AD
A. Daalman
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
1 records found
1
Reliable DOA estimation and detection of closely spaced targets are important for automotive radar, particularly when highly diverse radar cross-sections (RCS) cause weak targets to be masked by nearby strong targets. Independent detection is therefore governed by both angular resolution and strong-target leakage into neighboring angles. High-resolution DOA estimators can improve resolution, but often require additional scene assumptions, sufficient snapshots, or increased computational complexity. Alternatively, Bartlett beamforming offers a simple, assumption-light approach, but its finite-aperture resolution and sidelobes limit target separability and impose higher detection thresholds. Coherent distributed subarrays can improve resolution by synthesizing a larger effective aperture without proportionally increasing the number of antenna elements, while sidelobe leakage can be mitigated through several approaches, including receive antenna-element weighting (apodization).
An optimized multi-apodization framework is developed to suppress sidelobes while preserving the angular resolution of the unweighted aperture. In contrast to conventional multi-apodization, where "standard" weights are combined, the proposed methods optimize complementary weights specifically for their pointwise-minimum response. Two approaches are developed: a convex formulation that prescribes complementary sidelobe-suppression regions, and a joint non-convex formulation that directly optimizes the combined multi-apodized response. For a 20-element uniform linear array, optimized multi-apodization reduces the peak sidelobe ratio from −13.19 dB to as low as −40.00 dB without increasing the 11.40° main-lobe width.
The framework is extended to coherent distributed dense–dense and dense–sparse subarrays, which combine sidelobe suppression with increased angular resolution. Optimization of the sparse-subarray element locations is additionally investigated. Although spatial optimization reduces the intrinsic array-factor sidelobes, it provides no consistent additional improvement when followed by optimized multi-apodization, indicating overlapping capabilities between analog aperture shaping through element placement and digital aperture shaping through complex weighting. Instead, regular dense–sparse arrays provide the most favorable investigated trade-off, allowing the increased angular resolution of a larger effective aperture while retaining sidelobe and grating-lobe structures that can be effectively suppressed by multi-apodization.
Finally, the multi-apodized beamforming is incorporated into a detection pipeline. The Bartlett output is used for candidate localization, after which, for dense–sparse subarrays, a hybrid dense-subarray response suppresses grating-lobe ambiguities. This is followed by a reduced-point Iterative Adaptive Algorithm (IAA) that refines only selected angular regions. The results demonstrate that multi-apodization increases the probability of detection (P_D) of weak targets, especially at small angular separations, by suppressing sidelobes and enabling the use of lower detection thresholds. The subsequent reduced-point IAA spectrum correction substantially reduces the associated false-alarm rate (FAR) by suppressing sidelobe-induced detections, mutual-interference-induced detections, and ambiguity-induced detections. The pipeline allows a coherent two-subarray dense-sparse topology to detect targets with RCS differences of up to approximately 36 dB at approximately 1° angular separability, while maintaining a low FAR after refinement. ...
An optimized multi-apodization framework is developed to suppress sidelobes while preserving the angular resolution of the unweighted aperture. In contrast to conventional multi-apodization, where "standard" weights are combined, the proposed methods optimize complementary weights specifically for their pointwise-minimum response. Two approaches are developed: a convex formulation that prescribes complementary sidelobe-suppression regions, and a joint non-convex formulation that directly optimizes the combined multi-apodized response. For a 20-element uniform linear array, optimized multi-apodization reduces the peak sidelobe ratio from −13.19 dB to as low as −40.00 dB without increasing the 11.40° main-lobe width.
The framework is extended to coherent distributed dense–dense and dense–sparse subarrays, which combine sidelobe suppression with increased angular resolution. Optimization of the sparse-subarray element locations is additionally investigated. Although spatial optimization reduces the intrinsic array-factor sidelobes, it provides no consistent additional improvement when followed by optimized multi-apodization, indicating overlapping capabilities between analog aperture shaping through element placement and digital aperture shaping through complex weighting. Instead, regular dense–sparse arrays provide the most favorable investigated trade-off, allowing the increased angular resolution of a larger effective aperture while retaining sidelobe and grating-lobe structures that can be effectively suppressed by multi-apodization.
Finally, the multi-apodized beamforming is incorporated into a detection pipeline. The Bartlett output is used for candidate localization, after which, for dense–sparse subarrays, a hybrid dense-subarray response suppresses grating-lobe ambiguities. This is followed by a reduced-point Iterative Adaptive Algorithm (IAA) that refines only selected angular regions. The results demonstrate that multi-apodization increases the probability of detection (P_D) of weak targets, especially at small angular separations, by suppressing sidelobes and enabling the use of lower detection thresholds. The subsequent reduced-point IAA spectrum correction substantially reduces the associated false-alarm rate (FAR) by suppressing sidelobe-induced detections, mutual-interference-induced detections, and ambiguity-induced detections. The pipeline allows a coherent two-subarray dense-sparse topology to detect targets with RCS differences of up to approximately 36 dB at approximately 1° angular separability, while maintaining a low FAR after refinement. ...
Reliable DOA estimation and detection of closely spaced targets are important for automotive radar, particularly when highly diverse radar cross-sections (RCS) cause weak targets to be masked by nearby strong targets. Independent detection is therefore governed by both angular resolution and strong-target leakage into neighboring angles. High-resolution DOA estimators can improve resolution, but often require additional scene assumptions, sufficient snapshots, or increased computational complexity. Alternatively, Bartlett beamforming offers a simple, assumption-light approach, but its finite-aperture resolution and sidelobes limit target separability and impose higher detection thresholds. Coherent distributed subarrays can improve resolution by synthesizing a larger effective aperture without proportionally increasing the number of antenna elements, while sidelobe leakage can be mitigated through several approaches, including receive antenna-element weighting (apodization).
An optimized multi-apodization framework is developed to suppress sidelobes while preserving the angular resolution of the unweighted aperture. In contrast to conventional multi-apodization, where "standard" weights are combined, the proposed methods optimize complementary weights specifically for their pointwise-minimum response. Two approaches are developed: a convex formulation that prescribes complementary sidelobe-suppression regions, and a joint non-convex formulation that directly optimizes the combined multi-apodized response. For a 20-element uniform linear array, optimized multi-apodization reduces the peak sidelobe ratio from −13.19 dB to as low as −40.00 dB without increasing the 11.40° main-lobe width.
The framework is extended to coherent distributed dense–dense and dense–sparse subarrays, which combine sidelobe suppression with increased angular resolution. Optimization of the sparse-subarray element locations is additionally investigated. Although spatial optimization reduces the intrinsic array-factor sidelobes, it provides no consistent additional improvement when followed by optimized multi-apodization, indicating overlapping capabilities between analog aperture shaping through element placement and digital aperture shaping through complex weighting. Instead, regular dense–sparse arrays provide the most favorable investigated trade-off, allowing the increased angular resolution of a larger effective aperture while retaining sidelobe and grating-lobe structures that can be effectively suppressed by multi-apodization.
Finally, the multi-apodized beamforming is incorporated into a detection pipeline. The Bartlett output is used for candidate localization, after which, for dense–sparse subarrays, a hybrid dense-subarray response suppresses grating-lobe ambiguities. This is followed by a reduced-point Iterative Adaptive Algorithm (IAA) that refines only selected angular regions. The results demonstrate that multi-apodization increases the probability of detection (P_D) of weak targets, especially at small angular separations, by suppressing sidelobes and enabling the use of lower detection thresholds. The subsequent reduced-point IAA spectrum correction substantially reduces the associated false-alarm rate (FAR) by suppressing sidelobe-induced detections, mutual-interference-induced detections, and ambiguity-induced detections. The pipeline allows a coherent two-subarray dense-sparse topology to detect targets with RCS differences of up to approximately 36 dB at approximately 1° angular separability, while maintaining a low FAR after refinement.
An optimized multi-apodization framework is developed to suppress sidelobes while preserving the angular resolution of the unweighted aperture. In contrast to conventional multi-apodization, where "standard" weights are combined, the proposed methods optimize complementary weights specifically for their pointwise-minimum response. Two approaches are developed: a convex formulation that prescribes complementary sidelobe-suppression regions, and a joint non-convex formulation that directly optimizes the combined multi-apodized response. For a 20-element uniform linear array, optimized multi-apodization reduces the peak sidelobe ratio from −13.19 dB to as low as −40.00 dB without increasing the 11.40° main-lobe width.
The framework is extended to coherent distributed dense–dense and dense–sparse subarrays, which combine sidelobe suppression with increased angular resolution. Optimization of the sparse-subarray element locations is additionally investigated. Although spatial optimization reduces the intrinsic array-factor sidelobes, it provides no consistent additional improvement when followed by optimized multi-apodization, indicating overlapping capabilities between analog aperture shaping through element placement and digital aperture shaping through complex weighting. Instead, regular dense–sparse arrays provide the most favorable investigated trade-off, allowing the increased angular resolution of a larger effective aperture while retaining sidelobe and grating-lobe structures that can be effectively suppressed by multi-apodization.
Finally, the multi-apodized beamforming is incorporated into a detection pipeline. The Bartlett output is used for candidate localization, after which, for dense–sparse subarrays, a hybrid dense-subarray response suppresses grating-lobe ambiguities. This is followed by a reduced-point Iterative Adaptive Algorithm (IAA) that refines only selected angular regions. The results demonstrate that multi-apodization increases the probability of detection (P_D) of weak targets, especially at small angular separations, by suppressing sidelobes and enabling the use of lower detection thresholds. The subsequent reduced-point IAA spectrum correction substantially reduces the associated false-alarm rate (FAR) by suppressing sidelobe-induced detections, mutual-interference-induced detections, and ambiguity-induced detections. The pipeline allows a coherent two-subarray dense-sparse topology to detect targets with RCS differences of up to approximately 36 dB at approximately 1° angular separability, while maintaining a low FAR after refinement.