Radar Publications Weekly Briefing (2–9 November 2026)

Radar weekly briefing infographic

Polarimetry

  • Polarimetric Scattering-Structure Contrastive Learning for Few-Shot HRRP Recognition
    Polarimetric scattering-structure contrastive learning transfers physical scattering mechanisms into HRRP representations to improve few-shot recognition under aspect variation.
    https://arxiv.org/abs/2611.04490

FMCW Radars

  • Prediction-Induced Sparsity for Tracked FMCW Radar: A Track-Conditioned Event-Gap FPGA Architecture
    Prediction-induced sparsity conditions FMCW processing on track predictions to reduce computational load and power on FPGA hardware.
    https://arxiv.org/abs/2610.41145

  • AFDM-ISAC Design Amalgamating FMCW Radar and Carrier Aggregation
    Carrier-aggregation-assisted AFDM waveforms with an FMCW-inspired dechirping receiver improve delay-Doppler sensing accuracy and resolution in ISAC systems.
    https://arxiv.org/abs/2610.40107

Target Detection

  • 6G ISAC Drone Detection: Signal Processing and Experimental Validation at FR1
    Coherent and non-coherent ISAC detectors using clutter whitening and adaptive coherence estimation enable real-time drone tracking on a commercial base-station platform.
    https://arxiv.org/abs/2611.05052

  • Data-Driven Likelihood-Ratio Construction for Radar Detection via Bernstein Induction Operators
    Data-driven likelihood ratios from clutter samples using Bernstein polynomials and induction operators improve weak-target detection in non-Gaussian sea clutter.
    https://arxiv.org/abs/2611.06192

  • Lightweight graph transformers for clutter and target classification in automotive radar
    Tiny graph transformer networks classify clutter versus stationary and moving targets with far smaller model size and faster inference.
    https://arxiv.org/abs/2611.04959

  • Joint Data Association and Tracking for Surveillance Radar Using Gaussian Processes
    Variational Bayesian joint data-association and tracking based on Gaussian processes outperforms Bernoulli and Kalman filters for weak drone targets.
    https://arxiv.org/abs/2611.03314

Other Topics

  • Payload-Blind Bistatic Delay-Doppler Sensing with Phase-Injected Chirps in CE-OFDM
    Injecting Zadoff-Chu chirps into CE-OFDM phase enables payload-blind bistatic sensing via residual carriers acting as implicit pilots.
    https://arxiv.org/abs/2611.09687

  • Reliable Bistatic Ranging With Unreliable Communication Payloads
    Reliable bistatic ranging in ISAC by exploiting unrecoverable high-rate data payloads yields higher detection probabilities than pilot-only methods at low SNR.
    https://arxiv.org/abs/2611.07803

  • 4D Point Splatting for mmWave Radar Micro-Doppler Differentiable Rendering
    Physics-based 4D point-splatting differentiable renderer fits explicit moving scenes to real mmWave radar captures for high-fidelity micro-Doppler simulation.
    https://arxiv.org/abs/2611.08342

  • A Universal Raw-IQ Radar Encoder for Cross-Sensor Human Sensing
    Convolutional beta-VAE pretrained on raw IQ radar data from multiple platforms produces transferable embeddings for human sensing across unseen sensors and users.
    https://arxiv.org/abs/2611.04534

  • P-SER: Physical Consistency-Constrained Scattering Expandable Representation for Class-Incremental HRRP Target Recognition
    Reconstructs scattering points from sum and difference HRRP channels to enforce physical consistency in an expandable representation for class-incremental recognition.
    https://arxiv.org/abs/2611.04254

  • CLOCs-R: Radar-Assisted Camera-LiDAR Fusion for Long-Range Object Detection
    Adds accumulated radar returns to rescoring camera-LiDAR detections, providing modest mid-range gains on automotive datasets.
    https://arxiv.org/abs/2611.04300

SAR Papers

  • High-order adaptive context association and domain-aware convolution for oriented ship detection in synthetic aperture radar images
    YOLO-SARX with hybrid polarimetric convolutions, orientation-aware pooling, and high-order adaptive context modules improves oriented ship detection under speckle and clutter.
    https://doi.org/10.1016/j.asoc.2026.116025

  • A fast imaging method for moving vehicles in traffic radar based on second-order Keystone transform
    Sub-aperture second-order Keystone processing enables real-time ISAR-style imaging of moving vehicles with millimeter-wave traffic surveillance radars.
    https://doi.org/10.1016/j.dsp.2026.106395

Papers are ordered by priority: polarimetry, FMCW radars, target detection, then other topics (SAR last). Sources include arXiv, IEEE, Elsevier, and MDPI.

The End of an Era

The End of an Era

The End of an Era - our lovely target for the validation of most ideas and signal processing algorithms during last 18 years is demolishing…

Our the first and the best target

This is the slide from my presentation in 2009 - the illustration of our unexpected finding that the PARSAX radar can provide two times better range resolution that was planned during the design time.

80 years ago we received our own signals from outer space for the first time

On January 10, 1946, a team of military and civilian personnel of the Project Diana at Camp Evans, Fort Monmouth, New Jersey, aimed a modified SCR-271 bedspring radar system at the rising Moon and transmitted high-power radio pulses in its direction. About 2.5 seconds later, the range-amplitude oscilloscope’s indicator for the first time in human history registered a faint radio echo that travelled to the Moon through the Earth’s atmosphere, outer space, reflected from the Moon’s surface, and returned to Earth.

First radar return from the moon – Project Diana. Credit - Wikimedia Commons and WorldRadioHistory.Com

The radar was operating at the frequency 111.5 MHz, transmitted 15 kW pulses with 250 microsecond duration at a 1/3 Hz pulse repetition rate (one pulse per 3 seconds). It uses an enormous (40x40-foot, 12.2x12.2 meters) double bedspring antenna consisting of an 8x8 array of 64 half-wavelength dipoles with reflectors. Such an array has an antenna pattern with 15 degrees main beam and provides a still not too high gain of about 23 dB (200 times). In available specification (unfortunately, engineering specifications were destroyed by the Army in 1971, so the exact design details are unknown), the maximum Doppler shift of echo was mentioned as 327 Hz with a comment that it was set in the radar for a target at maximum range 980 MPH = 1577,16 km/h = 438,1 m/s (it has the exact Doppler frequency shift 326 Hz at 111.5 MHz frequency)…

More details about the project Diana can be found in the following sources:

An article "Radar Reaches the Moon". Credit - WorldRadioHistory.Com