VADUM INC — Department of Defense STTR Phase II: A20B-T002
VADUM INC — STTR Phase II award from Department of Defense.
Phase II STTR prototype / development signal
- Phase II is where Department of Defense funds deeper R&D after feasibility. Incumbents with Phase II history are serious competitors on adjacent topics.
- Use this award as past-performance context and to map customer organizations for STRATFI/TACFI-style transition planning.
- Obligated amount $1,149,875 is consistent with substantial Phase II-scale effort; compare to related awards from the same agency.
- Topic code A20B-T002 links this award to a solicitation family — search the same topic stem for incumbents and recompete timing.
- Amount
- $1,149,875
- Agency
- Department of Defense · Army
- Program / Phase
- STTR · Phase II
- Topic
- A20B-T002
- Solicitation
- 20.B
- NAICS
- —
- Place of performance
- NC
- Period
- 2022-06-01 → 2024-05-31
Description
Vadum and North Carolina State University (NCSU) will develop Simulation of Communications Circuits in the Time Domain using Reinforcement Learning (SCOUTER) – a novel machine-learning-enhanced RF circuit simulator that rapidly and accurately analyzes transient circuit behavior using complex time-frequency communications waveforms. SCOUTER will have the capability to simulate modern RF transceivers in the time domain with extremely high-dynamic range (> 160 dB), while capturing full RF device nonlinearity and multi-physics effects. The use of macro-models of nonlinear RF device components significantly shortens simulation execution time with minimal loss in fidelity. A novel neural-network-based multi-scale transient simulation enables high dynamic range for analysis of nonlinear effects in the presence of complex waveforms. The core simulation component will be augmented with automated reinforcement learning to discover novel RF phenomena in representative RF circuits of interest. The learning approach searches the multi-dimensional space of input waveform parameters, resulting in a capability that rigorously characterizes modern, complex RF circuits more rapidly and accurately than existing state-of-the-art techniques.