Burst Transmission Detection

The foundational energy-based approach that integrates received power over a time-frequency window and compares the output against a calibrated noise-only threshold. Burst signals with sufficient energy-to-noise ratio exceed this threshold, triggering a detection event. This method is non-coherent and requires no prior knowledge of the signal structure.

• Key metric: Energy integration gain proportional to √(TW)
• Vulnerability: Fails against LPI waveforms with negative SNR
• Implementation: Channelized radiometers split wide bandwidths into parallel narrowband integrators

A blind spectrum sensing technique that computes the sample covariance matrix of multi-antenna or oversampled received signals and analyzes its eigenvalue distribution. The presence of a correlated burst signal alters the ratio between maximum and minimum eigenvalues, enabling detection below the noise floor without requiring noise variance estimation.

• Advantage: Robust against noise uncertainty that plagues radiometers
• Methods: Maximum-Minimum Eigenvalue (MME), Energy with Minimum Eigenvalue (EME)
• Application: Detecting DSSS bursts where spreading codes induce spectral correlation

Exploits the periodic statistical properties inherent in modulated burst transmissions. By computing the Spectral Correlation Density (SCD) function, this technique isolates cyclic frequencies corresponding to the symbol rate, chip rate, or carrier offset—features absent in stationary noise. Burst signals are detected when cyclostationary signatures emerge from the noise floor.

• Discrimination: Distinguishes between modulation types by their unique cyclic signatures
• Resilience: Functions effectively in negative SNR environments
• Computational cost: High; mitigated by strip spectral correlation algorithms

Transforms burst signals into a joint time-frequency representation using the spectrogram (STFT magnitude squared) or Wigner-Ville Distribution. Short-duration emissions appear as localized energy concentrations against a diffuse noise background. This method is particularly effective for detecting frequency-hopping bursts where energy hops across discrete channels over time.

• Resolution trade-off: STFT time vs. frequency resolution governed by window length
• Advanced transforms: Wavelet scalograms for multi-resolution burst analysis
• Post-processing: Image processing techniques (edge detection, morphological filtering) applied to spectrograms

A non-coherent detection architecture specifically designed for DSSS burst signals. The received signal is multiplied by a delayed version of itself, producing a spectral line at the chip rate that can be detected with a narrowband filter and threshold comparator. This technique requires no knowledge of the spreading code or carrier phase.

• Mechanism: Exploits the cyclostationarity induced by the spreading waveform
• Output: Discrete spectral line at 1/Tc (chip rate frequency)
• Limitation: Performance degrades with low processing gain or severe multipath

Acquires wideband burst signals at sub-Nyquist sampling rates by exploiting their sparsity in the time-frequency domain. The sparse signal is reconstructed from compressed measurements using ℓ₁-minimization or greedy pursuit algorithms. This enables detection of short-duration emissions across GHz of bandwidth without requiring prohibitive ADC sampling rates.

• Enabler: Analog-to-Information Converters (AIC) with random demodulation
• Reconstruction: Basis Pursuit Denoising (BPDN), Orthogonal Matching Pursuit (OMP)
• Advantage: Simultaneous detection and parameter estimation from compressed samples

A class of signal processing transforms that map a signal's energy distribution across both time and frequency axes simultaneously. This is the foundational visualization and analysis tool for burst detection.

• Spectrogram: A squared magnitude of the Short-Time Fourier Transform (STFT), revealing how the frequency content of a signal changes over time.
• Wigner-Ville Distribution: Offers superior joint time-frequency resolution compared to the spectrogram but introduces cross-term artifacts.
• Wavelet Transform: Provides multi-resolution analysis, using variable window lengths to capture both short bursts and long-duration signals effectively.

A fundamental energy-based detection method that integrates the power of a received signal over time and bandwidth, comparing the output to a noise-only threshold to declare signal presence. It is the simplest and most common burst detection architecture.

• Channelized Radiometer: Splits a wide bandwidth into parallel narrowband channels, integrating energy in each to detect and characterize frequency-hopping signals in real time.
• Sensitivity: Performance is limited by the noise floor uncertainty; a 1 dB error can significantly degrade detection probability.
• Double-Threshold Detection: Uses a two-stage thresholding process to improve detection of weak bursts while maintaining a constant false alarm rate (CFAR).

Exploits the periodic statistical properties of modulated signals for robust identification, even at low signal-to-noise ratios. Burst transmissions often embed unique cyclic signatures.

• Spectral Correlation Density (SCD): A two-dimensional transform that measures the correlation between spectral components separated by a cyclic frequency, revealing hidden periodicities.
• Chip Rate Detection: A blind signal processing technique that extracts the fundamental clock frequency of a direct sequence spread spectrum code by detecting spectral lines in the SCD.
• Noise Rejection: Cyclostationary processing inherently rejects stationary noise and interference, making it superior to radiometry for detecting weak LPI bursts.

A blind spectrum sensing method that computes the eigenvalues of the received signal's sample covariance matrix to detect the presence of a signal without requiring knowledge of the noise floor. Ideal for burst detection in dynamic noise environments.

• Maximum-Minimum Eigenvalue (MME): The ratio of the maximum to minimum eigenvalue is compared against a threshold derived from random matrix theory.
• Covariance Matrix: Constructed from time-domain samples; the presence of a correlated signal (a burst) increases the spread of eigenvalues.
• Advantage: Overcomes the noise uncertainty problem that plagues radiometric detectors, providing reliable detection for Low Probability of Intercept (LPI) waveforms.

A signal acquisition framework that reconstructs sparse wideband spread spectrum signals from sub-Nyquist rate samples by exploiting their inherent structure. This is critical for monitoring vast spectrum ranges for intermittent bursts.

• Sub-Nyquist Sampling: Analog-to-digital converters (ADCs) sample at rates far below the Nyquist criterion, reducing hardware cost and data throughput.
• Sparsity Assumption: Relies on the fact that burst transmissions are sparse in the frequency domain; only a few narrow bands are active at any instant.
• Reconstruction Algorithms: Techniques like Basis Pursuit or Orthogonal Matching Pursuit are used to recover the full wideband spectrum from the compressed measurements.

The process of synchronizing a non-cooperative receiver with the exact switching instants of a frequency-hopping transmitter to enable subsequent demodulation and analysis. This is the critical step after detecting a burst.

• Dwell Time Estimation: Determining the fixed duration a transmitter remains on a single carrier frequency before switching.
• Transition Detection: Identifying the phase discontinuities or amplitude nulls that occur at the boundary between two hops.
• Pattern Prediction: Once timing is recovered, the hop set and pseudo-random pattern can be analyzed to predict future dwells for proactive interception.