This PDF file contains the front matter associated with SPIE Proceedings Volume 12311, including the Title Page, Copyright information, Table of Contents, and Conference Committee Page.

Microwave photonic systems process the electrical signal in optical domain so that the characteristics of the output optical signals include the transfer function of the signal processing link. In this paper, we review our recent works about a microwave photonic repeater, self-interference cancellation, and microwave signal coupling techniques. First, based on a photonic repeater consisting of a dual-polarized dual-parallel Mach-Zehnder modulator (DP-DPMZM) and a Polarization Controller (PC), four different down-conversion states can be performed. Two separate Local Oscillator signals (LO) also enable frequency down-conversion of microwave signals in different frequency bands. Secondly, a Dual-Parallel Dual-Drive Mach-Zehnder Modulator (DP-DDMZM) and a low-frequency LO signal source are used to achieve Self-Interference Cancellation and Harmonic Down-Conversion (SIC+HDC) for an In-Band Full-Duplex (IBFD) system. Finally, for the microwave photonic coupling technique, a Photonic Microwave Hybrid Splitter (PMHS) and a Photonic Microwave Hybrid Combiner (PMHC) are proposed, respectively. The PMHS and PMHC individually controls the amplitude and phase of the two signals by adjusting the Direct Current (DC) biases of modulators. Our proposed signal processing techniques are very promising for flexible microwave signal processing, radars, and wireless communications.

The nonlinear effects in single-mode fiber and the constraints in device technology, time division multiplexing and wavelength division multiplexing cannot increase the capacity of optical fiber communication system unlimitedly. The development of space division multiplexing technology has increased the capacity of existing optical fiber communication systems by at least an order of magnitude. With the continuous improvement of transmission rate and system bandwidth in the long-haul communication, low cost and small size are also regarded as important factors in the future. In this work, the transmission performance of the low-cost self-coherent receiver and the mature coherent receiver in the long-haul mode division multiplexing system is compared. In the 32-Gbaud 6-mode dual-polarization QPSK transmission system with a fiber length of 80 km as the single span, two receiver schemes are compared considering the same configurations of the transmitter and the optical fiber link components. Compared the use of eight photodetectors integrated in the coherent receiver, the Kramers–Kronig (KK) receiver only requires two photodetectors to demodulate the dual-polarization transmission, while the phase recovery algorithm based on Hilbert transform in the KK receiver will increase the complexity of digital signal processing. Numerical results indicate that the KK receiver scheme has the advantages of lower cost and more compact size and shows the similar performance as the coherent receiver for the transmission of less than 2500 km, despite the requirement of larger transmitted power and algorithm complexity. It can also be concluded that, self-coherent receiver based on the KK algorithm can be a complementary detection solution to the coherent receiver for next-generation long-haul transmission networks with low-cost transceivers.

Self-mixing interferometry (SMI) is superior to other laser interferometry methods due to its simplicity and compactness. However, SMI signals are often complex and difficult to process due to interference in the form of variations in the effective reflectivity of the target, noisy signals, complex signal shapes, and other dependencies. Deep neural networks have been a very popular area of research in computer artificial intelligence in recent years, allowing more implicit features to be uncovered than traditional shallow machine learning. It has been shown that convolutional and back-propagation neural networks can be used for SMI signal processing. There are also studies that have used machine learning genetic algorithms for absolute distance measurement. Based on the above, this study used convolutional neural networks to form a deep neural network for absolute distance measurement based on SMI technology. We first trained the deep convolutional neural network at different feedback strengths. The results of the Convolutional Neural Network (CNN) model showed a coefficient of determination of 0.9987. which is consistent with the required model. The trained network was then used to estimate absolute distances with and without the addition of noise. The comparison proves that the proposed method is noise-proof and has high adaptability for measurements under different conditions.

An approach to measuring high-frequency responses of electro-absorption modulated lasers (EMLs) is proposed based on fixed low-frequency pilot analysis. The fixed low-frequency pilot is inserted into the microwave driving signal loaded on the EML through amplitude modulation. Then, the high-frequency response of EML can be obtained by extracting and analyzing the pilot (kHz level) after photodetection, thereby realizing the low-frequency detection for EML measurement. Moreover, the method is independent of the responsivity fluctuation of the photodetector due to the fixed frequency analysis and enables the self-calibrated frequency response measurement of high-speed EML.

An improved self-reference photonic sampling method is proposed to measure the frequency response of photodiode (PD) chips. In the proposed scheme, the uneven response of the Mode-Locked Laser Source (MLLS) is eliminated by using the half-frequency photonic sampling measurements. The microwave de-embedding under short-open-load-device termination is used to realize on-chip de-embedding of the adapter network connected to the receiver of the microwave network analyzer in terms of the transmission loss and the impedance mismatch. The proposed on-chip measurement method is free of any extra electro-optical transducer standard, and an accurate measurement can still be realized without an impedance match.

Self-Mixing Interferometry (SMI) is a promising interferometric measurement technology with unique system structure. It has an advantage that conventional two-beam interferometry do not have, i.e., movement directions of the target to be measured can be determined by a single-channel interferometric signal due to the existence of linewidth enhancement factor in lasers. However, movement directions are difficult to be determined when the optical feedback is weak. In this work, an algorithm is proposed for determination of SMI-based displacement directions based on Convolution Neural Network (CNN). We used Python language and the third-party libraries NumPy to complete numerical calculation as well as TensorFlow to establish the CNN. The simulation results shows that displacement directions are able to be determined with the accuracy higher than 94.8% when the optical feedback factor is low to 0.1.

During the process of annealing, the warpage of conduction cooled diode laser bars (CSs) packaged with indium solder is studied according to temperature difference along the direction of thickness of CSs. This investigation reveals that the smile of CSs is related to the annealing rate and the thickness of the packaging fixture. In theory, the smile can be eliminated by optimizing the annealing rate and the thickness of the packaging fixture when using two-sided isokinetic annealing method. Compared with the one side annealing, when the thickness of the fixture is more than 15 mm, the smile of CSs still can be well controlled by the two-sided isokinetic annealing. It provides theoretic evidences for eliminating smile during annealing and improving the performances of CSs.

Photonic crystal laser diodes are characterized by low divergence angle and high brightness, but thermal effects have become a major obstacle to further improvement of output power and efficiency. The thermal characteristics of high- power photonic crystal laser diodes are of great importance to improve the output power and increase the lifetime. In this paper, the physical heat dissipation model of a single photonic crystal laser diode with CS-mount package is established. Steady-state thermal characteristics simulations are performed using the Finite Element Method (FEM) and the influences of different parameters, such as solder, transition heat sink and heat sink on the thermal characteristics are analyzed. The simulation results show that the thickness and thermal conductivity of the heat sink materials are the main factors impacting the heat dissipation of the laser. The thermal resistance of the laser can be reduced effectively by using heat sink materials of high thermal conductivity. On the premise of ensuring wettability and reliability, the thickness of the solder layer should be decreased. A photonic crystal laser diode with a cavity length of 4 mm and a stripe width of 350µm based on an optimized heat dissipation structure is designed and fabricated. The CW output power of 41.9W, the vertical divergence angle of 18.48° and the thermal resistance of 1.54 K/W are obtained under the injection current of 50A at 20 ℃.

The lateral leakage current is influenced by the height of ridge waveguide. We design two structures to restrict the lateral leakage current of ridge diode lasers, called ‘step-ridge’ structure and ‘groove-ridge’ structure. In order to obtain a better output electrical characteristic, we optimize the geometry of the two structures. For ‘step-ridge’ structure, we simulate various step-widths (including up-step widths and down-step widths). For ‘groove-ridge’ structure, we simulate different widths of groove and distances between injection section and groove. The lateral leakage currents of both two structures were calculated under the same injection current. In conclusion, both two structures can effectively reduce the leakage carrier by at most 80%, the ‘step-ridge’ diode lasers can improve around 6.5% wall-plug efficiency, but the ‘groove-ridge’ diode lasers would reduce the wall-plug efficiency at the same time.

Photonic crystal laser diode bars have the advantages of low vertical divergence angle and high resistance to catastrophic optical mirror damage. However, with the increase of output power, the waste heat problem is becoming more serious, affecting the further improvement of laser performance. Therefore, it is of great significance to study the thermal characteristics of bars. In this paper, the fluid-solid coupling conjugate heat transfer model of a microchannel cooled photonic crystal laser diode bar is established through the Finite Element Method (FEM) and Computational Fluid Dynamics (CFD) numerical methods. The transient thermal behavior, steady-state characteristics, and temperature distribution of photonic crystal laser diode bars under continuous (CW) operating states are studied in detail. The simulation results show that the junction temperature is 55.48°C, and the thermal resistance is 0.48 K/W. The closer the emitter is to the bar center, the easier the thermal crosstalk occurs. In the experiment, the continuous output power of the photonic crystal laser bar is 112.13 W at 120 A, the junction temperature is 57.14 °C, and the thermal resistance is 0.50 K/W. The simulations of bars are consistent with the experiment.

The smile effect of the laser bar increases the difficulty of laser beam coupling and limits its application. In this paper, the smile effect in the high-power laser on the microchannel cooler (MCC) and the method of reducing the smile effect by balancing the thermal-induced stress are studied. The model of diode laser packaging with MCC is established based on the finite-element method (FEM). The effect of the thickness of the N-foil on the smile effect and stress is analyzed. The bar is bent into a convex shape after the bar is bonded to the heat sink, according to the simulation. With the increase of the thickness of the N-foil, the bar deformation gradually decreases, and then the middle part of the bar reversely increases. The thermal-induced stress on the bar is balanced by optimizing the thickness of the N-foil. The minimum deformation was less than 0.2 μm.

We experimentally demonstrate that InAs/GaAs quantum dot (QD) lasers exhibit high reflection insensitivity in the wide temperature range of 293 K to 353 K, which is due to the low and thermal stable linewidth enhancement factor. This work shows the potential of QD lasers as uncooled and isolator-free on-chip laser sources for next-generation photonic integrated circuits.

We design a 976nm fiber coupling module with 20 single-emitter diode lasers by ray tracing. Each emitter has an output power of 10 W. This is achieved by beam collimation of the fast and slow axis, fast-axis beam stacking using overlapping mirrors, and polarization beam combining. Through polarization multiplexing, the output power can be nearly doubled with no loss of beam quality. The core diameter of output fiber is 105μm with a numerical aperture (NA) of 0.22. Finally, the simulated result indicates that the module can have an output power over 190W. At the same time, the brightness of 14.43 MW∙cm-2∙str-1 and the coupling efficiency of 95% can be achieved.

Laser range finder is an important device in the field of deep space exploration and earth mapping. Laser is widely used in Laser range finder as an active detection source. After years of development ,the laser index is getting higher and higher, ranging accuracy limits the narrow pulse width output. Detection dynamic range determines the large energy output of laser. The power supply and thermal control ability determined the high conversion efficiency>5%. Therefore the laser has the characteristics of high repetition rate, larger energy , narrow pulse width, high conversion efficiency output. This project has realized the key technologies. The principle verification and engineering prototype of the 10kHz laser have been completed and it has passed the thermal vacuum ,mechanical and thermal environmental tests.

Distributed feedback (DFB) lasers that are widely used in high-speed communication nowadays usually use end-facet coating in order to seek a balance between fabrication difficulty and device performance. This paper investigates DFB lasers with a partial grating structure that is immune to the phase uncertainties at the end-facet with high-reflection (HR) film and the improvement of its output chirp characteristics. It is found that, after parameter optimization, the carrier concentration and photon density distribution along the partial grating DFB(PG-DFB) laser cavity can be more uniform, which helps to obtain a better production yield. In addition, a compromised choice of the length of grating region and non-grating region can ensure stable single-mode output, as well as sub-10 mA threshold and high slope efficiency (SE). We also find that the PG-DFB laser can reduce the transient chirp by about 5 GHz compared to conventional DFB laser with uniform grating.

We have experimentally investigated the generation of tunable and broadband Optical Frequency Comb (OFC) based on a gain-switching vertical-cavity surface emitting laser driven by a square wave signal under optical injection. During the experiment, the influences of modulation frequency f_m and injection light wavelength λ_i on OFC performances are analyzed systematically. The experimental results indicate that under suitable operation parameters, an OFC with bandwidth of 77.0 GHz within 10 dB power variation and single sideband phase noise of -115.7 dBc/Hz @ 10 kHz can be generated. Also, during the modulation frequency range of 1.5 GHz - 2.8 GHz, broadband OFCs with bandwidth exceeding 70.0 GHz can be obtained based on the square-current modulated VCSEL under optical injection.

Semiconductor laser (SL) with optical feedback presents rich nonlinear dynamics, e.g., Period-One Oscillation (POO), quasi-periodic oscillation, and chaotic states. It is of significance to study the state boundary between different dynamic states for an SL with optical feedback from the perspective of both suppressing and using these dynamics, especially the boundary of POO state. This paper reveals the POO boundary of an SL with Dual Optical Feedback (DOF). The modified Lang-Kobayashi equations were first numerically solved to develop a bifurcation diagram to determine the boundary between POO and other states. Then the POO-DOF boundary was compared with that of an SL with single optical feedback. Moreover, the effect of system parameters on the POO-DOF boundaries is researched. The results obtained are helpful to promote the development of the applications of POO dynamics.

We have performed a theoretical study and made an experimental realization of a multi-frequency self-injection locking of an external cavity laser, composed of a gain-chip and an external mirror, to a high-Q-chip-scaled ring microresonator. We use a numerical model based on the rate equation system that accounts for spontaneous emission to describe a semiconductor laser and optical feedback from a high-Q cavity. It allowed us to investigate dynamics of the system and to find out regimes when several locked lines are emerged simultaneously via power redistribution over the frequency domain. Due to the well-known phenomenon of mode competition, the multifrequency regimes appear only when several conditions are met, otherwise only single-frequency locking regimes may occur. The performed experimental investigations have shown that it is possible to achieve such states deterministically if these conditions are fulfilled.

The application of high-performance VCSELs is extending from consumer electronics to automotive applications. Wet oxidation is an important technology in the fabrication of VCSELs. In this paper, we studied the wet oxidation process and mechanism in order to accurately control the oxidation aperture and improve the power and the conversion efficiency. Current density distributions of VCSELs with different oxide apertures are simulated based on COMSOL Multiphysics. In the experiment, the output power, conversion efficiency and threshold current of single junction and five-junction 940 nm VCSELs varying with oxide apertures are measured. Five-junction VCSELs exhibit maximum power conversion efficiencies are more than 60% and slope efficiency are more than 5.28W/A with oxide aperture from 9 to 15 μm under room temperature pulse condition (50 µs pulse width, 0.5% duty cycle). In addition, 385-element five-junction VCSEL array exhibits maximum power conversion efficiency of 53.45%. The five-junction VCSELs can be used as the basic laser source for the automotive applications.

Among several sensors, photonic RADAR has shown potential performance in range detection, radial velocity, “allweather” system, and distance resolution. With an increase in the applications of the photonics RADAR, disturbances such as interference, intrusion, and jamming due to multiple radars operating simultaneously in “close-proximity”, misdirecting, and misleading in a multi-radar environment can be substantial unless a suitable mitigation technique is employed. Therefore, in a multi-radar environment, analyzing the impact of disturbances on the radar system is of utmost importance. Hence, in this paper, we mainly focus on (i) analyzing disturbances and their effect on the output performance of the widely used LFM radar waveforms, (ii) generating a robust waveform, photonics-based frequency hopping LFM (Ph-FHLFM), and (iii) detecting target objects in the presence of disturbances. In analyzing the effect of disturbances, we focused on the interference with different radar waveforms and spoofing in the Victim’s radar performance, followed by the generation of the Ph-FHLFM. The basic principle involved in the demonstration of the PH-FHLFM is the usage of the abrupt intensity modulation of the optical beam for the abrupt redshift in the emission wavelength of the semiconductor. Four-step Ph-RHLFM and its reconfigurability in hopping steps and center frequency are demonstrated. Further, mitigation of interferences by Ph-FHLFM through the detection of target objects in the presence of disturbances and the comparison with conventional LFM is demonstrated. Hence, the proposed technique can be used for the multi-radar environment, such as autonomous vehicles, navigation, and defense system for safety and security.

All-optical switches have the advantage of significantly reducing the cost of data-center and improving the transmission characteristics of the system, which has led to many different optical switching technologies for data-center. For this application, we demonstrate a superfast wavelength switching drive design for DFB laser array, and realize a fast tunable laser with 5 ns switching latency. The laser array (C-band, 16-channel, 100 G-space) we used is based on the reconstructed equivalent chirp technology. During the process of tuning, the output wavelength of each channel is within the channel wavelength error range specified by DWDM under the ITU-T standard. Based on the above light source, we build a complete prototype of all-optical switching transceiver integrating transceiver and transmitter, and demonstrate the stable transmission of 10.24 Gb/s data under the condition of four wavelength arbitrary switching and routing.

High power mid-infrared GaSb-based lasers are desired for many applications, however, the self-heating in the active region is still one of the main influence factors for practical application. In this paper, we report on fabrication and characterization of high-power GaSb-based lasers. The temperature dependence of output performance of the device was investigated. Due to the high quality of epitaxy and wide waveguide design, the lasers exhibited a high-power capability from 288 K to 318 K. Devices with a cavity length of 1.5 mm and an aperture of 100 μm delivered a power of 1.46 W at a current of 7 A at 288 K and remains 1.10 W at 318 K under CW operation limited by thermal rollover. The characteristic temperature T0 is 151 K and 68 K below and above 298 K, respectively.

We experimentally investigated multi-channel chaos synchronization characteristics based on two asymmetrical mutually coupled Weak-Resonant-Cavity Fabry-Perot Laser Diodes (WRC-FPLDs). Experimental results show that, through adjusting the center wavelength of the Tunable Optical Filter (TOF) and the injection power, different modes can be selected and induced into chaotic state with wideband. Under proper asymmetrical injection power and frequency detuning, stable leader-laggard chaos synchronization with the maximal correlation coefficient about 0.90 between two asymmetrical mutually coupled WRC-FPLDs can be achieved. In addition, the effects of injection power and frequency detuning between the two lasers on chaos synchronization performance have also been discussed.

The integrated Vertical-Cavity Surface-Emitting Lasers (VCSELs) modules have been widely researched and manufactured accompanying with the rapid development of compact atomic magnetometers, atomic gyroscopes, atomic clocks, and the other atomic sensors. For atomic magnetometers operating in the Spin-Exchange Relaxation-Free (SERF) regime, the vapor cell should be heated to a high temperature, which may cause the built-in laser chip over-heated and module structural or optical component deformation, lowering the performance of the built-in laser module. Meanwhile, due to the space constraints, the laser module needs to achieve a large collimation beam diameter and the non-magnetic structure should be optimized to have high temperature tolerance and stable thermal dissipation. In this study, a compact non-magnetic VCSEL module is developed based on the non-magnetic structure with the abilities of optical path alignment, beam collimation, and polarization conversion. Compared with the common TO-can packaging, the proposed VCSEL module achieved low residual magnetic field generated. And the entire volume is less than 1 cm3 with the collimating beam diameter of 2 mm. The experiment evaluation result shows that the laser module could work stably in high temperature with stable thermal dissipation and sufficient thermal margin (60±10℃) for precise wavelength tuning and maintain optical performance and structural for meeting the demand of pump laser in the SERF atomic magnetometers.

GaSb-based narrow Ridge Waveguide (RW) laser diodes providing high optical power with low lateral beam divergence single-transverse-mode operation are fabricated and characterized. The typical Separate-Confinement-Heterostructure (SCH) Multi-Quantum-Well (MQW) structure is grown by the solid-state Molecular Beam Epitaxy (MBE). The 1 mm long uncoated RW lasers yield single-transverse-mode output power exceeding 170 mW in the 1950 nm wavelength range under continuous-wave (cw) operation at an injection current of 800 mA and room temperature of 20 ℃. The shallow-etched 7 μm width RW design produces a lateral beam divergence angle as narrow as 9° Full Width at Half Maximum (FWHM) with an excellent beam quality of M² factor < 2 at the maximum output power, enabling it for simple and inexpensive bulk coupling into the typical SM1950 or PM1950 fiber which has a core diameter of 7 μm and numerical aperture (NA) of 0.2. The RW lasers with high output power, good beam quality, and low divergence are promising candidates for a wide range of demanding and advanced applications including pumping fiber amplifiers and solid-state lasers, seeding external cavity lasers, and frequency conversion.

With the increase of output power, more heat generation and higher operation current have become important issues, which affect the electrical-optical performance and reliability of high power semiconductor lasers. For the past several years, high power semiconductor laser chips utilizing double or triple quantum wells have been developed to achieve higher output power. However, the operation current of diode laser chips with double or triple quantum wells is much higher than that with single quantum well. Diode laser chips with double or triple quantum wells could only operate at a much lower duty cycle. In this paper, a compact quasi-continuous wave (QCW) high power semiconductor laser array based on dual-chip integration techniques has been developed. For this packaging structure, two diode laser bars were welded above and below a micro-channel heat sink, without significant increase in volume. By means of this integration method, the output power of the semiconductor laser could reach kilowatt-level at a lower operation current. The thermal behavior of the semiconductor laser array with different operation parameters was carried out using finite element method. The structure parameters of semiconductor laser array based on dual-chip integration were optimized and characterized. The output power is 1485 W operated at a current of 700 A and the maximum electro-optical efficiency is 75%, which is the record-high level for a high power semiconductor laser array.

High output power GaSb-based diode lasers are critical component for 2μm laser systems. We compare four structures with different layer thickness combinations to optimize lower cladding layer thickness. Four structures have similar optical confinement factor of active region. As the lower cladding layer thins, the threshold current increases and the series resistance slightly reduces. Among the four structures, laser with 370nm waveguide layer and 1200nm n-type cladding layer functions the best. An output power of 1.21W at 3A is obtained, the threshold current is 0.11A, the series resistance is 0.25Ωthe slope efficiency is 0.42W/A.

In this work, a filter-free scheme of photonic generation of a high-quality 32-tupling millimeter-wave signal based on four dual-port MZMs (DP-MZMs) is proposed and simulated. In this scheme, four DP-MZMs are recombined into the two parallel DP-MZMs and the two cascaded DP-MZMs, respectively. After being modulated by a local oscillation signal, the light from the two parallel DP-MZMs is converted by a photodetector into an electrical signal, which is imposed on the electrodes of the two cascaded DP-MZMs. The simulation results show that the performances of the photonic microwave signal are affected by the modulation index, extinction ratio, and bias voltage of the DP-MZMs. Under the optimized operation parameters, a high-quality 32-tupling millimeter-wave signal with 46 dB radio frequency sideband suppression ratio (RFSSR) and 40 dB optical sideband suppression ratio (OSSR) is generated without using the filter. Because the two cascaded DP-MZMs are modulated by a small signal and there is no filter in the system, the high stability and high utilization rate of the sideband can also be realized in this scheme.