, I can tell you how to steal a car, from those that need keys to those that don't. That doesn't mean i'm going to do it or that i'm going to educate others about it.

Besides, I'm not the one who you should be talking to about that, I'm just a lowly robotic engineer with ideas and useful friends who i haven't seen in months. I can't seem to recall who is the person you can talk to either.

Personally, do what they did in the fast and the furious: monitor the police freuqneces. Its only a crime to do it if you're using it to commit a crime (they'll have a hard time proving that you were using the CB to speed). Depending on your area, the system and how smart you are, see what you can do from there. Plus you get some interesting entertainment...

http://www.hamuniverse.com/shotsfired.wav

May i suggest Uniden?

8. So, a month ago I posted, and you are now responding.

How do you figure it would be "stupid" for you to tell me, the rest of the forum member how to do this. So you really think you are the smartest one here, and therefor the only one with the knowledge to do this, even though it can't be done. And how mature, you know but you're not telling anyone. WTF are you talking about. You know how to do it but you are not telling. Well, then why the hell are you posting.

How about this, it can't be done. Prove me wrong, you and your "engineering" degree can't change the laws of physics. Go ahead, show us ALL how it can be done.

Michael

9. its nothing about being the smartest one here, its about protecting oneself. I tried to use the car theft as an example of what i tried to illustrate. Hell i even said i wasn't the only brain on it. Just because you can DOESN"T MEAN YOU SHOULD. beyond that, not htat you obviously care, but recent events have made it temporarily impossible to help you as far as you seem to want me to.

Seriously though, i dont' follow up on all my posts, i've got other things to do. Hell i've got threads i've STARTED and have forgtten about for TWO MONTHS. Doesn't mean i don't care, it means i forgot. Hell, the ONLY reason why i searched for this thread is because i actually got messages for people looking fo the solution. I can't provide one, nor will i attempt to, as stated about. We never finished. The little i do recall form the project, I dont' even want to remember. Good enough for you or are you just mad that i didn't give you an answer? I mean seriously...

But to your condescentding attitude, it has NOTHING to do with physcis, because, as you stated, that method wouldnt work, however, there is more than one method out there to solve a problem. So how about this: It can't be done. if that makes you sleep better tonight, just attackign someoen who doesnt' want to divulge too much about something that will undoubtly get LOTS of people in trouble, and shuts you up, have a good rest. Other than that, there's nothing more i want to say on this topic. The project is dead.

But geez, calm down.

10. In my area, alot of the emergency vehicles emit a single to all other emergency vehicles, adn to the traffic light system.... could one tap/listen for that to build a radar like system

Page 2 of 7 First 1234567 Last

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•