AERIS-10 Deep Dive: How an Open-Source Phased Array Radar Brings Echolocation from Military to Makers

> "What I cannot create, I do not understand." — Richard Feynman

Forget jargon like "phased array," "pulse compression," and "Doppler FFT" for a moment. Start with a more fundamental question: what does a radar actually do?

The Essence of Radar: Echolocation Like a Bat

When a bat flies in darkness, it emits ultrasound and listens to the echoes to judge obstacles—their distance and direction of motion. Radar does exactly the same thing, just with electromagnetic waves instead of ultrasound.

You emit an electromagnetic pulse; it reflects off an aircraft, car, or drone and comes back. Measure the time from transmission to reception and you know the range. Measure the frequency shift (Doppler effect) and you know whether the target is approaching or receding. Measure the tiny arrival-time differences across multiple antennas and you know the target's direction.

It's that simple. All the complex terminology wraps around three basic questions:

• Range: time × speed of light / 2
• Velocity: amount of frequency change
• Direction: phase difference of signals received across antennas

Phased Array: Magic Without Rotating Antennas

A traditional radar is like a searchlight—the antenna physically rotates to sweep the sky. That's slow, clumsy, and mechanically wearing.

The phased array insight: with a row of antennas, if you precisely control the time delay (phase) of each antenna's signal, the beam adds constructively in one direction and cancels elsewhere. Nothing needs to rotate—pure electronic control.

Imagine a row of people clapping in unison. If each claps slightly offset in time, the sound combines into a thunderclap in one direction while canceling in others. Controlling that offset controls where the sound points. That's beamforming.

AERIS-10 uses 16 antenna elements with ADAR1000 phase shifters to achieve ±45° electronic scanning, plus a stepper motor for 360° mechanical rotation to cover the full airspace.

PLFM: Why Chirp Signals

AERIS-10 uses PLFM (Pulse Linear Frequency Modulated) signaling. Instead of a fixed-frequency "beep," it emits a chirp whose frequency sweeps from low to high, like a bird call.

Why? The resolution trade-off. To see far you need long pulses (more energy), but long pulses mean poor time precision—two closely spaced targets produce overlapping echoes you can't separate.

The clever part of LFM chirps: the frequency differs at each moment in time. Even when two targets' echoes overlap in time, their frequency content differs, so pulse compression (matched filtering) can "untangle" them—like distinguishing a violin from a piano playing simultaneously because their pitches differ.

AERIS-10's FPGA performs this pulse compression in real time: the ADC captures echo data, then I/Q downconversion, decimation, FFT, and matched filtering yield high-resolution target information.

Two Versions: Nexus and Extended

• Nexus: 3 km range, 8×16 patch antenna array, ~1W × 16 power
• Extended: 20 km range, 32×16 slotted waveguide antenna, 10W × 16 GaN power amplifiers

The split is practical. Phased array radar costs concentrate in the antenna array and power amplifiers. Nexus uses patch antennas—cheap, easy to manufacture, suited to short-range applications (drone obstacle avoidance, close-range monitoring). Extended uses slotted waveguide antennas—more complex, more efficient, and with GaN PAs reaching 160W total, suited to long-range search. Users pick the version matching their scenario instead of paying for capability they don't need.

Why Full-Stack Open Source Matters

What struck me most isn't any single spec but the project's completeness:

• Hardware: full schematics, PCB layout, Gerber files, BOM
• Firmware: FPGA signal processing (VHDL/Verilog), STM32 control code
• Software: Python GUI, map integration
• Documentation: system architecture, debugging guide, test reports

The licensing is also well chosen:

• Hardware: CERN-OHL-P (explicitly addresses patent risk for hardware designs)
• Software: MIT (maximum flexibility)

This isn't a "toy" project. From the AD9523-1 clock generator (low-jitter clock distribution) to the INA241A3 current sensing (PA bias calibration) to the CIC/FIR filter chain in the FPGA—every link is serious engineering.

The 16.4k stars say one thing: many people want to play with radar, but until now there was no door.

The Signal Processing Pipeline: What Happens in the FPGA

After echoes enter the system, the FPGA (XC7A50T) executes:

1. ADC acquisition: digitize the raw RF signal 2. I/Q downconversion: shift RF to baseband, separate real and imaginary components 3. Decimation: reduce data rate while preserving effective bandwidth 4. CIC/FIR filtering: remove out-of-band noise 5. Pulse compression: "untangle" overlapping echoes via matched filtering 6. Doppler FFT: extract velocity information 7. MTI (Moving Target Indication): filter out static background (ground, buildings) 8. CFAR (Constant False Alarm Rate): adaptive thresholding so only real targets are reported

This pipeline runs in real time on the FPGA—not post-processing; results must appear within microseconds of a pulse.

The STM32 handles "housekeeping": power sequencing, GPS/IMU interface, PA temperature monitoring, stepper motor control. FPGA focuses on signal processing, STM32 on system management—a clean division of labor.

Limitations and Blind Spots

1. Alpha stage. The README is marked Status: Alpha; some features are still in development. Not a criticism—that's how open source works—but "buy and use" expectations will be disappointed.

2. Assembly barrier. The project requires "PCB assembly experience." This is not an Arduino—there's 10.5 GHz RF circuitry, PA thermal management, FPGA debugging. Even with everything open-sourced, few will successfully bring it up.

3. Regulation. 10.5 GHz at 160W total power requires radio transmission licensing in many countries. I didn't see band-compliance discussion (e.g., FCC Part 15, ETSI) in the docs. "Can use" doesn't mean "legal to use" for open-source hardware.

4. Still not cheap. Despite being "low-cost," the 16 ADTR1107 front-end chips, 4 ADAR1000 phase shifters, ADF4382 frequency synthesizer, and XC7A50T FPGA aren't inexpensive—the full BOM likely runs into thousands of dollars. "Low cost" is relative to military radar (millions).

5. Competitive ecosystem. The project overlaps with Analog Devices reference designs (e.g., the ADAR1000 evaluation board) and SDR community projects (e.g., USRP + GNU Radio). AERIS-10's advantage is integration (full stack, complete system), but it may be less flexible than pure SDR solutions.

A Feynman-Style Summary: What Is Truly Understood

AERIS-10's core insight is small but fundamental:

The high barrier to phased array radar isn't the complexity of the principles—it's the absence of a complete, reproducible, end-to-end engineering implementation.

The theory is in textbooks. The chips are purchasable. But stringing together clock distribution, phase calibration, pulse compression, thermal management, PA bias closed-loop control, and GPS attitude correction—and publishing it all as open source—that's something nobody had done before.

As Feynman said: "What I cannot create, I do not understand." This project makes "creating" a radar system possible. Understanding follows.

As for PLFM, CFAR, MTI—those are just names. Knowing a chip is called ADAR1000 and understanding "how phase control steers a beam" are two different things. AERIS-10's value is that it gives you the chance to genuinely understand the latter.

---

Reference: Richard Feynman (his lecture style on waves and interference)

Project: AERIS-10 / PLFM_RADAR (https://github.com/NawfalMotii79/PLFM_RADAR)