This PDF file contains the front matter associated with SPIE Proceedings Volume 11192, including the Title Page, Copyright information, Table of Contents, Author and Conference Committee lists.

Dual-wavelength semiconductor lasers have various potential applications in microwave photonics and laser radars fields. Monolithic integrated dual-wavelength DFB laser with equivalent-chirp sampled grating is proposed and investigated theoretically in this paper. The grating of the dual-wavelength DFB laser is designed by the reconstruction-equivalentchirp technique. The effects of the grating with different phase-shifts and different chirp ratios on the characteristics of the dual-wavelength DFB laser are theoretically demonstrated. The results proposed in this paper have great reference value for fabricating dual-wavelength semiconductor lasers.

In conventional optical communication systems, the transmission signals are regular digital signals. Due to the openness of optical fiber networks, it is rather easy for eavesdroppers to obtain the transmission signals from the public fiber-link and then intercept the message directly. To address this crucial security challenge, in this work we propose a high-speed secure optical communication system in virtue of physical random temporal encryption based on private physical random phase modulation and phase-to-intensity conversion. The proposed physical random temporal encryption is performed by a module composed of a phase modulator (PM) and a chirped fiber Bragg grating (CFBG), and the corresponding decryption is achieved with a similar module composed of an inverse-phase driven PM and an oppositedispersion CFBG. By distributing a constant-amplitude random-phase signal to the local semiconductor lasers deployed in the encryption module and decryption module, a pair of synchronized physical random PM driving signals that are not exchanged on public link can be independently generated, which guarantees the receiver end can correctly decrypt the original transmission message. Our numerical results demonstrate that with the proposed encryption scheme, the regular transmission signal is encrypted as a noise-like signal that can greatly enhance the security of message, and moreover, based on the private synchronized physical random phase modulation, the privacy of encryption and decryption are guaranteed, which prevents the eavesdroppers from intercepting transmission message. This work provides a promising strategy for the implement of high-speed high-security physical-layer optical communication.

We propose and fabricate a rapidly wavelength switching DFB laser array based on Reconstructed-Equivalent-Chirp technique. A Semiconductor Optical Amplifier(SOA) is applied to enhance and balance the output optical power. The module covers 8 channels from 1554.5nm to 1566.1nm with an interval of 1.6nm. To tune the wavelength on the microsecond scale, we adopt a combination of a MCU and a FPGA as the controlling core to turn on and off the driving current of 8 lasers on the DFB laser array through a collector feedback circuit, and the switching time between 2 channels is well controlled within 300ns. The side mode suppression ratios(SMSRs) of all channels are above 50dB and the output power are guaranteed above 10dBm with the SOA providing 14dBm saturation output power.

Electroluminescence based on inelastic tunnelling has been investigated for many years in scanning tunnelling microscopy (STM) and plasmonic antenna platforms. Here, we report an electrically excited nanosource with a bowtieshaped tunnel junction that has achieved an output power. The tunnel junction in our experiments is formed via an electromigration (EM) process. The benefits of our new structural design include higher tunnel currents, ultra-strong electrical field enhancement and a higher Purcell factor. By varying the geometric parameters of the plasmonic antenna, the energies of these LSP modes can be tuned for different confinement lengths.

A photonic-based approach for multifunctional microwave signal generation and processing is demonstrated based on equivalent phase modulation. The key component of the system is a dual-polarization quadrature phase-shift keying (DPQPSK) modulator. One dual-parallel Mach-Zehnder modulator (DP-MZM) in the DP-QPSK modulator is biased to function as an equivalent phase modulator (e-PM), while the other DP-MZM is biased as a carrier-suppressed singlesideband (CS-SSB) modulator. The two optical signals from the two DP-MZMs are combined and detected in a photodetector. With different driving signals applied to the two DP-MZMs, different functions can be achieved. When the e-PM is driven by a direct current signal to phase shift the optical carrier, and the CS-SSB modulator is to generate a first-order optical sideband of the driving RF signal, a wideband microwave phase shifter is implemented, which can introduce arbitrary phase shift to the electrical driven signal applied to the CS-SSB modulator. Under the above condition, if the CS-SSB modulator is to generator a first-order and an opposite third-order optical sidebands of the driving RF signal, a repetition rate tunable triangular and square waveform generation scheme can be realized. When the e-PM is driving by an electrical coding signal, and the CS-SSB modulator is to generate a first-order optical sideband of the driving RF signal, a reconfigurable pulse compression signal generator is achieved. Experimental verifications are made to demonstrate the multifunctional system, which has the potential to be used in a variety of microwave systems.

A triple-frequency microwave photonic link is proposed based on a polarization-multiplexing dual-parallel Mach- Zehnder modulator (PM-DPMZM). The lower sub-DPMZM is biased at the maximum transmission point to obtain the 2nd-order RF sidebands. Meanwhile, the RF signal modulates the upper sub-DPMZM through an electrical 90° hybrid coupler. The upper sub-DPMZM works at the carrier-suppressed single-sideband (CS-SSB) modulation to obtain the +1st order RF sideband. An optical band pass filter is used to filter out the +2nd-order RF sideband, and only the +1storder and -2nd-order RF sidebands are output for the frequency beating at a photodetector (PD). The frequency triple signal of 3ω_RF is obtained correspondingly. The experimental link is built. The results show that the RF signal from 8 to 12 GHz is tripled, and frequency tripled signal at 24 to 36 GHz is acquired. The minimal spur suppression ratio of triple-frequency signal is 20dB. A relatively low-frequency signal can be used to generate a high-frequency signal with well quality by this triple-frequency link. The proposed method can be applied in various microwave photonic transmitting systems.

A photonic RF self-interference cancellation (SIC) scheme for full-duplex communication is proposed and demonstrated experimentally. It is based on phase modulation to convert the RF signal into optical domain. The interference cancellation performance of the photonic RF SIC system under different delay deviation (Δτ) and amplitude deviation (Δα) is analyzed. The cancellation depth of 34.5 dB is measured for 10 GHz signal with bandwidth of 50MHz. According to experimental results, the interference cancellation performance affected by the time delay deviation, the amplitude deviation and the phase response is investigated. The results give a direction for the improvement of system performance.

Characterizing chemical and physical aerosol properties is important to understand their sources, effects, and feedback mechanisms in the atmosphere. In this study, a classification scheme based on multidimensional polarization central spectrum (including P-Hdop, P-Pdop, R-Hdop, R-Pdop at scattering angle 30°, 60°, 85°, 115° respectively) were developing to classify aerosol type. The scheme is obtained thanks to the outstanding set of information on aerosol composition and morphology these polarization properties contain. All these polarization data are recorded in real-time and on each individually flowing aerosol particle in our equipment. Several continuous particles signals are averaged to reduce the measurement uncertainty and size dependence, then high temporal resolution and polarization feature resolution can be obtained simultaneously in our method. Several Standard atmospheric aerosols generated in laboratory are identified and their polarization scattering signatures were evaluated. Experiment results were interpreted based on numerical simulations (Mie theory) to investigate polarization properties resolved by particle complex refractive index and shape. Results of this exploratory study proved useful in drawing a general spectrum database build on multidimensional polarization index to optically identify aerosol types in real time.

In this work, we present an in situ online aerosol recognition scheme by synchronized parallel polarization scattering analysis. We develop a measurement system based on multi-angle optical scattering and multidimensional polarization analyzing technique.Using the synchronized polarization information measured by this system, we can obtain the correlation coefficient, which is the representation of the relative relationship between the two polarization indices. We use the twodimensional information represented by correlation coefficient to describe the polarization characteristics of different types of aerosols instead of the average or histogram of one-dimensional polarization index.The number of correlation coefficients among more than three indicators exceeds the number of original indicators themselves, so the original information can be further interpreted and expanded.Since our system can simultaneously detect four normalized polarization indices online, we also synthesize the correlation coefficient groups among these indices to form the correlation matrix.The correlation matrix contains abundant data information, and some special elements or element combinations of the matrix are effective for identifying specific types of aerosol particles.We constructed an evaluation vector to distinguish different types of aerosols represented by different correlation matrices.In addition, the potential physical significance of each element of the vector and the aerosol physical properties it represents are preliminarily studied.The experimental data confirm the application prospect of correlation matrix in the rapid and effective identification of atmospheric pollutants.

In this work we propose a new physical realization of optical neural network (ONN) based on a recently appeared technological platform of synthetic photonic lattices (SPL), and demonstrate its capabilities for deep learning. The system operates with time series of optical pulses with ability to easily control their parameters and possesses the architecture that well suits the ONN paradigm. We have also shown that such an ONN can be potentially utilized for signal processing in optical communication lines for signal distortion compensation.

We propose a vibration detection system based on Phase-sensitive Optical Time Domain Reflectometry (φ-OTDR) and 3D printed sensors. The sensor is composed of a cylindrical elastomer wounded by bend insensitive optical fiber. The cylindrical elastomer is made of flexible material by 3D printing machine. We have demonstrated that the 3D printed sensor is highly sensitive to vibration. Experimental results also show that the sensors in our vibration detection system can distinguish different distances from the vibration source, indicating an accurate locating approach of structural damage in health monitoring for large scale civil engineering structure.

Optofluidic time-stretch imaging system has enabled high-throughput phenotyping of cells with unprecedented high speed and resolution. However, significant amount of raw image data is produced, which requires recognition algorithm with not only high accuracy but also high speed to analyze image data efficiently. In this paper, we compare the performance of popular feature extraction methods and learning-based classification algorithms on time-stretch microscopy image recognition. The applied image recognition system comprises an outlier detection step, feature extraction method and classification. The main concept of outlier detection uses DBSCAN (Density-Based Spatial Clustering of Applications with Noise) to eliminate error images. Gabor wavelet, HOG (Histograms of Oriented Gradients), LBP (Local Binary Pattern) and PCA (Principal Components Analysis) are applied and compared as the feature extraction methods. Finally, with a set of extracted features, the computing time and accuracy of SVM (Support Vector Machines), LR (Logistics Regression), ResNet (Residual Neural Network) and XGBoost (Extreme Gradient Boosting) classification algorithms are evaluated. The tested cell image datasets are acquired from high-throughput imaging of numerous drug-treated and untreated cells (N = ~21,000) with an optofluidic time-stretch microscope. Results show that PCA feature extraction and XGBoost classification proves to be the fastest algorithms with the highest level of accuracy. DBSCAN outlier detection helps to improve the recognition accuracy by 2% approximately. Therefore, we propose a recognition algorithm consisting of DBSCAN outlier detection, PCA feature extraction and XGBoost classification as a promising solution to process the image data of high-throughput optofluidic time-stretch microscopy accurately and rapidly.

Single-pixel imaging system can get image through a detector without spatial resolution. It is a lower-cost alternative to multi-pixel detectors. In addition, the single-pixel detector can provide higher performance, such as higher detection efficiency, lower noise and faster time response. However, single-pixel imaging often needs to be modulated by expensive digital micromirror device (DMD) or spatial light modulator (SLM) to produce structured light. Moreover, DMD or SLM makes the imaging system more complex and brings difficulties to integrated design. Therefore, we propose a method of replacing DMD and SLM with an arbitrary display screen (computer screen, mobile phone screen, liquid crystal display, organic light-emitting diode and so on) to realize structured light illumination. A variety of display screens with low price are widely used in the market. And it is convenient to drive and control, it has a long service life and high stability. By programming the processor, we can control the display screen to generate any desired structured light directly. We also compared the recovery effects of different algorithms. The results show that display illumination can replace DMD or SLM in single-pixel imaging system.

The harmful effects of Cryptosporidium oocysts and Giardia cysts in drinking water have been widely concerned by the international community. Currently, the EPA1623 method is one of the most mature and authoritative methods for detecting Cryptosporidium oocysts and Giardia cysts internationally. However, this method has the limitations of high cost of time and human labor. Based on ultrashort pulse time-space-frequency mapping principle, ultrafast time-encoded flow imaging can reach high speed and high resolution. Therefore, it is proposed for replacing the last three steps of EPA1623, which are immunomagnetic separation, fluorescent staining and enumeration. Specifically, mixed with immunomagnetic beads, the liquid quality sample of Giardia cysts and Cryptosporidium oocysts flow through the microfluidic channel with high throughput of 100 particles/s. With ultrafast time-encoded flow imaging system, images are acquired including oocysts and cysts which are magnetized by attachment of magnetic beads or not, and only magnetic beads. Extracted appearance and shape features, images are classified by K-means cluster algorithm. It is shown in results that, ultrafast time-encoded flow imaging method costs less than 10 minutes and maintains recovery at more than 80%, compared to the last three steps in EPA1623 which need almost 2 hours at less recovery. The proposed method makes full use of the biological properties of immunomagnetic beads, Cryptosporidium oocysts and Giardia cysts, and maintains high percent recovery with much shorter detection time.

Ultrafast imaging flow cytometry can be realized by time-encoded single-pixel imaging technique, with high imaging speed (<10million frame/s) and high throughput (<10,000 cells/s). However, the signal of background image without cells occupies a large part of the acquired data and takes up a lot of storage space. In this paper, a FPGA-based triggering and storage system is proposed, which allows real-time storage of signal of cells with blank background neglected. Moreover, it is easy to implement and of high accuracy, as well as adaptivity to different sampling rate. This system reduces the required storage space and enables efficient storage for ultrafast imaging flow cytometry.

A multi-longitudinal mode (MLM) laser sensor system based on software radio demodulation is proposed, which realizes the simultaneous measurement of vibration and temperature. The software demodulation is realized through the SDR (Software Defined Radio) technique due to the applied vibration can be analyzed as the superposition of a series of frequency modulation (FM) signals. Furthermore, the absorption of erbium-doped fiber (EDF) decreases with the increase of temperature, so the output power of the laser can be measured as another sensing signal. As a result, by demodulating these two sensing signals, the vibration from 10 Hz to 2 kHz can be successfully measured and the sensitivity of the temperature is around 0.011 dBm/°C.

A panoramic monitoring system is designed to achieve continuous monitoring of the surrounding environment. The image acquisition module of the system is composed of five fixed-focal-length cameras and a variable-focal-length camera, which realizes 360 degree environmental monitoring. Usually, the background of continuous photography changes due to fluctuations of ambient light, humidity and wind. Therefore, a dynamic adaptive threshold is used to dynamically update the background template in order to better accommodate various weather changes. Further, a motion-aware algorithm based on background updates is applied to effectively detect whether an intruding target exists and determine the direction of the target. Once an intrusive target is found, the deep convolution neural network Yolo is employed to recognize the target quickly. It shows the advantages of less computation and preferable detection accuracy. In addition, according to the preset warning level, when the intrusion target needs to be alarmed, the target orientation is transmitted to the platform through the central control processing unit, so that the variable-focal-length camera can take real-time snapshots. we propose an end-to-end lightweight siamese convolution neural network to achieve fast and robust target tracking. The network structure replaces the hand-crafted features by the multi-layers deep convolution features of the target, so that higher precision can be achieved. The experiment result shows panoramic surveillance system can effectively and robustly perform security tasks such as panoramic imaging, target recognition and fast target tracking.

A monitoring system for nitrate concentration in seawater based on ultraviolet absorption spectrum is discussed. The monitoring system can realize in-situ real-time monitoring and provide a large amount of monitoring data for Marine environmental analysis, which has fast measurement speed, simple operation, long time continuous monitoring and reagent-free. Savitzky-golay (SG) convolution smoothing pretreatment is used in the calculation, which can remove noise through spectral pretreatment of the acquired absorption spectral data. According to the relationship between the concentration of nitrate and the absorbance in ultraviolet, Partial least square method (PLS) is used in the model established to measure the nitrate concentration. Based on the interval Partial Least Square (iPLS), the ultraviolet absorption spectrum of 219-244 nm is selected in the model to represent the whole band for modeling, which can reduce calculation time and increase model accuracy. The optimal number of principal component was determined to be 3 on the basis of cross-validation Q_h². By laboratory system evaluation, for the artificial seawater with NO₃^--N concentration of 30-750μg/l, the nitrate concentration using PLS model is higher in linear correlation to its actual concentration (R²=0.999) in which the Root Mean Squared Error(RMSE) is 10.27, mean absolute error is 8.02 μg/l, average relative error is 2.4%, indicating that the system has high detection accuracy and good stability, and is suitable for the continuous monitoring of nitrate concentration in low-nutrient seawater.

IMECAS is developing a silicon photonics process platform based on existing 22nm CMOS platform. Developing this platform requires continuous process optimization and design verification, so the wafer-level test solution presented in this paper plays an extremely important role in process validation and optimization. We design a test station which enables manual and semi-automatic for optical and electro-optical testing of passive and active silicon photonics components and circuits, including waveguides, grating couplers, splitter, photo-detectors, modulators etc. It is compatible with 200mm wafer-level testing and Die-level testing. Meanwhile, it has two coupling ways: horizontal coupling and vertical coupling. The measured repeatability of S-parameters and IV is within 6α.

Prevailing object detection algorithms such as RCNN, YOLO and SSD usually are not suitable for high definition surveillance systems because of the fixed size of network input and masses of object candidate regions in the select search process. This paper proposes a fast moving object detection and recognition method for video surveillance system, which applies background extraction and frame difference to fulfill select search process, followed by a pretrained CNN model inference to complete object recognition. Proposed method was proved to be fast and effective in our experimental results which is more suitable for moving object detection for video surveillance system, compared to current other object detection algorithms.

A microwave mixer based on a tunable microwave photonic filter is presented and experimentally demonstrated. The tunable microwave photonic filter consists of a broadband optical source and a dual-parallel Mach–Zehnder modulator employing a variable optical carrier time-shift method. The central frequency of the tunable single-passband filter can be tuned by adjusting an optical variable delay line. The filter can select the wanted mixing components and suppress the other mixing spurs. Experiments are performed and the results show that the out-of-band rejection ratio of the filter is over 45 dB. The up-converted and down-converted signals are successfully selected, and the unwanted mixing spurs are effectively suppressed.

A photonic-based approach for microwave spectrum sensing is proposed based on optical carrier-suppressed single-sideband (CS-SSB) modulation and coherent detection. Two dual-parallel Mach-Zehnder modulators (DP-MZMs) function as two CS-SSB modulators. A local oscillator (LO) signal with its frequency swept at a fixed step of Δf is applied to one DP-MZM, while the microwave signal to be detected is applied to the other DP-MZM. The two CS-SSB modulated optical signals from the two DP-MZMs are sent to a coherent receiver, which consists of a 90° optical hybrid and two balanced photodetectors. The two outputs of the coherent receiver are combined by a 90° electrical hybrid and filtered by an electrical low-pass filter with a bandwidth of Δf. Different frequency components in the unknown microwave signal are all frequency downconverted to within Δf and the signals in different frequency bands appear in different time periods, which can be used for spectrum sensing by sampling and processing the IF signal. In addition, due to the balanced detection, the direct current components are suppressed, and the frequency-swept LO signal only detects the microwave signal on its left or right side, avoiding the interference from the image frequency. To verify the proposed technique, spectrum sensing in a frequency range from 1 to 20 GHz is demonstrated by simulation.

A tunable dual-frequency optoelectronic oscillator (OEO) based on a tunable dual-passband microwave photonic filter (DPMPF) is proposed and demonstrated. The DPMPF is based on phase-to-intensity modulation (PM-IM) conversion and stimulated Brillouin scattering (SBS). Two pump lightwaves are generated through carrier-suppression doublesideband modulation (CS-DSB) in a Mach-Zehnder modulator (MZM) to generate two SBS gain regions in a single mode fiber. Two SBS gain regions act on the phase modulation signal to achieve a dual-passband filter. Through simply varying the frequency of the radio frequency (RF) signal used for CS-DSB, the two central frequencies of the OEO can be tuned with the frequency interval kept constant. In addition, adjusting the frequency of tunable optical source launched to the MZM, the frequency interval of the OEO can be tuned. In the experiment, the dual-frequency OEOs with frequency intervals of 0.2 GHz and 1.2 GHz are achieved, respectively. The frequency tuning range from 3 to 8 GHz is demonstrated. The dual-frequency OEO with the frequency interval tuned is also achieved.