SIGNAL PATTERNS

#135 SI aka AI and patterns

Making a digital FM 10mw signal play tones which a radio can “hear”

Sound is just vibrations through the air, some we can hear and others we can’t. Different animals can hear different ranges. A dog can hear some things humans can’t.. but we can create things like SDRs that allow us to see and hear sound. Now we can make use of these two new senses unfortunately we can’t smell sound but seeing and hearing sound is good enough for a computer.

In this example I have 3 ICs 1 8-bit and 2 32-bit microcontrollers they are made by different manufacturers and are slightly different in some ways and very different in other ways. These 3 ICs are transmitting the same signal in the same way using the same code and the same transmitter with the same wire lengths and tolerances.

The variable is the IC. I have tried to isolate it as best i can, as much as possible with the equipment I have at this point in time.

Now I play the signal and record the pattern the pattern looks the same at a glance but using audacity we can “zoom in” (there’s other fancy software, but I’m using audacity for this example) we can see there is very subtle changes in some parts of the signal other parts are very easy to differentiate.

To do this analysis of the same signal generated from different devices: of the same type e.g. all using stm32 or different types stm32,py32,ch32 has to be done with a SBC or PC as decent computing power to analyse the signal is required but it’s very affordable these days…

So what happens when we analyse these subtle signal differences?

Well we are able to fingerprint the previously non fingerprintable electronic device, just by measuring these differences from a known signal that is emitting the device has already told us what it’s unique ID is…

The devices imperfections.. the devices personality becomes the fingerprint.. that unique crystal that’s slightly off frequency? Well that’s another tattle-tail telling someone a unique characteristic of your device whatever the electronic device may be.. mobile phone, bank terminal, aircraft avionics, google glasses or.. your pace maker.. your sick family members medical equipment. The vulnerable always become easy prey first.

Similar types of fingerprinting has been done in the past though now with AI there’s been a huge boom.. AI has made it so easy to control through technology on scales that have never been seen before using patterns that were very difficult to find or process in the past.

Philosophical tangent starts here XD

With the explosion of AI(SI) it is now possible for a minority group to implement a global device tracking system with much less employees, resources, restriction, laws and accountability… but most important of all.. humans will struggle to ~prove the use or non-use of these tools without a reasonable doubt.. without the use of another AI.. and can it be trusted? Especially when lives are on the line? used as a tool with a human master who has strong ethics, morals and accountability with extraordinary judgment… well maybe.. but lets be honest.. how many of these people exist?

It may sound very negative and anti AI but on the contrary I find it extraordinary useful with the potential to be a net positive.. it’s just that humans have a pattern written in the blood of many that shows without a reasonable doubt that this will be used for evil in some way or form and we are not ready for the consequences of that..

It can’t be taken lightly this is the biggest evolution since the internet, and it’s extremely hard for the average joe to build any business or life or use it (AI SI) without requiring payment of some kind to the AI overlords. I feel as if this will be used more like a new kingdom making weapon, subjugating the world to a new boot… the artificial intelligence boot.. we will see the return of a new world one ruled by AI monarchies… the British empire will look like child’s play to the new AI empire and the minority group of elitists that hold the power like gods with everyone being connected.. controlling anyone through “the click of a button” will most certainly become the norm… There are a lot of scary things in the world.. real scary things and this is one of them no nonsense this is real not science fiction.

Luckily.. as with everything this outcome “depends” on events occurring a certain way. There is a chance that things can go the right way for once. Though the circle will go on and on and on history repeating itself or rather rhyming with itself and mocking the current Ignorant humans in power, the time it takes to be born go through school and start a life just takes too long.. and there is too few people passing down genuine knowledge to their offspring, getting them up to speed as fast as possible without the poison of other educators is getting more and more difficult. Here’s where I’ll just say it.. if a vampire lives forever it would be of great use to find one and have a quick chit chat around this time… guaranteed you’d get something useful out of the interaction.. as long as you survive.

Of course I hope I’m wrong, I want to be wrong.. but our fellow man never cease to disappoint us.. just as we disappoint ourselves sometimes. I really hope that a lot of smart people can figure out a way to make this work and actually benefit everyone.. kind of like electricity benefits all.

However, perhaps we should go off to Mars, make it realistically possible, two worlds are better then one, just like people thought aircraft were impossible and just like a caveman couldn’t even comprehend or fathom the use of what a mobile phone is other then maybe to use it as a light source until the battery went flat then perhaps it would be used as a decorative piece. It may be that there is a technology, a force or a power right under our noises and we just don’t have the understanding.. the drive.. just not ready yet but its there.. patiently waiting to be found as long as there is someone to find it.

~C.A Torino

PORTABLE ALLSTAR NODE

#134 Making a better Allstar Node

Soldered node and PI Zero 2W
Kicad design
Front
Back

While looking for solutions to connect to my radio clubs repeater systems. I considered a few solutions, eventually I settled on the Allstar asterisk system. This allowed us to connect analog radio signals over the internet similar to traditional repeaters without sounding too much like a VOIP system.

The system was created by a ham for hams with lots of settings and capabilities as well as tons of customization options. This sounds complicated but it doesn’t have to be. The beauty of the system is that you can decide to just buy a node or voter or make your own hardware. There’s options to cater to a large range of ham operators, there’s even commercial licensing (asterisk) and systems for business operators which is also nice. Having options is always good in a robust communication network.

Pogo pins for data with a choke to filter the data lines

After making a few DIY nodes I wanted to make my own complete professional PCB node unit with quite a few requirements.

Namely:

  • 0805 parts.
  • Affordable and efficient as possible but also prototype friendly.
  • Easy layout for quick soldering.
  • Extra fuses and protection.
  • Reverse polarity protection USB-C.
  • Built in battery (LiPo pillow cells) commonly available & large capacity.
  • Short or limited no (wires flopping around)
  • Good grounding.
  • Good 50v or 35v capacitors.
  • Calculate PCB thickness/shielding and filtering for 50ohm impedance.
  • Solder bridge points and lots of headers for testing and debugging.
  • Antenna fits in a monopole way so try have decent ground plane.
  • Built in efficient linear charging circuit (avoid noise as much as possible)
  • Power path circuitry to allow charge and power (need good wall adapter)
  • Built in step up to 5v high current for PI and radio.
  • 5v step up converter with high efficient and frequency (min 500khz)
  • Use Raspberry PI zero 2W hat approach
  • Pogo pins for USB data & choke
  • CH340 type of UART to TTL IC for programming SA818.
  • USB differential pair tuning.
  • 4v linear regulator for filtering before powering radio module.
  • Uses SA818 modules (UHF/VHF FM capability)
  • Ability to connect any radio if needed (so not forced to use the SA818 module)
  • Proper radio filtering for UHF/VHF FM options.
Blank PCB top & soldered box below

After taking all these points into consideration I was able to do a bit of research and just a tiny bit of testing and then some more testing and then some more… Finally I was happy with the PCB I had sourced a decent box and design concept and all the parts allocated. I calculated all the costs at time of purchase and manufacture and sent to Gerbers off.

A few weeks later I was greeted with some decently manufactured PCBs.

On fitting the actual PCB to my enclosure I noticed the edges needed to have a more aggressive chamfer… Which I was able to do manually with a metal file tool. I did this outside with eye protection and a proper mask with respiratory particle filter. FR-4 carbon fiber dust in no joke.

chamfered edges
PI close up

After about a year of testing and playing around with different configurations I’m quite happy that this project worked out really well.

The two biggest concerns are chip availability and time designing, researching and developing.

Creating a commercial version of any project or product has a few other challenges, the big one being “bureaucracy” sadly there’s not much that can be done and “the process” can never be trusted.. we can only suffer through it and hope to be a better person at the end of the day.

Video demonstration here.

ROBOGUARD PROTOCOL

#133 Decoding an ASK Alarm Protocol

A few years ago I wanted to add multiple devices to my Roboguard HQ. Unfortunately I had to make my own custom solution because the original product can only allow up to 8 zones for sensors. I created a prototype using the classic Atmega328p but since then I’ve made 3 PCB designs all with different ICs: Attiny, STM8s and STM32 versions, all working with great success

Now I would like to talk a bit about the RF side of things and since I am a licensed ham radio operator as well. I think sharing a blog post will be quite interesting.

So first of all I needed to use an SDR with the correct antenna for the correct frequency in this case it was the tried and true good ol standard (ISM 433.92mhz). Since I’m located in a rural area the air waves are relatively uncongested so for me this is great.

Now I needed to record the signal I used a few tools mainly SDRSharp (because I like writing my own private plugins in C# ) Universal Radio Hacker and Audacity. Honorable mentions go to paint.net I used it for the picture editing

So firstly I observed my device for a few days, triggering it and pressing buttons etc. all while locked on to the frequency. Each time it did something I would record the time and make a high quality .wav recording of the signal.

Once I had learnt a few of the patterns and habits of the device i was able to use this in conjunction with the recorded wave files in audacity to start putting meaning to the patterns of bits and bytes.

Though to make my life a bit easier I calibrated Universal Radio Hacker for my RTL SDR and mad some recordings using some .complex16s files.

The nice thing about URH is that it tries to figure out the patterns for you but I find there’s always a bit of manual labour needed so clankers-0 meatsacks-1 in this regards

URH helped clean up the signal a bit but I still needed to make some fine tuning adjustments and re record the signals a few times with different crystal offsets.

Eventually I got decent results and was able to easily read the bitstream.

So the basic signal is 1x preamble then 8x repeating pattern containing the id and type of signal being sent.

looks like this:

Now everything is about timing. The preambles timing is very different to the payloads timing sure you can just grab the payload and call it a day but the preamble is critical in this day and age. It helps prevent false positives and also adds a bit of time so that the listener can “wake up” and grab the signal reliably so it has and important purpose.

So I used URH since it captures the timings in μs where Audacity can’t really show you accurate timings.

And I came up with the values like in the picture.

This is a nice and simple AM signal to reverse engineer and so far I have not seen anyone else do this for this specific signal so I think I’ll add it to the RTL_433 devices list.

Now once we understand the signal we need to send that signal and receive that signal.

I have done this in bare-metal C programming for the Attiny, STM8 and STM32 microcontrollers as well as added the decoding library to the RTL_433 project on Github.

The cool thing about this was that now I got to extend my bit banging knowledge using bit arrays and shifting bits around. Plus really simplicity is genius in this case. complexity just confuses an already confusing pattern of radio waves, the simple solution is always there as humans we are mostly just to dumb to see it and require time running our biological processors thinking about it until one night suddenly there’s a solution.

The main purpose was to make a cool system for myself at the time but now the secondary objective is to get more people interested in Radio and signals, SDR making plugins etc.

Reverse engineering this signal required multiple different skills and the ability to learn new things so it’s a wonderful educational and useful thing to do with a youngster.

In my opinion it opens the door to curiosity about further complex signals in a moderate way, the individual could then take up more demanding tasks like decoding FSK and other more complex signals like digital voice DMR as an example.

RADIO REPEATER MONITOR

#132 Designing a Radio Repeater Monitoring PCB

3D model of the PCB

About a year ago I realized the need for a dedicated repeater monitor that’s isolated from my clubs ham radio repeater equipment. E.G (power supply VOTER and Radio equipment). Having a dedicated monitoring PCB with a few extra complimentary features like remote control switching and reading/triggering capabilities would be very useful for largely remote and isolated places, also I get to test out and make some cool stuff.

PCB design

So I convinced myself to make the RRM V1.0 radio repeater monitor based around the well known and beloved ESP8266. Well what about the ESP32 you may ask? Yes I have made a new and improved version with asynchronous reporting and many other hardware improvements however I intend to sell that as a commercial product and thus I will not be giving away to many details for free but for a small fee you too could have one in your hands so now hopefully you see my simple strategy.

Soldered PCB

Though I designed this board for monitoring repeaters it can also be used to measure current temperature and control inputs and outputs for a number of other appliances like monitoring the current, temperature and input outputs of a vehicle. With the WiFi capability the device can be connected to as an Access Point as well for mobile operations, of course connections to the internet through WiFi is required to send the sensor and peripheral data to the server so that reports and nice colourful charts can display the data in a useful logical manner.

Here you can watch a YouTube video where I discuss the design at our Ham radio bimonthly meeting a few months ago

I decided to make a version 1 SMD PCB with large 1206 components as a prototype just to see that everything works correctly. I chose KiCad as my design software and got to work.

My requirements were:

  • 1206 SMT Parts
  • Current Measurement
  • Internet Capability
  • Reed In Detect
  • Digital Input
  • Digital Output
  • Voltage In Detect
  • Battery Back Up
  • Temperature Sensors
  • High Resolution ADS1115 ADC
  • IO Expander
  • OLED Screen
  • Reset Button
  • Control Button
  • M3 sized Screws
  • USB-C
  • Buzzer
  • Notification LEDs
3D back of PCB

So I got down to work and created a prototype around the an ESP8266 module. Note: although this module has been around for a while there’s a lot of different variants available however the EPS32 will be used in my commercial version as it has much better performance and features as well as long term support.

Designing the PCB took a few weeks of fulltime checking and double checking and then triple checking XD. Then eventually I send the PCBs to be manufactured. I soldered everything myself and confirmed everything was working correctly. Next I just experimented a bit with different components, led colours and temp sensors etc.

Bare PCB

Now I was ready to Install the module for tests at my local repeater site. We have WiFi and power at the site so I was able to connect all the hardware up and make things look neat.

Everything worked well and has been for over a year now so the project is a success I just need to monitor it long term.

FLUX DANGERS

#131 Flux causing strange IC behaviour

Low solids runny liquid flux is the best

When I ran out of my professional electronics flux I decided to try and use some off the shelf flux from the local construction store.

The flux was quite thick and seemed to work quite well when soldering parts but after cooling down some time later it forms a gunky wax that is conductive and causes all kinds of interference(Guess how I know) and worst of all it gets under IC’s and is almost impossible to remove entirely without de-soldering cleaning and then re soldering while using a very runny flux.

I tried cleaning with paint thinners and alcohol multiple times but in the end I had to remove the whole IC and clean it thoroughly

Even after multiple cleans you can still see the stubborn flux gunk on the pins as soon as i desoldered the IC
Side view of the gunk shorting out pins (Very difficult to remove when the IC is soldered on the PCB even after multiple washes and scrubs)

When the flux dries it can cause a few issues: gunk up of the pins and getting trapped under the IC

Backside is also gunky
This flux caused electrical issues on my IC’s pins…

RADAR SENSORS

#130 A Few DIY Radar Sensors

Testing the current (multimeter was used for accurate uA measurments)

One sensor that’s been incredibly useful to me is the radar sensor. Over the past few years I’ve made quite a few different versions with great results and many improvements especially in power consumption. One of the most useful features is the ability to place wood or plastic over the entire sensor effectively covering the entire unit from sight. This allows the device to be placed covertly in very effective positions. The biggest issue is ensuring the alarm signal can be transmitted from these locations and that’s where LoRa technology comes into play.

My first wireless Radar prototype used a 12V 23A battery and used a lot of power

RF 2.4GHZ and 433MHz,Wifi and LoRa are some of the most well known and common low bandwidth digital wireless communication methods, of course we could use a classic analogue radio to send digital square wave signals like the very early alarm systems but that tends to make the device a power hog and increases the size of the device however more power can also have great pros like increasing the transmission and, having less interference affecting the signal and even some level of immunity to jammers.

My second radar sensor rechargeable uses less current and made use of deep sleep but still not good enough for me…

However I always focus on low power and low current applications I want my sensors to do the job in remote areas with solar power or hefty batteries running them for years without breaking the bank.

During my journey I started with basic breadboard projects moved to more permanent perfboard and strip board projects and eventually started creating fully fledged PCBs for these devices. I encountered various problems like matching antennas to increase the transmission effectiveness, waterproofing and powering the devices with solar and batteries, the effect the blazing hot African sun has on enclosures outside for years, consuming the least amount of current and running the microcontrollers in the most effective configuration suited to their purpose and I can go on.. there’s always something new to learn and I bet there will be even more advancements in Radar technology assuming solar flares or nuclear war or maybe even aliens don’t destroy our electronic and electrical technologies. we have our ancestors to thank for creating and sharing this power with us over the many many decades and hopefully we will eventually evolve to colonize the stars… well ahem I guess that’s a bit ironic coming from a South African but ideas are stronger than any country government or religion I can only hope we keep moving forward.

Front view of my 3rd radar sensor uses 100uA when in deep sleep mode and about 600uA when running and around 10mA when TXing for about 1.2s. Uses HC7333 regulator and a rechargeable LiPo battery. SYN1115 used to TX ASK alerts. Rd-04 Module Ai-Thinker X-band radar is used.
Back view

DIY ROBOREMOTE REMOTE

#129 Building a DIY Robo remote

My prototype using a LAN cable breakout enclosure made it small and nifty.

A few years ago while using many wireless systems on a large remote property I had the problem of testing the signals in varies areas before installing my hardware like: Roboguards or my own custom devices that have the ability to communicate with the RoboGuard ecosystem.

Front view of the PCB. It fits perfectly into an ABS enclosure.

So I decided to create a dedicated remote that uses the ASK protocol used with these devices. This makes my life so much easier because it allows me to test multiple requirements at once.

I’m able to test:

  • Signal strength in the area
  • 433 modules I want to use
  • Antennas I want to use
  • The transmission logic I want to use
  • Different ASK protocols

All this with a wireless, rechargeable and easily customizable package. At the moment this is not a commercial product however I have many other devices that work with the Roboguard ecosystem that have commercial applications.

Back view of the PCB. This was made around the ATtiny212 MCU and the SYN1115 transmitter.

DIY ENEGIZER

#128 making a simple energizer

Since I’ve been living in a rural area for the past few years I’ve had to come up with some affordable solutions to problems unique to my situation. One of these problems is keeping animals out of certain areas with methods that wont cause permanent harm but will definitely be effective. So the logical solution was to use electricity in the form of electro shocks. I searched for commercial solutions and found a few but they were a bit overkill/overpriced and of course not very hacker friendly… so I decided to make my own which I can always scale up with microcontrollers/relays/monitoring and all those nice to have features but for this one I wanted it to be plain old dumb without any programming required.

So I did my research and came up with a suitable schematic found the correct parts and settled on the good old 555 timer. Designed a very small PCB and got it manufactured a voila the idea became a reality and after a few tests it works perfectly fine.

Now the shock is created by the collapsing fields of the ignition coil I used for the project. I got the most basic simple coil that is driven by a suitable N-FET at a frequency and strength that I can fine adjust via a potentiometer on the PCB. The whole system runs on a 12V battery and consumes very little current.

I took inspiration for this device and modified and created my own flavour. Like everything in life no one can do everything by themselves so I would like to credit the source which got me started here.

With that being said I had a lot of fun making this device and it’s been working well I was also able to adjust the voltage to a very humane jolt so smaller animals know not to enter the area but also don’t get fatally hurt.

DUAL BAND ANTENNA FIX

#127 Fixing a Faulty Handheld Antenna

A while ago I purchased a few hand held radio antennas from a reputable source. Unfortunately one of them had a great SWR for 2m but a terrible SWR for 70cm. So I began investigating this and started to take apart the antenna to find out more. Just as a side note: I would not recommend this for beginners or businesses (if you have a faulty product immediately contact the seller and get a replacement or refund)

So during my autopsy of the antenna I was able to determine a few things. Firstly the plastic connectors in the middle of the antenna is supposed to have a loaded coil but there was just a crimped thinner antenna wire. Second on opening up the base of the antenna I was surprised to find out that they had a metal enclosing case which was nice.

Then looking at the SMA connection to the antenna I was pleased to see a base tunning coil and capacitor, this is a good sign. potentially meaning the antenna is a “GOOD COPY” of whatever the original antenna was. the antenna has zero markings but we can take a guess that its a good copy of the Diamond RH951

Finally I found the culprit.. the capacitor on the coil seemed to be destroyed.. after de-soldering it and testing it turned out to be a faulty part. So in order to fix the antenna I soldered another cap with a close enough value.. (13pf instead of 12pf) and sure enough after putting everything back together a gluing it the antenna worked perfectly but did have a slight frequency shift when Compared to the others I had purchased.

still this is a mission success the antenna works perfectly in both bands and I’ve been using it for over a year with both bands and no issues.

ALLSTAR NODE

#126 Creating An Allstar Node

Display prototype on an actual Bread Board 🙂

In my previous article I was focused on the RTCM and the VOTER PCB which are used with repeaters. however I found out you could make a node instead for personal use. So basically a node is a radio connected through a sound card to a raspberry pi acting as an Allstar server.

You setup and connect this to the internet and you have a home made personal gateway to the Allstar servers. now you can listen or chat with anyone on the Allstar network

With that being said I was very interested in making my own version but I wanted it to use LiPo battery power for portability and a solar panel as power intake. I wanted a complete stand alone unit that could be setup once and placed somewhere and almost forgotten just remembered for maintenance and check ups.

I also wanted clean audio which could be a whole article in itself but for this write up I’ll be brief.

Testing out different filters and super capacitors
Eventually settled on a case with very short wires

I got the initial setup working using a UHF only Analog radio connected to a CM108 sound card that is plugged into a raspberry pi 2w with an OTG adapter. (it’s very important to keep all the wires as short as possible to prevent them from becoming antennas!!)
This was connected to a 5V step up converter and charger in 1 module. Now the radio needs max around 4.2V so I had to add an additional stepdown converter to accommodate this.

CM108 FOB

Everyting worked well and the system was portable but there was a very annoying whine. with some filtering I was able to reduce this significantly and the setup was almost done.

One annoying issue was that when the 5vstep up converter switched from main to battery power there was a very brief short delay where power to the raspberry pi could be lost casing a brownout and reboot of the pi. The solution was to add super capacitors in parallel with the supply to keep the volts flowing during this dip in power.

Also using a linear voltage regulator with a high ripple rejection circuit helped reduce the audio whine and noise

Over all this was a fun success of a project I will be creating a permanent case for my DIY version and I will be creating my own custom Nodes with a professional PCB for sale in future!

COS wires location on the radio
RX, PTT and GND solder areas

Passionate about technology!

Test