Ham radio is well worth getting into if you come from a software background but want to get more hands-on with embedded electronics. Radios are ubiquitous in modern technology, and getting a deeper understanding of how they work can have surprising career benefits too!
The RF fundamentals stay the same, but the gulf between ham radio and modern RF comms is truly vast.
Those TDM'd bands 40MHz wide, with digital data and modulation past the limits of sanity, and the entire RF system being integrated into one die somehow? Oh boy.
What really blows me away is the range that you can achieve with almost no power on tiny little antennas. For instance, ELRS uses a transmitter/receiver that is less than a gram, that can keep a link with a drone alive across 30 km or even more. And the antenna is so small you might toss it away with the packaging if you're not paying attention.
It's a centimeter on a side, and easily goes more than 10km. It's just mind-blowing that this exists. 0.9 grams, IIRC.
I used to follow the balloon projects that hams would launch. A mylar balloon with a tiny 50 milliwatt transmitter and GPS, solar powered on the 10Mhz band tracked thousands of miles away.
Yep, its called LoRa.
Ive been able to decode as low as -26 SNR.
Theres LoRa chips for 2.4GHz, 900MHz, 868MHz, 433MHz, and 144MHz.
> Radios are ubiquitous in modern technology, and getting a deeper understanding of how they work can have surprising career benefits too!
Indeed.
The problem with many modern ham radios of any sufficiently complex feature set - especially when it comes to cheap hackable radios or digital radios - is that a lot of the functionality is hidden away in blackbox ASIC hardware blocks that have no public datasheets (e.g. BK4819 powering Quansheng's radios, Si4732, or for anything DMR, the AMBE-2020 vocoder).
It's truly a miracle what the hacker community has gotten out particularly out of the Quansheng chipset.
Job well done! I tried reverse engineering the encryption on Yamaha's midi files. I thought it would be super complex but it turned out to be ridiculously easy. It's funny when you're preparing mentally for some long slog and turns out to be an hour at best. In case you're interested: they used a fixed block of 256 bytes that they xor'd the data with in a cyclic fashion.
That's more like obfuscaton, you got lucky there!
I've reverse engineered lots of things, but the one time I actually got paid for it (this is more a hobby to me), I got the exact opposite of what happened to you.
I quoted some small amount to document the protocol to configure some embedded device that I thought would take a day or so, and it turned into a two-week nightmare. Turned out there was no configuration protocol, it was firmware updates always -- and internal parameters were just overwritten along with the code. So I ended up having to disassemble a big chunk of the firmware before I could configure the device.
Pro-tip, state your assumptions baked into the estimate. If one of them is wrong you can renegotiate price, although depending on the client, you may not always want to do that to show good will and whatnot.
Ham radio is well worth getting into if you come from a software background but want to get more hands-on with embedded electronics. Radios are ubiquitous in modern technology, and getting a deeper understanding of how they work can have surprising career benefits too!
The RF fundamentals stay the same, but the gulf between ham radio and modern RF comms is truly vast.
Those TDM'd bands 40MHz wide, with digital data and modulation past the limits of sanity, and the entire RF system being integrated into one die somehow? Oh boy.
What really blows me away is the range that you can achieve with almost no power on tiny little antennas. For instance, ELRS uses a transmitter/receiver that is less than a gram, that can keep a link with a drone alive across 30 km or even more. And the antenna is so small you might toss it away with the packaging if you're not paying attention.
One example:
https://rcmaniak.pl/userdata/public/assets/images/SpeedyBee/...
Oh, and it also speaks WiFi, just in case and it has its own little onboard computer and a web server.
I use this one, with an onboard antenna:
https://imgaz.staticbg.com/thumb/large/oaupload/banggood/ima...
It's a centimeter on a side, and easily goes more than 10km. It's just mind-blowing that this exists. 0.9 grams, IIRC.
I used to follow the balloon projects that hams would launch. A mylar balloon with a tiny 50 milliwatt transmitter and GPS, solar powered on the 10Mhz band tracked thousands of miles away.
Yep, its called LoRa.
Ive been able to decode as low as -26 SNR.
Theres LoRa chips for 2.4GHz, 900MHz, 868MHz, 433MHz, and 144MHz.
> Radios are ubiquitous in modern technology, and getting a deeper understanding of how they work can have surprising career benefits too!
Indeed.
The problem with many modern ham radios of any sufficiently complex feature set - especially when it comes to cheap hackable radios or digital radios - is that a lot of the functionality is hidden away in blackbox ASIC hardware blocks that have no public datasheets (e.g. BK4819 powering Quansheng's radios, Si4732, or for anything DMR, the AMBE-2020 vocoder).
It's truly a miracle what the hacker community has gotten out particularly out of the Quansheng chipset.
Job well done! I tried reverse engineering the encryption on Yamaha's midi files. I thought it would be super complex but it turned out to be ridiculously easy. It's funny when you're preparing mentally for some long slog and turns out to be an hour at best. In case you're interested: they used a fixed block of 256 bytes that they xor'd the data with in a cyclic fashion.
That's more like obfuscaton, you got lucky there!
I've reverse engineered lots of things, but the one time I actually got paid for it (this is more a hobby to me), I got the exact opposite of what happened to you.
I quoted some small amount to document the protocol to configure some embedded device that I thought would take a day or so, and it turned into a two-week nightmare. Turned out there was no configuration protocol, it was firmware updates always -- and internal parameters were just overwritten along with the code. So I ended up having to disassemble a big chunk of the firmware before I could configure the device.
Pro-tip, state your assumptions baked into the estimate. If one of them is wrong you can renegotiate price, although depending on the client, you may not always want to do that to show good will and whatnot.
[dead]