Narrow Results

An efficient vapor-redissolution technique is used to greatly reduce sidewall scattering loss in the polymer arrayed waveguide grating (AWG) fabricated on a silicon substrate. Smoother sidewalls are achieved and verified by scanning electron microscopy. Reduction of sidewall scattering loss is further measured for the loss measurement of both straight waveguides and AWG devices. The sidewall loss in straight polymer waveguide is decreased by 2.1 dB/cm, the insertion loss of our AWG device is reduced by about 5.5 dB for the central channel and 6.7 dB for the edge channels, the crosstalk is reduced by 2.5 dB, and 3-dB bandwidth is narrowed by 0.05 nm after the vapor-redissoluton treatment.

Aiming at the low efficiency of antenna modeling and solving the problems of slow training speed and high resource consumption when multiple antenna structural parameters are optimized simultaneously, a new deep multi-layer convolutional neural network (DMCNN) is proposed. The DMCNN model uses four layers of ladder-type fully connected layers combined with convolutional layers to improve the learning and classification capabilities of data features. It adds a LeakyReLU activation function to solve the problem of neurons disappearing in negative value areas and alternately uses maximum pooling and average pooling operations to update training parameters to reduce the amount of calculation and maintain feature invariance. Apply Adam optimizer combined with dropout technology to adjust the learning rate and reduce weight fluctuations. The DMCNN model extracts samples from the geometric construction parameters of the ultra-wideband multiple-input multiple-output (UWB MIMO) antenna as feature input to predict the return loss $S_{11}$ and insertion loss $S_{21}$. Simulation results prove that the DMCNN model’s prediction fit for $S_{11}$ and $S_{21}$ reached 98.48% and 95.02%, respectively. Compared with the multi-layer perceptron (MLP) and Fully Connected Neural Network (FCNN) models, the training effect was improved by at least 11.66%. It solves the shortcomings of traditional methods, and has excellent UWB MIMO antenna modeling capabilities.

Push-pull electro-optic polymeric Mach-Zehnder modulators with photo-bleaching induced waveguides and dual-driving electrodes operating at 1.55 μm wavelength have been demonstrated. The nonlinear electro-optic polymer used in this work was based on a phenyltetraene bridged chromophore in polycarbonate, APC-CLD1. It has been shown that the predominant photo-bleaching mechanism was a photochemical decomposition of the chromophore. The half-wave voltage of the integrated polymeric modulator was 4.5 V in a push-pull configuration with a 1.5 cm interaction length. The extinction ratio was greater than 20 dB, and the fiber-to-fiber insertion loss was 8 dB for the TM polarization at 1.55 μm wavelength.

In this paper, we report the insertion loss and adjacent crosstalk for tapered and conventional configuration of double S-shaped design of arrayed waveguide grating (AWG) using 1.2 μm and 1.0 μm core width. The 13-channel AWG on silicon-on-insulator (SOI) was simulated using beam propagation method (BPM) at central wavelength of 1.49 μm, producing a transmission spectrum ranging from 1310–1620 nm. Tapered AWG for both core widths gave the lowest insertion loss of < 7 dB while only the AWG with the 1.2 μm core width produced the lowest crosstalk at < -19 dB compared to the other configurations. A 13-channel transmission spectrum of the AWG device has been produced which fits closely the standard ITU-T CWDM wavelength grid.