SBIR TOPIC SOCOM254-007 • THORNVEIL LLC
AURORA
Acoustic Swarm Coordination
GPS-denied drone swarm coordination using acoustic sensing. Simulation-validated across 80+ experiments with corrected methodology. COTS hardware. TRL 3+ demonstrated.
Key Results
10–50 m
Operational Range
100 m in ideal conditions
20–40 bps
Data Rate
validated; higher achievable
2.7°
DOA Accuracy
actual MUSIC, 4-mic array
12+
Drone Scaling
BER < 0.05 at 12 drones
<$25
Cost / Drone
all COTS components
80+
Experiments Run
11 core + methodology validation
RIGOROUS, SELF-CRITICAL TESTING
80+ experiments with methodology corrections applied throughout: CFAR detection thresholds (no oracle), 1000-bit BER measurements with Wilson score 95% confidence intervals, and actual MUSIC DOA processing on simulated array signals. Every claim was re-validated under corrected methodology.
CFAR Detection (No Oracle)
Matched filter threshold set via constant false alarm rate estimation from noise-only statistics — not by looking at signal-present data. Eliminates circular reasoning in processing gain claims.
+23 dB at 200 ms, +29 dB at 500 ms, +31 dB at 1 s
1000-Bit BER with Confidence Intervals
All BER measurements use 1000+ transmitted bits per test point with Wilson score 95% confidence intervals. Previous tests used as few as 50 bits, making zero-error results statistically meaningless.
Threshold confirmed at −10 dB; zero-error results bounded at BER < 0.004 (95% CI)
SNR Defined at Microphone
SNR is measured at the microphone (pre-filtering), connecting the link budget directly to BER performance. Confirmed that filtering provides 0 dB gain within the communication band — the quiet-band advantage is intrinsic.
Mic-level and post-filter SNR identical at 3–5 kHz; link budget uses mic-level threshold
Actual MUSIC DOA Processing
DOA estimates computed from simulated multi-microphone array signals with propagation delays and noise — not from synthetic angle-plus-noise. Full eigendecomposition and spectral scan.
2.7° mean error (median 0.3°); 0 collisions in integrated Boids test
Why Acoustic?
EMCON Compliant
Any RF emission is detectable by hostile SIGINT/EW systems far beyond communication range. Acoustic propagation is confined to the local environment, making interception beyond the swarm perimeter impractical. Zero RF signature during coordination.
Jam-Resistant
RF jamming degrades all RF-dependent systems simultaneously — GPS, datalinks, and inter-drone communication. Acoustic channels are spectrally orthogonal to RF jamming. Processing gain of +23 dB provides robust performance against interference.
GPS-Denied Capable
Acoustic sensing enables both communication and spatial awareness. Direction-of-arrival estimation (2.7° accuracy) and acoustic ranging (0.29 m error) provide relative positioning without any external infrastructure. The system degrades gracefully as conditions worsen.
SBIR Compliance
| Metric | Target | Result | Status |
|---|---|---|---|
| Acoustic communication range | 5–20 m | 10–50 m operational, 100 m ideal (simulation-validated, hardware pending) | MEETS |
| Data rate | Coordination-sufficient | 20–40 bps validated; higher achievable with shorter symbols | MEETS |
| GPS-denied positioning | Sub-meter | 0.29 m range error, 2.7° DOA error in integrated test | MEETS |
| DOA accuracy | ±5° | 2.7° mean (actual MUSIC, 4-mic square array) | MEETS |
| Multi-drone scaling | 5+ drones | 12 drones at BER < 0.05; hybrid MAC scales further | MEETS |
| Doppler tolerance | Operational velocities | Resilient to 30 m/s relative velocity (BER = 0.028) | MEETS |
| Wind resilience | Outdoor operations | BER below measurement threshold to 12 m/s | MEETS |
| SWaP | Minimal | < 25 g, < $25 per drone, all COTS | MEETS |
| Formation keeping | Collision-free | Zero collisions in 5-drone integration test; passive sensing achieves 90%+ performance | MEETS |
| Hardware validation | TRL 4+ | Simulation-validated (TRL 3). Hardware Phase 0 measurement is next step. | PARTIAL |