Things you can do with a Software Defined Radio (2024)

A few months ago when there was a lot of emergency services activity in my area and I didn't know why, I was reminded that no-one in my region is contributing a feed to Broadcastify.

I went down the tunnel of using SDR to recieve those transmissions, and share them online.

Then I went a bit further.

What if you could transcribe the broadcasts into something like a text feed? What if you could add location information somehow to monitor where things were going on in your region? Could you use AI to somehow organize the data into a more useful format?

What if this data was valuable? Maybe you could sell this as a service? Who would buy it? Public safety organizations? Hospitals? News organizations?

I spent a few days worth of freetime figuring out how you'd do someting like this, and got to a place where I figured it was conceptually possible.

Then somewhere in my googling, I stumbled across this site: http://citizen.com/ - and realized that someone had already turned my idea into what looks like a pretty mature product.

Ahh well. I'm sure my billion dollar idea will come later.

In the meantime, I'd still like to mess with SDR at least so I can know what's going on around me next time there's a fire or other public safety incident, before it gets reported on.

I don't know about billion dollar ideas, but I encourage you to make a product even if something similar exists.

If you squint enough there is nothing new under the sun and chances are that you will take a very long time to find something that hasn't already been done!

But doing your own product does several things - you learn a lot, you position yourself for future success, you see future ideas differently. And maybe you're okay for something to not be a billion dollar idea and you can outlast a venture funded product.

Maybe I'm just projecting, because I've put of building something for such a long time!

Citizen is really enshittified, so any alternative would be better. I don’t object to them charging for their service, but they use all kinds of predatory editorial tactics and push notifications and marketing copy to instill primal fear that your neighborhood is imminently burning down and you will get shot if you don’t subscribe to a higher tier of service from them. Crime is way down in the US but you really don’t feel that way when you are a subscriber/user of Citizen.

I was joking about the billion dollar idea.

My actual "MVP" was some kind of automated neighborhood newsletter, that'd monitor emergency services radio traffic, and put together some kind of "here's what happened in your neighborhood" daily newsletter.

Maybe I could get it packaged in a hardware/software package that let anyone set one up in their neighborhood.

But I mostly got stuck in privacy concerns. I'm not sure it's a valuable public service to let people know that, for example, someone had a heart attack a few blocks over.

I did think about the scientific value of some kind of statistical database that process and recorded emergency services calls though. But mostly, my ideas for commercial and moral opportunities were half-baked at the point that I discovered citizen.

One of the technical challenges I came up against was finding transcription software that could semi-accurately transcribe UHF/VHF radio traffic. However, it looks like there's some progress that's been made there since I last checked: https://www.rtl-sdr.com/radiotransciptor-real-time-radio-spe...

The fact that there's a mature product doing what you want to do is a good thing. It means there's a market for what you want that's large enough to sustain development.

You can easily distinguish yourself from Citizen by targeting a different demographic, different branding, different UX, interpreting the data in a different way.

Just look at how many businesses there are in any industry that deliver the same outcome for their customers but in a slightly different way.

What you're describing could be a really good news source giving live on-the-ground information to people.

> I encourage you to make a product even if something similar exists.

This is very good advice: we often give up on "great ideas" once we find that they have already been done.

But the vast majority of people we consider successful did not invent anything completely new, they just made a better kind of XYZ, sometimes not even that dramatically different. If you think about it, it's a much more logical path to success than expecting to be the next DaVinci.

> But I mostly got stuck in privacy concerns. I'm not sure it's a valuable public service to let people know that, for example, someone had a heart attack a few blocks over.

In the moment, notifying people who know CPR and may be nearby and able to get to a nearby location and start CPR before emergency services arrives is the base of PulsePoint [1], which seems like a useful public service.

As a digest, yeah, I don't think any usefulness outweighs the invasion of privacy. Maybe just a count of health emergencies responded to for observing trends.

[1] https://www.pulsepoint.org/

That was the vibe I was getting when visiting the site, they seem to understand fear pretty well. Stay away for your own mental health :-)

I’d encourage you to pursue it. I remember the old @breakingnews on Twitter when it first started, people listening to police scanners and typing info-dense one liners on what they heard. To this day the best news service of my life (until someone bought it).

A real time, AI snips version for my area in a running feed would be amazing. There are lots of formats and use cases; and the info is already out there.

It’s a great idea. Don’t let citizen sway you away from it.

The privacy concern is real and not something I'd want to think about too hard myself. One night I heard sirens and checked one of the local scanner type sites. I could hear enough about the call, and that combined with a record of previous calls to that address made me wonder if I really wanted that information. Maybe some obfuscation of the previous codes to the address would have been enough to reduce the feeling of knowing too much.

None of that is to say it isn't a good idea. I appreciate the ability to see roughly what is going on when I hear sirens. Even if the sites aren't always able to show the calls. I think highway patrol doesn't show up for me.

Many metro police are moving towards encrypted communications but it varies by location.

Regarding medical emergencies, I'm pretty sure EMS just says "medical emergency" and gives the address. I've never heard them say specific patient conditions, although sometimes the ambulance can forward that to the ER.

If there were any risk, it would be making it too easy for criminals to monitor and allow them to commit crime more effectively.

> A few months ago when there was a lot of emergency services activity in my area and I didn't know why, I was reminded that no-one in my region is contributing a feed to Broadcastify.

Maybe instead of emergency services activity, it could be other types of activities (hazards, local events, nextdoor alerts, local business/SIGs rss feeds, etc), it is all just local info and knowledge aggregation endpoints and archives that have over-the-air and terrestrial distribution channels.