DIRECT TO PHASE II: Frequency-Based Persistent Artificial Intelligence Tracking and Surveillance Capability

The Navy faces acute operational challenges in intercepting and characterizing covert, highly agile adversarial emissions across dense electromagnetic operational environments. Modern threat systems increasingly utilize Low Probability of Intercept and Low Probability of Detection (LPI/LPD) techniques—such as Direct-Sequence Spread Spectrum (DSSS), ultra-fast frequency hopping (UFLH), wideband Linear Frequency Modulation (LFM), Costas frequency coding, and polyphase pulse compression (e.g., Frank, P1–P4 codes)—deliberately depressing emission power densities below the thermal noise floor [Refs 1, 2]. Consequently, conventional Electronic Support Measures (ESM) and wideband radiometer/energy-detection receivers exhibit severe performance degradation, failing to detect stealth radar transmissions, covert datalinks, and distributed sensor relays. This visibility shortfall undermines the tactical sensor-to-shooter timelines foundational to Combined Joint All-Domain Command and Control (CJADC2) and Distributed Maritime Operations (DMO). To counter advanced stealth emitters, an AI-driven signal processing framework can draw upon several modular algorithmic pathways and candidate techniques, such as: Sub-Noise-Floor Cyclostationary Analysis: For instance, computing the spectral correlation density (SCD) or cyclic autocorrelation can expose covert signal periodicities (e.g., symbol rates, chipping sequences, and carrier modulation artifacts), isolating structured LPI/LPD waveforms from stationary thermal background noise and Gaussian interference [Refs 3, 4]. Time-Frequency Deep Learning Architectures: As an example for intra-pulse modulation recognition, raw wideband In-phase/Quadrature (I/Q) samples can be transformed into high-resolution time-frequency representations—such as Wigner-Ville Distributions (WVD) or Continuous Wavelet Transforms (CWT)—and processed by deep convolutional or recurrent neural networks (CNN/RNN) to classify polyphase or frequency-coded pulses without requiring prior waveform templates [Refs 5, 6]. Adaptive Multi-Channel De-interleaving: For high-density pulse environments, candidate approaches include reinforcement-learning-guided spectral search or adaptive filter banks to isolate agile hopping channels and separate co-channel, overlapping LPI signals across wide instantaneous bandwidths [Ref 7]. Calibrated Operator Decision Metrics: To support tactical watchstanders, Bayesian neural networks or conformal prediction methods can be applied to generate uncertainty bounds alongside extracted parameters (e.g., pulse repetition intervals, modulation bandwidths, chirp rates), providing validated, interpretable indicators of stealth emitter activity. SWaP-Constrained Edge Optimization: Potential implementation paths for tactical platforms include neural network quantization, weight pruning, and tensor acceleration to facilitate low-latency inference on Size, Weight, and Power (SWaP)-constrained naval edge computing hardware. Work produced in Phase II may become classified. Note: The prospective contractor(s) must be U.S. owned and operated with no foreign influence as defined by 32 U.S.C. § 2004.20 et seq., National Industrial Security Program Executive Agent and Operating Manual, unless acceptable mitigating procedures can and have been implemented and approved by the Defense Counterintelligence and Security Agency (DCSA) formerly Defense Security Service (DSS). The selected contractor and/or subcontractor must be able to acquire and maintain a secret level facility and Personnel Security Clearances. This will allow contractor personnel to perform on advanced phases of this project as set forth by DCSA and NAVWAR in order to gain access to classified information pertaining to the national defense of the United States and its allies; this will be an inherent requirement. The selected company will be required to safeguard classified material during the advanced phases of this contract IAW the National Industrial Security Program Operating Manual (NISPOM), which can be found at Title 32, Part 2004.20 of the Code of Federal Regulations.