A Data-Driven Approach to Complex Voxel Predictions in Grayscale Digital Light Processing Additive Manufacturing Using U-Nets and Generative Adversarial Networks.

Data-driven U-net machine learning (ML) models, including the pix2pix conditional generative adversarial network (cGAN), are shown to predict 3D printed voxel geometry in digital light processing (DLP) additive manufacturing. A confocal microscopy-based workflow allows for the high-throughput acquisition of data on thousands of voxel interactions arising from randomly gray-scaled digital photomasks. Validation between prints and predictions shows accurate predictions with sub-pixel scale resolution. The trained cGAN performs virtual DLP experiments such as feature size-dependent cure depth, anti-aliasing, and sub-pixel geometry control. The pix2pix model is also applicable to larger masks than it is trained on. To this end, the model can qualitatively inform layer-scale and voxel-scale print failures in real 3D-printed parts. Overall, machine learning models and the data-driven methodology, exemplified by U-nets and cGANs, show considerable promise for predicting and correcting photomasks to achieve increased precision in DLP additive manufacturing.

Bi-Criteria Radio Spectrum Sharing With Subspace-Based Pareto Tracing.

Radio spectrum is a scarce resource. To meet demands, new wireless technologies must operate in shared spectrum over unlicensed bands (coexist). We consider coexistence of Long-Term Evolution (LTE) License-Assisted Access (LAA) with incumbent Wi-Fi systems. Our scenario consists of multiple LAA and Wi-Fi links sharing an unlicensed band; we aim to simultaneously optimize performance of both coexistence systems. To do this, we present a technique to continuously estimate the Pareto frontier of parameter sets (traces) which approximately maximize all convex combinations of network throughputs over network parameters. We use a dimensionality reduction approach known as active subspaces to determine that this near-optimal parameter set is primarily composed of two physically relevant parameters. A choice of two-dimensional subspace enables visualizations augmenting explainability and the reduced-dimension convex problem results in approximations which dominate random grid search.

Timing Offset and Timing Stability for a Dual-Clock Channel Sounder.

We describe a mathematical framework for evaluating timing offset and timing noise in channel sounders based on a second-order deterministic model and a stochastic metric based on the Allan Deviation. Using this framework, we analyze the timing offset and noise for a 1-6 GHz correlation-based channel sounder that uses rubidium clocks to provide a common timebase between the transmitter and receiver. We study timing behavior in three clock-distribution configurations. In the "untethered" configuration, the transmitter and receiver each have a rubidium clock, and no physical timing cable is connected between the clocks. In the "tethered" configuration, a coaxial cable synchronizes timing between the two separate clocks. Finally, a benchmark "single-clock" configuration is used where a single rubidium clock drives the transmitter and receiver.

Evaluating Uncertainty of Microwave Calibration Models With Regression Residuals.

We present a sensitivity-analysis and a Monte-Carlo algorithm for evaluating the uncertainty of multivariate microwave calibration models with regression residuals. We then use synthetic data to verify the performance of the algorithms and explore their limitations in the presence of correlated errors. The uncertainties we evaluate can be used to estimate the total uncertainty of a calibrated measurement when combined with the prediction intervals for that measurement.

Evaluating Uncertainty of Nonlinear Microwave Calibration Models with Regression Residuals.

We investigate the performance of a recently developed algorithm that evaluates the uncertainty of nonlinear multivariate microwave calibration models using regression residuals. We apply the algorithm to synthetic data consisting of both random and systematic errors and show that the algorithm can account for both types of errors even in the absence of accurate models for the random errors. We also verify the algorithm with measured data.