Hands-on look at Radenso's upcoming new detector and AI

Token

Learning to Fly
Beginner User
Joined
May 28, 2017
Messages
53
Reaction score
85
Location
Mojave Desert, CA, USA
With the caveat that I haven’t tested it yet, I’m not so sure that is the case.


If I record encrypted police recordings and non-police recordings and then feed them to RAI to train on, I would bet money that the AI would pick up on “tells” or patterns in the police recordings that don’t exist in the other recordings. For example, if they always start their transmissions with a certain identifier for dispatch, etc.


These are things that we wouldn’t need to break the encryption to see - the AI doesn’t know what needs to be said during the transmission to tell the difference, it just needs to see something different between police and non-police transmissions.
While I think this is an interesting discussion I don't want to overemphasis it. I think it is neat to brainstorm on it, but I think it may be too large a problem to be something real anytime soon.

But with that said, you may want to take a closer look at the format of the data used in most such systems. Keep in mind I am NOT an expert on these systems, I do not work with them professionally, I have just touched on them in passing, but what follows is my understanding of them from my observations. I would be glad to be wrong, and if I am...yaaahhh! Maybe it would work as you are thinking and that would be very cool.

If you record encrypted police and encrypted non-police recordings the data format will be indistinguishable until you decrypt. The data outside the encrypted information, of course, will be usable. Something like NAC, unit ID, and TGID for P25 transmissions. But unless you have a database of known stations (probably correlated with specific frequencies) that is not going to do you any good, because the non-police transmissions will also contain the same classes (but different specific data) of information. None of that data is specific to police transmissions, it is specific to the network being received, which may be specific to agency or municipality.

What I am basically saying is that once they are digitized and encrypted there are no repeatable differences between police and non-police uses of the same encrypted digital system. The only differences are the words being said, and since that is digitized and encrypted, and not all police agencies use the same formats, that will be a problem.

I think without using frequency as a discriminator it is going to be a very tough nut to crack. Not undoable, but pretty tough. Fortunately, freq databases are readily available.

One way to use encrypted LEO transmissions is to watch only the repeater or system input freqs, vs the output freqs. The repeater outputs are going to have a large footprint, covering wide areas. But the mobile and hand held units, on the input freq, are typically going to have much more limited areas of coverage. If you hear one it is not too far away, depending on terrain and situation maybe only a couple of miles or less. Make the system trigger only on very strong signals, or not worry about developing highly sensitive receivers and antennas, and you could trigger only on things inside shorter ranges. Again, for this to work you need to know the input freqs, but like I said, the databases are widely available.

T!
 

InsipidMonkey

Premium Monkey
Advanced User
Premium Member
Joined
Mar 22, 2017
Messages
6,125
Reaction score
12,895
Location
New England
Something like NAC, unit ID, and TGID for P25 transmissions. But unless you have a database of known stations (probably correlated with specific frequencies) that is not going to do you any good, because the non-police transmissions will also contain the same classes (but different specific data) of information.
This information all exists in the RadioReference database, and is used by Uniden for their BearTracker technology. I don't know offhand what the patent status of it is, but with access to some sort of SDR API (or even a separate SDR dongle connected via USB hub), it would be fairly trivial to implement. Note that this would only work to detect active transmissions, however.

With an ultrawideband SDR and AI at your disposal, maybe it would be possible to capture the spectral fingerprint of a police cruiser based on the myriad electronic devices most contain? This is a much more difficult problem, but would be extremely interesting to work on. I'm sure there are some parallels in the defense industry, but I don't know of any specifically.
 

Token

Learning to Fly
Beginner User
Joined
May 28, 2017
Messages
53
Reaction score
85
Location
Mojave Desert, CA, USA
This information all exists in the RadioReference database, and is used by Uniden for their BearTracker technology. I don't know offhand what the patent status of it is, but with access to some sort of SDR API (or even a separate SDR dongle connected via USB hub), it would be fairly trivial to implement. Note that this would only work to detect active transmissions, however.

With an ultrawideband SDR and AI at your disposal, maybe it would be possible to capture the spectral fingerprint of a police cruiser based on the myriad electronic devices most contain? This is a much more difficult problem, but would be extremely interesting to work on. I'm sure there are some parallels in the defense industry, but I don't know of any specifically.
This is MASINT, and sometimes called hullteching. It is quite common at certain levels of activity.

Hmmm, I know I can recognize all of our local cruiser radars by their radars frequency. And yes that frequency does shift a bit with temps. I wonder if it would be possible to, with the level of detail that Radenso brings to each signal detection, "mark" each source as one would a GPS muted false alarm? Once you see a signal you mark it as whatever it is, city PD, county PD, CHP, etc, maybe even by cruiser number, and each time the system sees that signal in the future it tells tou what you marked it as in the past?

Just a thought, but to do such a thing you would have to record certain specific parameters of the signal, chirp rate, settling time, etc, and that would probably use a lot of memory.

T!
 

NVR2FST

Advanced User
Lifetime Premium Member
Joined
Aug 1, 2011
Messages
1,276
Reaction score
885
Location
Somewhere in cyberspace.
Good point, "Token"! When I cruise a specific area and I see the exact same frequency again 20 minutes later, I say to myself, "That's our old friend 34.587 again." (Or whatever the actual frequency was.)

Of course this only works over short time periods since there is only frequency to work with. Knowing what agency it is would be interesting to enthusiasts such as myself, but not crucial by any means.

Something like, "Seminole County Sheriff's Department Ford Explorer, white in color." would be way cool.
 

G37X Jockey

Advanced User
Premium Member
Joined
Jul 22, 2012
Messages
1,426
Reaction score
1,302
Location
Behind my Steering Wheel
Good point, "Token"! When I cruise a specific area and I see the exact same frequency again 20 minutes later, I say to myself, "That's our old friend 34.587 again." (Or whatever the actual frequency was.)

Of course this only works over short time periods since there is only frequency to work with. Knowing what agency it is would be interesting to enthusiasts such as myself, but not crucial by any means.

Something like, "Seminole County Sheriff's Department Ford Explorer, white in color." would be way cool.
Love it. We RDF guys don't ask for much do we?
 

Dallara

Learning to Drive
General User
Joined
Nov 11, 2019
Messages
31
Reaction score
95
Love it. We RDF guys don't ask for much do we?


Man, if it could do all that then Radenso could change the name from Theia to Crowhorse...





~
 

Token

Learning to Fly
Beginner User
Joined
May 28, 2017
Messages
53
Reaction score
85
Location
Mojave Desert, CA, USA
Good point, "Token"! When I cruise a specific area and I see the exact same frequency again 20 minutes later, I say to myself, "That's our old friend 34.587 again." (Or whatever the actual frequency was.)

Of course this only works over short time periods since there is only frequency to work with. Knowing what agency it is would be interesting to enthusiasts such as myself, but not crucial by any means.

Something like, "Seminole County Sheriff's Department Ford Explorer, white in color." would be way cool.
I live in a small community, about 90 minutes from the next larger community, with a handful of patrol vehicles with mounted radar and several hand held units. Few enough Ka sources that you can remember all of their frequencies for a given temperature, and the approximate range of change with temp. When I see 34.684 on a 70 degree morning (lower at warmer times) I know it is the mobile sign that the local PD puts out beside the road. When I see 34.728 and alternating with 34.671 I know it is the newest Explorer Pursuit and they are switching between the front and rear radars.

It would be very neat if the detector took the features of the signal, combined with the frequency, and allowed you to note that when "this shape signature, this fingerprint, combined with this frequency, is this user noted ID". Yeah, I know, that is something that not 1 in 100 users would care about, but hey, for the OCD among us....

;)

There is no doubt that this ability is potentially within the hardware and AI as described so far, but of course there is probably not enough interest in such an ability to warrant the effort to develop the logging / database / user interface required.

T!
 

G37X Jockey

Advanced User
Premium Member
Joined
Jul 22, 2012
Messages
1,426
Reaction score
1,302
Location
Behind my Steering Wheel
I live in a small community, about 90 minutes from the next larger community, with a handful of patrol vehicles with mounted radar and several hand held units. Few enough Ka sources that you can remember all of their frequencies for a given temperature, and the approximate range of change with temp. When I see 34.684 on a 70 degree morning (lower at warmer times) I know it is the mobile sign that the local PD puts out beside the road. When I see 34.728 and alternating with 34.671 I know it is the newest Explorer Pursuit and they are switching between the front and rear radars.

It would be very neat if the detector took the features of the signal, combined with the frequency, and allowed you to note that when "this shape signature, this fingerprint, combined with this frequency, is this user noted ID". Yeah, I know, that is something that not 1 in 100 users would care about, but hey, for the OCD among us....

;)

There is no doubt that this ability is potentially within the hardware and AI as described so far, but of course there is probably not enough interest in such an ability to warrant the effort to develop the logging / database / user interface required.

T!
Jon has said that Theia will be open sourced to allow developers to create apps with specific functionality like has been done for the V1. We have some very talented people on RDF, so we'll see what developers can come up with.
 

Dragons

Breathing fire
Advanced User
Joined
Nov 21, 2013
Messages
886
Reaction score
597
Jon has said that Theia will be open sourced to allow developers to create apps with specific functionality like has been done for the V1. We have some very talented people on RDF, so we'll see what developers can come up with.
I'm hoping it's open enough, to allow addition of our own radar sources, esp with Redflex NK7 as one is rarely the same as the next.
 

Jon at Radenso

Manufacturer
Advanced User
Premium Member
Manufacturer
Joined
Oct 9, 2012
Messages
1,534
Reaction score
10,831
Age
31
I'm hoping it's open enough, to allow addition of our own radar sources, esp with Redflex NK7 as one is rarely the same as the next.
What do you mean one is rarely the same as the next? If they are the same hardware emitter, the AI will view them all as the same.
 

andi

Learning to Fly
Beginner User
Joined
Oct 3, 2015
Messages
88
Reaction score
136
Location
Germany, Bavaria
Redflex NK7 is using 2 Radar Antennas simultaneously, with 24,1 GHz CW and 24,2 GHz FMCW. Imho Theia will see both signals and recognice them individually as NK7.
 

Attachments

Dragons

Breathing fire
Advanced User
Joined
Nov 21, 2013
Messages
886
Reaction score
597
What do you mean one is rarely the same as the next? If they are the same hardware emitter, the AI will view them all as the same.
Okay, I don't have massively advanced understanding in RF, so excuse the ignorant language..

The problem as I understand has been frequency hopping, the wave form out of one NK7, is different to the next, and changes again at the next restart.

@beingaware can probably explain this a little better.
 

Jon at Radenso

Manufacturer
Advanced User
Premium Member
Manufacturer
Joined
Oct 9, 2012
Messages
1,534
Reaction score
10,831
Age
31
That wouldn’t be a problem for Rai since she doesn’t care what frequency the signal is, or what the waveform looks like. She is looking at emitter characteristics on a hardware level.
 

Dragons

Breathing fire
Advanced User
Joined
Nov 21, 2013
Messages
886
Reaction score
597
That wouldn’t be a problem for Rai since she doesn’t care what frequency the signal is, or what the waveform looks like. She is looking at emitter characteristics on a hardware level.
Fair enough, then I take back my previous statement.
 

CarefulDriver

Learning to Fly
Beginner User
Joined
Jun 13, 2019
Messages
92
Reaction score
54
This might be a wrong thread to ask in, but can someone explain me what's the difference between Multaradar C and CD? And what frequencies they operate on?
 

Randy at Radenso

Radenso Radar
Advanced User
Lifetime Premium Member
Manufacturer
Joined
Sep 9, 2011
Messages
615
Reaction score
2,334
Location
Cincinnati, OH

jumperthumper

Learning to Drive
General User
Joined
Jan 21, 2019
Messages
15
Reaction score
15
i hope the production unit could be just labelled "Electronic Compass" and for the built in display just have the ability to show the direction of travel. and radenso changes the credit billing info on statements to say "safety products inc" if they sell direct with this product. no reasons
 

PaSt

Learning to Drive
General User
Joined
Dec 20, 2018
Messages
14
Reaction score
13
Location
Chicago
I very much want to do that, but it just comes down to whether we can afford to without additional NREs. We have to make a bunch of difficult design decisions when routing stuff and all of them have tradeoffs. And BOM cost is another challenge - the cogs of this detector are going to be 3-4x what that of the R7 is - at some point, we need to draw a line in the sand and be a business instead of a charity.

That being said, since we did this in a modular way, there exists the possibility for us to come up with different drop in boards. For example, for a couple hundred grand we could probably produce a different “motherboard” that gave people direct access to the baseband I/Q data. But I honestly don’t know how big that market is, so I would be open to some other form of pre-sales and then development. Maybe something like Massdrop would be a good method to see if it’s viable or not. If enough people want it, I’ll build it.
I'm all for a modular design! OMG, I am all about this market! It's like an R/C, PC hobby but for radar detectors. If the design is 1000% modular based, "but" along with changing your current line up XP, SP, Pro M, to modular, I believe you would completely dominate!

Like a case with different versions:
Case/Shell Version 1.0 allows hardware such as MB's, Horns, etc. up to Versions X.x
It would be cool for your entry level detectors to be modulars giving entry level users the ability to upgrade as days pass by.
I know it's completely [email protected], but if the design will not change for many generations, people like me spend on limited edition items, like a limited edition carbon fiber blacked out case, just to bring in additional revenue.

Though I dont need entry level, I would love modular internals!

Sign me up for pre-sales.
 

AnthonySzum

Learning to Drive
General User
Joined
Oct 28, 2019
Messages
25
Reaction score
5
This detector was built for worldwide support from day 1 of development. It is part of the reason we have so much storage and such a wide bandwidth front end. A huge chunk of the world has been completely ignored by radar detector manufacturers. There is no such thing as a truly high end radar detector being sold in Asia or Russia right now.
Have you seen youtube videos of people driving in Russia? I don't think they need a device to help get away with speeding LOL
 

Discord Server

Latest threads

Forum statistics

Threads
79,902
Messages
1,215,696
Members
20,227
Latest member
Pinhook
Top