Wangzhe Li

Piezo-photoelastic tuning provides a capacitive, carrier-free route to optical resonance control, making it attractive for data-center interconnects and energy-efficient photonics. However, piezoelectric materials that simultaneously provide lead-free composition, CMOS compatibility, and high piezoelectric coefficients remain elusive. To address this, we first propose and experimentally demonstrate AlScN-integrated Si ₃ N ₄ microring resonators for efficient piezo-photoelastic tuning. The device employs a planar AlScN-on-Si ₃ N ₄ stack designed to enhance stress-optical interaction. Experimental results show a passive loaded quality factor of 4.7×10 ⁴ and tuning efficiencies of 0.9 and 1.0 pm/V at the drop and through ports, respectively. These results demonstrate the potential of AlScN-integrated waveguides for enhanced piezo-photoelastic tuning.

Multiple-input-multiple-output inverse synthetic aperture radar (MIMO-ISAR) is widely used in three-dimensional (3D) imaging due to its excellent angular resolution. After range compression, by applying super-resolution algorithms, i.e., sparse Bayesian learning (SBL), on the Doppler dimension and the height dimension sequentially, the 3D point cloud of the target is obtained. However, this cascaded processing approach leads to error accumulation and may even impair the interference phases between antennas. In contrast, joint multi-channel processing can achieve better performance but incurs heavy computational costs. Hence, the SBL method with space-time adaptive dimensionality reduction (STADR SBL) is proposed for 3D imaging efficiently. Firstly, the space-time domain dictionary is constructed for joint sparse recovery in the angular (height) and Doppler frequency dimensions. Then, edge detection is performed on the angle- Doppler spectrum, and the frequency search range of the dictionary is reduced to within the edge boundaries. Finally, the dimensionality-reduced dictionary is further split into a spatial and a temporal dictionary utilizing the Kronecker product property, and super-resolution 3D imaging is achieved efficiently by the fast SBL method. Experiments with both measured real data and simulated data verify the algorithm's effectiveness.

Advanced deception countermeasures now enable adversaries to inject false targets into synthetic-aperture-radar (SAR) imagery, generating electromagnetic signatures virtually indistinguishable from genuine targets, thus destroying the separability essential for conventional recognition algorithms. To address this problem, we propose a versatile front-end Multi-Feature Extraction and Spatial Transformation Network (MFE-STN), specifically designed for the task of discriminating between true targets and deceptive false targets created by SAR jamming, which can be seamlessly integrated with existing CNN backbones without architecture modification. MFE-STN integrates three complementary operations: (i) wavelet decomposition to extract the overall geometric features and scattering distribution of the target, (ii) a manifold transformation module for non-linear alignment of heterogeneous feature spaces, and (iii) a lightweight deformable spatial transformer that compensates for local geometric distortions introduced by deceptive jamming. By analyzing seven typical parameter-mismatch effects, we construct a simulated dataset containing six representative classes—four known classes and two unseen classes. Experimental results demonstrate that inserting MFE-STN boosts the average F1-score of known targets by 12.19% and significantly improves identification accuracy for unseen targets. This confirms the module’s capability to capture discriminative signatures to distinguish genuine targets from deceptive ones while exhibiting strong cross-domain generalization capabilities.

Multiple-input-multiple-output inverse synthetic aperture radar (MIMO-ISAR), which features a large real aperture, exhibits excellent angular resolution. However, the coherence of the MIMO array is severely impaired by various errors (i.e., amplitude-phase errors, position errors, and time synchronization errors). To tackle this issue, a multi-error calibration method based on joint space-time phase gradient autofocus (JST-PGA) is proposed. First, the estimation of the three types of errors are simplified into the correction of the channel amplitude-phase errors. Subsequently, the channel amplitude errors (CAEs) are estimated by comparing the envelopes of high signal-clutter-ratio (SCR) samples. Finally, with the space-time (channel-pulse) equivalence, the spatial PGA (SPGA) method is utilized to rectify the channel phase errors (CPEs). To mitigate the error accumulation caused by signal cascaded processing, the JST-PGA method is further developed. Specifically, angle- Doppler domain autofocusing is performed on the space-time snapshot samples to jointly correct the translational phase and CPEs after compensating the quadratic phase coupling terms. The proposed method innovatively uses the two-dimensional PGA to restore the coherence of the MIMO array, which avoids the requirement for external cooperative scatterers or internal calibration circuits. Real measured and simulated data validate the effectiveness and advantages of this method.

Autofocus serves to mitigate the residual motion error (RME) in the airborne synthetic aperture radar (SAR). However, the 2D frequency dependent RME remains unresolved for ultra-high-resolution (UHR) imaging up to the centimeter-scale resolution. To address this issue, a modified 2D autofocus approach is proposed in this paper. Regarding the range frequency dependent error (RFDE), an efficient azimuth resampling strategy combined with the Chirp-z transform (CZT) is devised, which avoids the approximation imposed by the existent subband division method. With respect to the azimuth frequency dependent error (AFDE), it is first reported in this paper, and removed efficiently by a special filter on the support region. The simulation results show that the deviation of the azimuth phase error (APE) can be reduced from ±5 rads to ±0.5 rads after the RFDE removal and ±0.05 rads after the AFDE removal. The real data from an airborne UHR microwave photonic (MWP) SAR is utilized for validation, and four existent autofocus methods are reproduced for comparison. In the cases without sensor-based motion compensation (MoCo), only the proposed method recovers the distinct textures of the manhole cover, demonstrating its robustness. In the cases with sensor-based MoCo, the proposed method achieves the best image quality with almost minimal processing time, demonstrating its efficiency.

Silicon nitride photonic integrated platforms, characterized by ultra-low propagation loss, broad transparency window, high power-handling capacity, and full CMOS compatibility, present a highly promising solution for high-performance optical computing networks. In this study, an on-chip silicon nitride photonic convolution framework is proposed based on fine-grained diffractive slabs with reduced error and wavelength insensitivity. The diffractive slab employs Fresnel approximation to map conventional optical lenses onto a two-dimensional plane. Through meticulous optimization of the structural parameters and application of wavelength-division multiplexing techniques, the slab effectively mitigates the scaling limitations associated with the number of diffractive ports. Compared with previous diffractive-based convolution accelerators, the proposed network demonstrates superior computational error, achieving a root-mean-square error of 0.021 for Fourier-based convolution operations. Notably, small error is achieved using only half the number of modulators and drivers, offering significant potential for accelerating complex processing tasks and high-performance computing applications.