VisionWave Holdings Inc. (NASDAQ: VWAV) filed a U.S. provisional patent application for its SDNN Symbiotic Deep Neural Network architecture, the company announced. The filing covers a proprietary neural network system designed for real-time data fusion and coordinated control of distributed intelligent platforms across defense, security, and robotics applications.

The provisional application, filed June 4, 2026, encompasses a 455-page specification and 23 engineering drawings covering the SDNN system architecture. VisionWave internally refers to the central core layer as "Mother." The company also filed a U.S. trademark application for SDNN, which remains subject to USPTO examination.

The SDNN architecture is intended to support fusion of data from heterogeneous sensors, unmanned ground vehicles, unmanned aerial vehicles, satellite feeds, and software agents. The system includes a qSpeed reasoning engine designed to prioritize mission-critical computations and a hardware security module called The Cube for system authentication.

"SDNN represents a fundamental rethinking of how AI can coordinate distributed intelligent systems," said Danny Rittman, inventor and chief technology architect for SDNN.

The patent application describes potential applications including counter-drone defense, missile detection systems, multi-robot industrial coordination, smart city operations, and autonomous spacecraft management. The company has not generated revenue from this architecture and acknowledges substantial uncertainty regarding successful development and commercialization.

VisionWave has 12 months from the provisional filing date to file a corresponding non-provisional utility patent application. The filing of a provisional patent application does not guarantee patent issuance, and the USPTO may reject or narrow claims during examination.

VisionWave Holdings develops AI-driven sensing and autonomy technologies for defense, homeland security, and commercial infrastructure applications.