hydrasdr.com

@amoswap @QVHenkel That swappable frontend is where all the cost hides. LNAs, filters, matching, layout iterations... You spend more on band-aids than on the ADRV itself. HydraSDR RFOne delivers the RF performance upfront, no frontend gymnastics, at a price that makes the ADRV route hard to justify

@hydrasdr @QVHenkel Most of the drone applications will have a custom frontend put in front of it - the flexibility is in just having that part easily swappable.

Who needs a WiFi module if you can have a 6GHz (40MHz BW) SDR instead 😄 I've received some requests to add a capable radio to the high-vibration resistant variant of the PiMX8, so this version will feature an ADRV9002- and XCAU7P-based SDR. The FPGA interfaces with the iMX8M plus via PCIe Gen3

@JamesMcPherson @QVHenkel Deepace spent 2+ years engineering a multi-band pre-filter bank and multi-stage front-end for the KC908. Same as Epiq. They're not proving these chips are fine, they're proving how much work it takes to make them usable outside the lab. deepace.net/products/kc908…

“Solid for the lab and prototyping” I feel like you have judged them on hobby grade SDR’s like the B200/210. A well designed front-end is like wise an “engineering problem” that has been well solved by SDR’s used in the real world. Epiq is just one example of many doing just that.

@JamesMcPherson @QVHenkel No hate on AD936x/ADRV9002 they're solid for the lab and prototyping. But anyone designing systems for the real world needs to understand: IIP3 and dynamic range are what keep your link alive when the RF environment gets hostile. And that's where these parts hit their limits.

@JamesMcPherson @QVHenkel Up/down converters extend frequency coverage while preserving front-end linearity. Frequency range is an easy engineering problem. Poor IIP3 from a direct-conversion architecture is a fundamental one.

@QVHenkel @JamesMcPherson Attenuators improve IIP3 but degrade NF by the same amount SFDR doesn't change. You just trade "can't handle strong signals" for "can't hear weak ones." Filters help out-of-band, but on a drone you can't pre-filter interferers you don't expect.

@hydrasdr @JamesMcPherson For that purpose any decent SDR has a bank of tunable filters and attenuators to offer a continuous filter range to target your specific band

@JamesMcPherson @QVHenkel +35 dBm (HydraSDR RFOne) vs -11 dBm IIP3 is a ~46 dB gap in linearity. That's not a spec sheet detail it's the difference between a drone link that works near a cell tower and one that doesn't. 40 MHz is useless if one strong interferer drives the front-end into compression.

@hydrasdr @QVHenkel “Beats any AD chip” is a bit over the top when the RFOne doesn’t have the IBW and also can’t tune to any of the frequencies of WiFi or the drone links that the platform Lukas is targeting.

@QVHenkel ADRV9002 (same architecture as AD936x) has IIP3 around -18 to -11 dBm and NF of 8-10 dB (per USRP B210 specs with AD9361). That gives you maybe 60-65 dB of SFDR in a 1 MHz channel. Any strong out-of-band signal generates intermod products that land right on your desired signal.

@QVHenkel The WiFi chipset isn't the concern, the concern is the ADRV9002's own receiver linearity. For drone applications, the RF environment is uncontrolled. You fly near cell towers, other drones, radar, ISM-band emitters, or even your own onboard electronics.

@QVHenkel Classic example: Try decoding ADS-B or air band voice on the field with a strong transmitter nearby and a distant aircraft. The weak signal gets crushed. That's where front-end dynamic range matters, and why tuner-based architectures with proper RF selectivity win outside the lab

@QVHenkel The HydraSDR RFOne (tuner-based architecture) beats any ADRV9xxx or legacy AD936x in sensitivity and IIP3. In a lab, wideband SDR chipsets look fine. In the real world, strong nearby transmitters hit that poor IIP3 and generate intermod products that bury weak signals.

@w1wra @FlUxIuS Thanks for the feedback ! Feel free to share some screenshots with different devices decoded always great to see it in action ! (Just make sure they don't contain any sensitive data)

@hydrasdr @FlUxIuS Doing some quick testing, looks pretty slick!

Introducing hydrasdr_433 a wideband ISM band decoder built on rtl_433, supercharged for HydraSDR. Decode the entire 433 MHz, 868 MHz, or 915 MHz band in a single pass. github.com/hydrasdr/hydra… #SDR #IoT #433MHz #868MHz #915MHz #ISM #HomeAutomation #SIGINT #RF #OpenSource

@w1wra @FlUxIuS You are welcome to write your ideas here github.com/hydrasdr/hydra…

@hydrasdr @FlUxIuS Sortable columns in stats view would be nice :)

@w1wra @FlUxIuS The Web UI is now available in latest nightly github.com/hydrasdr/hydra…

@w1wra @FlUxIuS Thanks for your feedback! Something nice is brewing hydrasdr_433 is getting a full Web UI: live event stream, activity charts, device registry, wideband presets, gain modes, protocol browser... all self-hosted, zero dependencies, works offline. Stay tuned !

hydrasdr_433 deep dive 2/2 Polyphase Resampler: per-channel rate conversion to match each decoder's native sample rate. 32 taps/branch, Kaiser window 60 dB design stopband (measured 74-76 dB) GCD-based L/M ratio reduction for minimal compute Bypass mode when rates already match

hydrasdr_433 deep dive 1/2 OS-PFB Channelizer: 2x oversampled analysis filterbank splits wideband IQ into M narrowband channels (M = 2, 4, 8, 16). 48 taps/branch, Kaiser window 80 dB stopband attenuation >41 dB adjacent channel rejection <0.1 dB passband ripple

Excited to announce that HydraSDR is sponsoring @ph0wn CTF 2026 ! Ph0wn is a one-of-a-kind Capture The Flag event dedicated to hardware & smart device security and it's free for all participants thanks to its sponsors. ph0wn.org #ph0wn #CTF #SDR #InfoSec

🚀 RF Swift images v0.1.3 out! Updated Ghidra, ImHex, RF tools + new libhydrasdr v1.1.0 for HydraSDR RFOne and our special SDR++ package & more! rfswift.io