RDFGS 2.0: Hardware Device... (1 Viewer)

xydrine

Vengeance. Justice. Fire and Blood.
Advanced User
Lifetime Premium Member
Joined
Oct 28, 2010
Messages
26,920
Reaction score
23,936
Location
/dev/null
I've been very interested in this and have been reading these forums to gather information, but I'm also still not really sure what is being built here.
I understand, you will be provided more information as time goes on. We're not quite ready to announce anything just yet.
 

InsipidMonkey

Premium Monkey
Advanced User
Premium Member
Acceptus
Joined
Mar 22, 2017
Messages
5,674
Reaction score
11,083
Location
New England
Arduinos/microcontrollers are definitely not a handicap. In fact, they are probably better for what we want to do.
If all you want to do is have a low-level serial interface with a detector it makes sense, but it wouldn't be capable of running SDRs for ADS-B/VASCAR or alerting you to nearby police radio traffic. It also wouldn't be able to support a display, store data or interface with a web API without a smartphone. Given the design specs/features proposed in the RDFGS 2.0 threads, a microcomputer seems necessary.

I wonder if there is an Industrial spec version of the RPi's that would be better suited for 24/7 automotive use?
 

SquirrelMaster

Has a subscription with CHP
Advanced User
Lifetime Premium Member
Acceptus
Joined
Dec 3, 2015
Messages
2,533
Reaction score
4,797
Location
Liberal California
If all you want to do is have a low-level serial interface with a detector it makes sense, but it wouldn't be capable of running SDRs for ADS-B/VASCAR or alerting you to nearby police radio traffic. It also wouldn't be able to support a display, store data or interface with a web API without a smartphone. Given the design specs/features proposed in the RDFGS 2.0 threads, a microcomputer seems necessary.

I wonder if there is an Industrial spec version of the RPi's that would be better suited for 24/7 automotive use?
Respectfully, I have to disagree with you. Maybe its because we are on the same page but not understanding each other so I'll start by explaining myself. I'm not talking about a $0.05 microcontroller that is only good for blinking an LED and spinning around a small fan. I'm talking about an ARM processor that is more than powerful enough to drive whatever we want. It is still a microcontroller but you have multiple cores, display interfaces, whatever. I believe with a properly selected micro, we will have no issues at all.

If we went RPI, we would be at the mercy of their hardware and software developers. Creating our own unit puts it in our complete control.

Haven't done much research on using SDRs so any comments on that from me you should not really listen to, but I do feel like it would not be an issue. An external DSP and better ADC would be good enough(?)? I'll have to buy some SDR modules and start playing around with them before I can make any claims tho.

So I guess what I am saying is that a microcontroller is almost necessary. From there we would have to either make it a beefy one, or add some microCOMPUTER to it. (in this case, you mentioned an RPi). To me personally, that does not make sense because of the extra hardware and software we have to deal with, as well as the HUGE spike in cost. Almost any device you buy today is low level microcontroller based for good reason.Hey, take a look at virtually any radar detector or laser jammer. They all have one or more microcontrollers in them. ALP has two beefy ones just in the CPU alone.

And to answer your last question, there is the raspberry pi compute which is designed to be embedded kind of. Very bare bones and features a memory module style connector to break out all of its pins. (not too good an idea for automotive because vibrations but whatever) There is also the raspberry pi zero. It is super small and features bluetooth and wifi. Could be soldered on as a daughterboard.


Let me know your thoughts :)
 

InsipidMonkey

Premium Monkey
Advanced User
Premium Member
Acceptus
Joined
Mar 22, 2017
Messages
5,674
Reaction score
11,083
Location
New England
There is also the raspberry pi zero. It is super small and features bluetooth and wifi. Could be soldered on as a daughterboard.
I was thinking along the lines of a Raspberry Pi Zero W. Built-in BT/WiFi, USB support for SDR dongles, display output, GPIO, etc. Microcenter periodically has these for $5. Alternately you could go with a full 3B+ for ~$30.

Can you link to some of the microcontrollers you are picturing?

--- DOUBLEPOST MERGED ---

If we went RPI, we would be at the mercy of their hardware and software developers. Creating our own unit puts it in our complete control.
How so? The RPi runs linux. It would be much easier to write complex software for a RPi than an embedded microcontroller. Hardware would be cheap and open as well.
 

dchemist

Lifetime RDF Contributor
Advanced User
Lifetime Premium Member
Joined
Jul 26, 2017
Messages
2,979
Reaction score
5,410
Location
Benton, AR
Respectfully, I have to disagree with you. Maybe its because we are on the same page but not understanding each other so I'll start by explaining myself. I'm not talking about a $0.05 microcontroller that is only good for blinking an LED and spinning around a small fan. I'm talking about an ARM processor that is more than powerful enough to drive whatever we want. It is still a microcontroller but you have multiple cores, display interfaces, whatever. I believe with a properly selected micro, we will have no issues at all.

If we went RPI, we would be at the mercy of their hardware and software developers. Creating our own unit puts it in our complete control.

Haven't done much research on using SDRs so any comments on that from me you should not really listen to, but I do feel like it would not be an issue. An external DSP and better ADC would be good enough(?)? I'll have to buy some SDR modules and start playing around with them before I can make any claims tho.

So I guess what I am saying is that a microcontroller is almost necessary. From there we would have to either make it a beefy one, or add some microCOMPUTER to it. (in this case, you mentioned an RPi). To me personally, that does not make sense because of the extra hardware and software we have to deal with, as well as the HUGE spike in cost. Almost any device you buy today is low level microcontroller based for good reason.Hey, take a look at virtually any radar detector or laser jammer. They all have one or more microcontrollers in them. ALP has two beefy ones just in the CPU alone.

And to answer your last question, there is the raspberry pi compute which is designed to be embedded kind of. Very bare bones and features a memory module style connector to break out all of its pins. (not too good an idea for automotive because vibrations but whatever) There is also the raspberry pi zero. It is super small and features bluetooth and wifi. Could be soldered on as a daughterboard.


Let me know your thoughts :)
I've got two SDRs that I played with for a bit. They're nothing fancy but they're yours if you want. If you're interested, drop me a PM with your address and I'll send them to you.

Posted from my Pixel 2 using the RDF Mobile App!
 

SquirrelMaster

Has a subscription with CHP
Advanced User
Lifetime Premium Member
Acceptus
Joined
Dec 3, 2015
Messages
2,533
Reaction score
4,797
Location
Liberal California
I was thinking along the lines of a Raspberry Pi Zero W. Built-in BT/WiFi, USB support for SDR dongles, display output, GPIO, etc. Microcenter periodically has these for $5. Alternately you could go with a full 3B+ for ~$30.

Can you link to some of the microcontrollers you are picturing?

--- DOUBLEPOST MERGED ---


How so? The RPi runs linux. It would be much easier to write complex software for a RPi than an embedded microcontroller. Hardware would be cheap and open as well.
There's a virtually infinite amount of chips we can use. Here is one example.
This one has a built in DSP, encryption accelerator and hardware rng, even canbus if we want to connect to vehicles' obd ports and change things based on speed/rpm/OEM buttons pressed, it has USB OTG, built in capacitive touch sensing, etc.
https://www.nxp.com/docs/en/data-sheet/K66P144M180SF5V2.pdf


The RPi running linux is my issue. There is so much there out of our control such as the RPi's hardware. **** me if I have to diagnose an issue on it. Then so much stuff going on in the background behind of our code. Yes I can 100% agree that with a rpi and a small micro, this can be quite powerful.

My thinking as to why I want to stay away from the RPI is all based off of my experience with it:
  • It was a lot more expensive than building standalone hardware that is more than capable of running what we'd need.
  • It was horribly unreliable from what I have seen. A combination of being fussy over power (2 amps btw for RPi 3B), relying on an SD card, and not being built to be very robust has made me not trust them. I constantly have them or their SD cards fail on me. Plus, I do not think they are built to withstand the vibrations and heat of being left/used in cars.
  • It would add a lot of bulk to add a whole RPi in (even the zero z) I was picturing something about as thick as an iphone and as wide and long as a credit card.
I totally agree that it would be great to have something like an RPi because of future possibilities of adding AI or something else crazy, but I dont think those benefits outweigh any of the drawbacks. I think a lot of it would depend on how much data analysis/computing happens on the rdfgs hardware device vs the RDFGS servers.
 

InsipidMonkey

Premium Monkey
Advanced User
Premium Member
Acceptus
Joined
Mar 22, 2017
Messages
5,674
Reaction score
11,083
Location
New England
There's a virtually infinite amount of chips we can use. Here is one example.
This one has a built in DSP, encryption accelerator and hardware rng, even canbus if we want to connect to vehicles' obd ports and change things based on speed/rpm/OEM buttons pressed, it has USB OTG, built in capacitive touch sensing, etc.
https://www.nxp.com/docs/en/data-sheet/K66P144M180SF5V2.pdf
That certainly looks plenty capable, but the chip itself is ~$20, on top of the PCB and everything else required for IO. Plus, IMO the development would be much more specialized and challenging. Is there a dev board available?

The RPi running linux is my issue. There is so much there out of our control such as the RPi's hardware. **** me if I have to diagnose an issue on it. Then so much stuff going on in the background behind of our code. Yes I can 100% agree that with a rpi and a small micro, this can be quite powerful.

My thinking as to why I want to stay away from the RPI is all based off of my experience with it:
  • It was a lot more expensive than building standalone hardware that is more than capable of running what we'd need. It seems like it would be cheaper ($5-$35) than building our own microcontroller board with one of the above chips.
  • It was horribly unreliable from what I have seen. A combination of being fussy over power (2 amps btw for RPi 3B), relying on an SD card, and not being built to be very robust has made me not trust them. I constantly have them or their SD cards fail on me. Plus, I do not think they are built to withstand the vibrations and heat of being left/used in cars. I have at least 5 RPi's, three of which run 24/7 and I've had no issues. Someone mentioned they ran one in their car for a year to collect ADS-B data and had no issues either. I guess it would depend on the components used and the environment, but in my experience reliability has not been an issue.
  • It would add a lot of bulk to add a whole RPi in (even the zero z) I was picturing something about as thick as an iphone and as wide and long as a credit card. The full 3B+ is slightly larger than this, but the Pi Zero W is quite compact (and much thinner due to the lack of headers and USB/LAN jacks).
I totally agree that it would be great to have something like an RPi because of future possibilities of adding AI or something else crazy, but I dont think those benefits outweigh any of the drawbacks. I think a lot of it would depend on how much data analysis/computing happens on the rdfgs hardware device vs the RDFGS servers.
This is a helpful discussion, but maybe it should be tabled until the necessary capabilities and features a RDFGS 2.0 device should have are discussed. Once we know the scope of the project, it will be much easier to figure out the hardware requirements.
 

SquirrelMaster

Has a subscription with CHP
Advanced User
Lifetime Premium Member
Acceptus
Joined
Dec 3, 2015
Messages
2,533
Reaction score
4,797
Location
Liberal California
That certainly looks plenty capable, but the chip itself is ~$20, on top of the PCB and everything else required for IO. Plus, IMO the development would be much more specialized and challenging. Is there a dev board available?
Again, it was just one example from an infinite list. Theres plenty more chips that are cheaper and moer capable. I picked this one because its from a dev board I have used before and really like.

This is a helpful discussion, but maybe it should be tabled until the necessary capabilities and features a RDFGS 2.0 device should have are discussed. Once we know the scope of the project, it will be much easier to figure out the hardware requirements.
Totally agree :) Thanks for the discussion and the points to think about.
 

dchemist

Lifetime RDF Contributor
Advanced User
Lifetime Premium Member
Joined
Jul 26, 2017
Messages
2,979
Reaction score
5,410
Location
Benton, AR
Gentlemen and scholars... There's the RDF I know and love!

Posted from my Pixel 2 using the RDF Mobile App!
 

xydrine

Vengeance. Justice. Fire and Blood.
Advanced User
Lifetime Premium Member
Joined
Oct 28, 2010
Messages
26,920
Reaction score
23,936
Location
/dev/null
Gentlemen and scholars... There's the RDF I know and love!

Posted from my Pixel 2 using the RDF Mobile App!
You better believe it!

RDF is going to be going HEAVY into development once we have XF2 up and firing on all pistons. I am hoping to make a portion of the forum specifically for development - similar to XDA-Developers.
 

dchemist

Lifetime RDF Contributor
Advanced User
Lifetime Premium Member
Joined
Jul 26, 2017
Messages
2,979
Reaction score
5,410
Location
Benton, AR
You better believe it!

RDF is going to be going HEAVY into development once we have XF2 up and firing on all pistons. I am hoping to make a portion of the forum specifically for development - similar to XDA-Developers.
Very off subject, but what would be the best place to start learning about the above mentioned? I had started with codeacademy but even that quickly went over my head. I've built computers, installed software, rooted phones, fixed macros, etc. but I flat out suck at reading code.

Posted from my Pixel 2 using the RDF Mobile App!
 

xydrine

Vengeance. Justice. Fire and Blood.
Advanced User
Lifetime Premium Member
Joined
Oct 28, 2010
Messages
26,920
Reaction score
23,936
Location
/dev/null
Very off subject, but what would be the best place to start learning about the above mentioned? I had started with codeacademy but even that quickly went over my head. I've built computers, installed software, rooted phones, fixed macros, etc. but I flat out suck at reading code.

Posted from my Pixel 2 using the RDF Mobile App!
Related to coding in general or as it relates to the RDFGSv2 stuff?
 

dchemist

Lifetime RDF Contributor
Advanced User
Lifetime Premium Member
Joined
Jul 26, 2017
Messages
2,979
Reaction score
5,410
Location
Benton, AR

SquirrelMaster

Has a subscription with CHP
Advanced User
Lifetime Premium Member
Acceptus
Joined
Dec 3, 2015
Messages
2,533
Reaction score
4,797
Location
Liberal California
@InsipidMonkey I know we left off this conversation with me saying we only need a decently powerful micro and you saying we need a raspberry pi but ultimately deciding to let the feature requirements dictate what we use.
I've been doing a bit of research towards your side and while I am still against the raspberry pi, I have had some other ideas.

I've looked at a lot of single board computers like the raspberry pi.
My criteria has been:
  • No SD card as main boot drive
  • Smaller footprint
  • Ideally open source hardware.
I looked at some devices such as the intel Edison which runs a yocto linux on it. I have used the Edison before for a video wall project and generally like it. It requires no external storage and has plenty of options and peripherals such as built in wifi. It is also very small. Issues with that are that all of its pins are on a connector. Generally trying to avoid connectors but the edison one is quite okay. Also, not sure what its current support is. IIRC they were talking about discontinuing it.

Another device I looked at is the VoCore2. Its a super tiny board that runs OpenWRT/LEDE. VoCore | Coin-sized Linux Computer Nice thing is how tiny it is as well as how opensourse it is. Not too sure how powerful it is nor how much power we would need it to have but I can order one and play around with it. It has built in wifi plus gpio pins and interfaces out the ass. Here is an example video of it "playing" doom:
Does not look too powerful, but does run a form of linux kind of...... That plus it being completely open source may make it a viable option.
If these small modules that run openwrt turn out to be usable, there is always gl-inet's modules such as: Domore Core - GL.iNet Docs

I can order a bunch of these and play around with them. @dchemist sent me some SDRs so I can see how many of these options will support one (or more ;) ) SDRs....

Plenty of other options that we can use but I just wanted to give you some examples. Let me know what you think!
 

jakevsnake

Premium RDF Member
Intermediate User
Premium Member
Joined
Oct 16, 2015
Messages
406
Reaction score
279
Location
Tampa, FL
I am just getting started in the Software Development field but I would be interested in participating in this project if its something I can handle. Right now most of my coding has been in Java. I know Python is the hot language right now especially with Arduino.
 

xydrine

Vengeance. Justice. Fire and Blood.
Advanced User
Lifetime Premium Member
Joined
Oct 28, 2010
Messages
26,920
Reaction score
23,936
Location
/dev/null
I am just getting started in the Software Development field but I would be interested in participating in this project if its something I can handle. Right now most of my coding has been in Java. I know Python is the hot language right now especially with Arduino.
Cool good stuff, right now it's really myself and @SquirrelMaster who will be doing the primary development/coding of the app/service/device but we've started talking briefly about how we want to do alpha/beta testing and possibly adding additional people to the project who would do more than just alpha/beta testing. I'm actually going to be posting a thread announcing all of this soon and having people start submitting their names/details/capabilities/interest because the general planning of the project is nearing the last few laps and announcement of the launch of development/all of that jazz will be coming very soon after we finish up February (RDF forum-related upgrades/cleaning month). So stay tuned for that post if this is still something that interests you!
 

mswlogo

Premium RDF Member
Advanced User
Premium Member
Joined
Nov 1, 2014
Messages
3,429
Reaction score
3,267
Location
MA
I'm a little lost on this RDFGS2 project. What is it trying to accomplish? Is it a database app? A Waze replacement? Replace the need for a phone to get live alerts?
I don't mean to depress @xydrine any further (I saw his response to your post above) but I'm a little confused too on exactly what it is.

I just clicked the "Details & Information" Sticky thread and it's empty?

I know there is a feature request thread with some ideas, but a feature request of what? I'll be honest I didn't read that whole thread. But I would have expected an outline of what RGFGS2 is before asking for feature requests.

I'll also be honest I have not read everything about this project, but I have seen quite a few posts pop up here and there and I kinda have some vague idea what it could be/might be taking RGFGS1 to another level with "live data".

But I'm a bit confused why specialize hardware is needed [based on what I'm guessing RGFGS2 is in my head] (tap RD/LaserJammer hardware?) and upload to database? So my guesstimate must be way off.

I think the project needs a good summary in the logical slot already allocated (i.e. Details & Information)

It needs a concise definition, as well as details. If it's already available, sorry I missed it. The only message I keep getting is it's going to be better than RGFGS1.

Believe me, if I'm confused I suspect others are too. And I'm posting this with good intensions for @xydrine but to share that "others" are still confused.
 

SquirrelMaster

Has a subscription with CHP
Advanced User
Lifetime Premium Member
Acceptus
Joined
Dec 3, 2015
Messages
2,533
Reaction score
4,797
Location
Liberal California
I don't mean to depress @xydrine any further (I saw his response to your post above) but I'm a little confused too on exactly what it is.

I just clicked the "Details & Information" Sticky thread and it's empty?

I know there is a feature request thread with some ideas, but a feature request of what? I'll be honest I didn't read that whole thread. But I would have expected an outline of what RGFGS2 is before asking for feature requests.

I'll also be honest I have not read everything about this project, but I have seen quite a few posts pop up here and there and I kinda have some vague idea what it could be/might be taking RGFGS1 to another level with "live data".

But I'm a bit confused why specialize hardware is needed [based on what I'm guessing RGFGS2 is in my head] (tap RD/LaserJammer hardware?) and upload to database? So my guesstimate must be way off.

I think the project needs a good summary in the logical slot already allocated (i.e. Details & Information)

It needs a concise definition, as well as details. If it's already available, sorry I missed it. The only message I keep getting is it's going to be better than RGFGS1.

Believe me, if I'm confused I suspect others are too. And I'm posting this with good intensions for @xydrine but to share that "others" are still confused.
Tl;Dr
Escort Live + Waze + RDFGS1.0 all on steroids.

Think of it this way: It is like a combination of RDFGS 1 as well as Waze on steroids. A waze made by and for enthusiasts. It will start off as an app where you can report LEOs with a bit more specific data than just present or not present (ie. what speed measurement equipment they are running). You are also not limited to ground LEOs. Should be support for VASCAR and just any way possible we can gather data on the fuzz. Based on reports and your location, you would also get data from rdfgs saying what the best settings for your detector are.

The projected future for that would also have it expand to communicate with hardware devices over bluetooth/wifi/usb/etc. You would be able to pair your V1 with the app and it would auto-report for you. Could allow you to change sweeps and bands automatically.
Could connect to a scanner and give alerts for nearby LEOs using their radios.
Could connect to laser jammers and give precise alerts to other users that a certain type of LIDAR is being used in this area.

This specific thread deals with the brainstorming of a hardware device for this so users can choose not to use an app. Hardware device would allow for everything the app does but be standalone as well as have support for Licence plate recognition to sniff out unmarkeds or whatever. Would support data from the vehicle such as speed/rpm/gear for more interesting stats. Could also be used to integrate/display data on the car's instrument cluster such as alerts from rdfgsv2 or alerts from a connected detector and use factory buttons as opposed to having extra controls for the detector, jammer, scanner, etc.
 

xydrine

Vengeance. Justice. Fire and Blood.
Advanced User
Lifetime Premium Member
Joined
Oct 28, 2010
Messages
26,920
Reaction score
23,936
Location
/dev/null
But I'm a bit confused why specialize hardware is needed [based on what I'm guessing RGFGS2 is in my head] (tap RD/LaserJammer hardware?) and upload to database? So my guesstimate must be way off.
It's not.

I think the project needs a good summary in the logical slot already allocated (i.e. Details & Information)
You guys will have a full proposal once we are done writing it - when we launch the development of the project.
 

Users Who Are Viewing This Thread (Users: 0, Guests: 1)

Latest threads

Latest posts

Social Group Activity

Forum statistics

Threads
76,597
Messages
1,166,924
Members
19,685
Latest member
lonelydesires
Top