This PDF file contains the front matter associated with SPIE Proceedings Volume 11048, including the Title Page, Copyright information, Table of Contents, Author and Conference Committee lists.

We propose adaptively loaded set-partitioned offset quadrature amplitude modulation (SP-offset-QAM) orthogonal frequency division multiplexing (OFDM) for low-cost intensity-modulation direct-detection (IM/DD) communication systems. We compare this scheme with conventional OFDM in 40-Gbit/s experiments over 50-km single-mode fiber. It is shown that the use of SP-QAM formats, together with the adaptive loading algorithm specifically designed to this group of formats, improves the performance of offset-QAM OFDM. It is also shown that the proposed scheme exhibits better performance than SP-QAM based conventional OFDM at back-to-back and after 50-km transmission.

A simple optical orbital angular momentum (OAM) mode generator based on silicon-on-insulator (SOI) strip waveguides is proposed, which is consisted of a coupled waveguide and a trench waveguide. The fundamental mode TE₀₀ is coupled to the second-order mode via an asymmetric directional coupler. Single-trench waveguide can support two orthogonal LP-like modes whose optical axes are rotated by around 45° with respect to the horizontal and vertical directions. We simulate and analyze the mode properties and propagation effects of OAM modes with charge numbers of 1or -1 by FDTD. When the phase difference between two LP-like eigenmodes is π/2,the second-order mode is further converted to the OAM mode over a wide wavelength range from 1.43μm to 1.58μm.The simulation results indicate that the loss can achieve approximately 0.16 dB. The proposed device is very compact with footprint of <47μm×2μm and the mode conversion efficiency is over 97%. Thus, such structure of OAM mode generator is a promising candidate for applying in OAM multiplexing system and other fields.

Compared to conventional orthogonal frequency-division multiplexing (OFDM), orthogonal frequency-division multiplexing offset quadrature amplitude modulation (OFDM/OQAM) system relaxes the orthogonal condition from the complexed field to real field, cyclic prefix (CP) could thus be removed to promote spectral efficiency. However, there exists intrinsic imaginary interference (IMI) for OFDM/OQAM systems when faced with multiple-path fading channel. Therefore, developing effective channel estimation method to suppress IMI effect is very important for maintaining the performance of the OFDM/OQAM systems. Recently, numerical discussions on channel estimation method for coherent optical OFDM/OQAM (CO-OFDM/OQAM) have been reported. For these works, although frequency domain channel estimation method based on the using of pseudo pilot have been established, the correlation of the noise on each subcarrier brought by the half period time offset between the real and imaginary part, have not been taken into consideration. In this case, the channel estimation accuracy might not be optimal. In this paper, we proposed intra symbol frequency domain averaging (ISFDA) channel estimation method based on the maximum-likelihood algorithm for this system to decrease the interference induced by the correlation property. We analyzed the correlation property of the received amplified spontaneous emission (ASE) noise on each subcarrier for CO-OFDM/OQAM system and performed weighed averaging method for the frequency domain channel transmission matrix. As shown in the theoretical analysis and simulation results, the interference induced by the correlation property of the ASE noise could be suppressed significantly thanks to the using of the ISFDA method.

Compared with the fiber channel, the changes of polarization state and the channel characteristics of the two polarization states are relatively slowly in space optical communication channel. At the same time, there have no serious PMD effect in the atmosphere. Because of the relative stability of polarization states in space optical communication systems, we propose and experimentally validate a method that performs low-speed polarization de- multiplexing processing first, and then performs low-speed linear channel equalization for each polarization channel in the DSP process. While the polarization de-multiplexing is realized by ICC algorithm, and the channel equalization is realized by CMA. What’s more, the tap coefficients of ICC algorithm and CMA are able to update by processing the received information sequence in low speed with low-speed logic cells in this method. Therefore, the advantage of this method is that the logic resource consumption in hardware implementation will be reduced by reuse of the low-speed logic cells during FPGA processing while the ICC algorithm and CMA are running. In order to verify the effectiveness of the scheme, we build a communication system that 20-GBd PM-QPSK signals are sent at the transmit-side, and a coherent receiver with local oscillator at the receive-side. By analyzing the constellation diagram and comparing the BER after the ICC algorithm and CMA processing with that after CMA processing, we know that this method simplifies the DSP effectively without significant impact on communication quality.

For the advantage of high spectral efficiency and lower out of band power radiation, orthogonal frequency-division multiplexing offset quadrature amplitude modulation (OFDM/OQAM) system have aroused wide interest. For OFDM/OQAM, prototype filter with promising time-frequency localization properties have been employed to suppress the out of band radiation. Time offset between the real and imaginary part of the QAM symbol and the phase offset module guaranteed the real filed orthogonal condition of the sub-carriers for OFDM/OQAM. Most recently, numerical experimental demonstrations for coherent optical OFDM/OQAM (CO-OFDM/OQAM) have been reported. Polarization-division-multiplexed structure could promote the spectral efficiency of CO-OFDM/OQAM further. However, chromatic dispersion (CD) and polarization mode dispersion (PMD) effect brought serious intrinsic imaginary interference (IMI) to PDM CO-OFDM/OQAM. Thus, effective channel estimation methods have to be developed to combat IMI and maintain system performance. Although numerical frequency domain channel estimation method based on the concept of pseudo pilot have been discussed, the correlation of the amplified spontaneous emission (ASE) noise brought by the time offset module decreased the channel estimation accuracy. To our best knowledge, channel estimation method based on weight averaging have not been discussed for PDM CO-OFDM/OQAM system. In this paper, we proposed joint intra symbol frequency domain averaging (JISFDA) channel estimation method based on the maximum-likelihood criterion for the PDM CO-OFDM/OQAM system. Compared to the conventional channel estimation method which only use single subcarrier for channel estimation, the JISFDA method considers the correlation of the ASE noise on each subcarrier and utilizes weight averaging method for the channel estimation matrix. As shown in the theoretical analysis and Montel Carlo simulation results, the interference induced by CD, PMD, and the correlation property of the ASE noise could be suppressed significantly with the using of the JISFDA method.

A grating-lobe-suppressed two-dimensional (2D) optical phased array (OPA) is proposed to realize optical beam steering based on unequally-spaced technique, in which the element distribution is optimized by a modified genetic algorithm to achieve a minimum peak side-lobe level (PSLL). Numerical simulations of 2D unequally-spaced OPAs are carried out. The results show that by optimizing the element distribution in an unequally-spaced OPA using the modified genetic algorithm, the grating lobes and side lobes can be well suppressed. Specifically, the PSLL of the far-field pattern reaches as low as 0.20 in 0° beam direction for a 2D 10×10 OPA with the inter-element spacing between 1 μm and 3 μm, which is much better than the traditional unequally-spaced OPA. The relationship between the optimized PSLL and the practical fabrication accuracy is also investigated. The results indicate that the proposed OPA can allow a certain fabrication deviation. Besides, the additional amplitude weighting method is applied to the optimized 2D unequally-spaced OPAs, and the simulation results show that the optimized unequally-spaced OPAs with amplitude weighting can achieve a better grating-lobe suppression. For the 2D 10×10 OPA, the PSLL can be further reduced to 0.18 in 0° beam direction. The proposed 2D low grating-lobe OPAs are expected to benefit many practical applications such as light detection and ranging, high-resolution display, and free-space optical communications.

Speech enhancement is a critical part of variety types of communication systems and automatic speech recognition (ASR) applications. In this study we propose a speech enhancement method for real time VoIP applications with stacked frames and deep neural network, a novel data preparation approach is also introduced. In contrast to many states of art learning-based method, we focused on real-time implement in VoIP applications. Experiments were conducted on speech degraded by different noise types and SNR levels which were not seen in the training stage of the deep neural network and achieved a significant improvement on PESQ. Important traditional real-time speech enhancement method and most recent states of art learning-based method were also tested and compared with proposed method. The results show that proposed method effectively improve the speech intelligibility, greatly outperform traditional real-time minimum-mean square error (MMSE) algorithm and real-time learning-based CNN method in PESQ. We also achieve comparable PESQ in comparison with most recent state of the art learning-based method, but outperform it in time complexity. Making this method attractive in VoIP communication system applications which is high demand on communication latency.

A simple photonic approach for the multiband phase-coded microwave generation based on a dual-drive Mach-Zehnder modulator (DDMZM) is proposed. By injecting the 5-bit barker code sequence to the bias of DDMZM, the binary phase-coded microwave signal for multiband radar application is generated. A proof-of-concept experiment is performed. The generation of phase-coded signals tuning from 1 to 9 GHz with 1 to 20 Mbit/s coding rates is verified.

We propose frequency domain shifting and time domain scrambling method based on 4D hyper-Chen chaos to improve physical-layer security of CO-OFDM systems. The proposed encryption scheme has key space of 10⁴²⁹ and can combat exhaustive attacks effectively.

Corrosion is a serious issue causing damage in steel facilities. Timely inspection and repair is essential to avoid unprecedented structural failures. Employing non-destructive methods of manual inspection for large number of antennas to detect corrosion and related damages is time consuming and expensive. In addition to this, safety of inspector to climb structures possibly weakened by corrosion. In such a situation non-contact approach of automated visual inspection for corrosion and related damage detection through image processing of aerial based images is a viable option. For robust corrosion segmentation and detection, we investigate color classification based on random forest. A random forest is a statistical framework with a very high generalization accuracy and quick training times. We evaluate random forest based corrosion detection and compare it to Bayesian network, Multilayer Perceptron, SVM, Naive Bayes and RBF network. Results on a database of real images with manually annotated pixel-level ground truth show that with the IHLS colour space, the random forest approach outperforms other approaches.

Optical network on chip (ONoC) is an emerging paradigm for multi-core processor system for providing high bandwidth and low power consumption. The scalability of the multi-core processor system is limited by process yield and power density of the single chip. Multi-chip ONoC, which aggregate several smaller chips together, is proposed to overcome this problem. Nesting Ring Optical Network on Chip (NRO) is a multi-chip architecture, which has good performance in terms of throughput and delay. When the NRO scales to a large size ONoC architecture, it may suffer from severe congestion. We propose a novel resource reservation scheme for NRO to alleviate this problem. The resource reservation includes resource planning strategy and a Resource Allocation strategy. With our observation, lots of contentions occur in intersection nodes due to the lack of resource, because intersection nodes have heavier traffic than normal nodes. The proposed resource planning strategy can offer more optical resource for intersection nodes specifically, and keep the original number of wavelengths in the normal nodes, which maintains good power efficiency of NRO. Additionally, unreasonable resource allocation method will lead to contentions. We employ backward resource reservation method in NRO architecture to avoid contentions in intermediate nodes of the path. The simulation results show that the proposed resource reservation scheme can reduce 85% blocking rate and 12% ETE delay compared with convention scheme at an injection rate of 0.4.

A novel phase modulator-based microwave photonics link (MPL) with improved spurious-free dynamic range (SFDR) is proposed, in which a parallel optical sideband processing path is used to generate the opposite third-order intermodulation distortion (IMD3) for destructive combination. By controlling the magnitude of the generated IMD3 term via attenuator in one path, the suppression of IMD3 term was achieved. A theoretical analysis is presented and the simulation experiment results indicate that the SFDR is up to 128.582 dB•HZ^2/3, which has an improvement of 23.66 dB compared with the nonlinearized link.

We investigate virtual-network-function placement and scheduling problem in optical datacenter networks, considering the installation/de-installation latency of VNF and the rapid variation of low-latency-demands. The proposed scheme achieves low blocking probability, latency, and spectrum resource consumption.

Photonic generation approach of linearly chirped microwave waveform(LCMW) with tunable frequency and bandwidth multiplication factor(FBMF) based on parallel Mach-Zehnder modulator(MZM) is proposed. Theoretical analysis show that LCMW with FBMF of 4, 8 and 12 can be obtained by properly adjusting the amplitude of linearly chirped microwave drive signal and direct current(DC) drive signal. The scheme greatly reduce the frequency and bandwidth of electrical linearly chirped microwave drive signal. Due to no filter is employed, so the generation LCMW has a large frequency and bandwidth tunable range. Furthermore, the feasibility of the approach is demonstrated by the simulation based on OptiSystem platform.

We have experimentally demonstrated a direct-detection (DD) 112-Gbit/s 16 quadrature amplitude modulation (QAM) transmission over single-span 140-km standard single mode fiber (SSMF) with Kramers-Kronig receiver and a sparse I/Q Volterra filter (VF). The sparse I/Q VF was proposed in our previous work and it is based on dual-input real-valued Volterra series and ℓ₁-regularization method. In this paper it is used for compensating the nonlinear distortion in a short-reach DD optical 16-QAM signal transmission system. In back to back case, sparse I/Q VF represents the great compensation ability to the saturation effect of the electrical amplifiers and the nonlinear sinusoidal transfer function of I/Q modulator. It provides around 1-order magnitude improvement of BER when reducing 84% complexity from full I/Q VF. For fiber transmission case, sparse I/Q VF can mitigate the fiber nonlinearity effectively and it achieves single-span 140-km transmission at hard-decision forward error correction (HD-FEC) threshold of 3.8 ×10^-3 with less than half complexity of full I/Q VF. Besides, optical signal noise ratio (OSNR) performance at 120 km is measured and sparse I/Q VF reduces the required OSNR at HD-FEC threshold by 1.3 dB. In a word, we investigate the performance of sparse I/Q VF in short-reach optical 16-QAM transmission system and sparse I/Q VF reveals its potential in the growing short-reach applications, such as data center inter-connection and metropolitan area network.

Time division multiplexing (TDM) is common in visible light communication system, which means a channel carries two or more services and different services have the same communication demand. If a channel carries two businesses that have different communication requirement, for example, the bit error ratio (BER) for one is 10^-3 and for the other is 10^-5 , what can be done to help solve the problem. This paper proposes a time-domain hybrid modulation scheme for visible light communication with different service requirements. By appropriately designing a time division multiplexing frame and time slot occupancy ratios of different modulation formats, a continuous adaptation between spectral efficiency and different tiers of BER requirements can be realized for an optical channel. Such a transmission approach is called time domain hybrid modulation (TDHM). Common on-off keying modulation (OOK) and pulse position modulation (PPM) are used in the time domain hybrid modulation scheme in this paper because its simplicity of OOK and high power efficiency of PPM, which can solve the problem of different communication requirements in VLC. In this paper, when the demand of BER for OOK is 10^-3 and for PPM is 10^-5 , SNR for OOK and PPM are 16dB and 12dB. But if Hybrid modulation is used, the system can satisfy the demand of BER=10^-3 when SNR is just 14dB.

A DC bias must be added to the multiple subcarrier modulated (MSM) system because of nonnegative performance of the optical intensity, which results in less power efficiency of MSM system. In this paper, three basic block code techniques and bias selection are described to improve the power utilization for MPSK subcarrier modulated optical system. The simulations demonstrate that the minimum power block code combined with time-varying bias has the best-performing performance when subcarrier number N≥ 3. In reserved subcarrier block code scheme, the minimum value of the normalized power requirement varying with the reserved subcarrier number L. The higher order of modulation, the more average power requirement could be got by block code than by choosing bias scheme. On the other hand, the performance of rate-compatible punctured convolutional code (RCPC) and combing code are also separately given. Compared to other coding techniques, RCPC combined with varying amplitude block code has lest normalized acquired power under time varying bias when N ≤ 4.

Although the ambiguity function of Beidou signal is roughly shaped like a pin, there still exists some sidelobes, which would inevitably reduce target detection performance of passive radar to a certain extent. However, this feature has been ignored in the existing literatures. In this paper, the signal characteristics of Beidou passive radar are analyzed, and a simultaneous range domain and Doppler domain window adding technique is proposed, which can reduce the sidelobes effectively and improve conventional target detection performance over 3 dB. Simulation results demonstrate the effectiveness of the proposed method.

In this paper, passive radar imaging analysis based on spatial spectrum theory is studied. It has a strong universality and can better reflect the intrinsic characteristics of passive imaging. We focus on the discussion of imaging performance from the perspective of spatial spectrum filling, and qualitatively analyze the influence of transceiver numbers, inversion algorithms and motion factors on the imaging results, so as to provide technical guidance for designing imaging system using non-cooperative illuminators. Theoretical derivation and simulation results confirm the effectiveness of related analysis.

We propose and demonstrate a segmental frequency domain pre-equalizer (SFDP) for OFDM-VLC systems. Experimental results present that the SFDP outperforms the conventional uniform pre-equalizer and the spectrum efficiency is increased by 21.2%.

This paper proposed a model of traffic control system which used the technology of visible light communication. By using some simulation software like Python or MATLAB, the optimized parameters can be found. For example, the best angle of the traffic light is 8.76 degree, the minimum and maximum distance of the light can received are 4.46 and 97.25 meters in this system respectively. With these parameters, we can also simulate the distribution of incident power on the road, including the crossing road or the roundabouts. As a result, incident power always centralizes on the center of the road. Eventually, some solutions are found to deal with these problems.

A digital signal processing (DSP) scheme based on Volterra equalizer (VE) combined with adaptive noise-whitening post-filter and maximum likelihood sequence detection (MLSD) is proposed to mitigate nonlinear impairments in vertical-cavity surface-emitting lasers (VCSEL) multimode fiber (MMF) system. Successfully transmission of 108 Gb/s, 100 Gb/s and 60 Gb/s 4-ary pulse amplitude modulation (PAM4) signal over 5 m, 160 m and 460 m OM3-MMF is demonstrated below the 7% overhead hard-decision forward error correction (HD-FEC) bit error rate (BER) threshold by using a 20-GHz class VCSEL at 850 nm. Linear pre-equalization is applied to mitigate severe bandwidth limitation of the system. Our experimental results show that the scheme can well mitigate modulation nonlinearity induced by VCSEL and fiber nonlinearity induced by MMF. The BER decreases about two order of magnitude compared to linear equalizer after 100 m OM3-MMF transmission for 100 Gb/s PAM4 signal.

A shared regeneration location scheme including an OSNR Deep Neural Networks calculation model and a shared regeneration location algorithm is proposed and demonstrated for regenerator placement problem. Simulation results show that it performs better on real scene simulation with OSNR and reductions of number of regenerations and regeneration sites.

Colorimetric characterization can reduce the distortion of image color information in the process of reproduction of the display device, so as to ensure that the same image can be accurately transmitted and reproduced. In order to realize the precise color characteristics of LCD, this paper uses the steepest descent method to optimize the parameters of GOG model, and establishes the color space conversion model from RGB to CIEXYZ and analyzes the model accuracy. The experimental results show that the maximum color difference of the model is 4.7494, and the average color difference is 2.7435, which can meet the accuracy needs of LCD colorimetric characterization.

In order to solve the problem that the same image has different display results on different monitors, the color characteristic of the display is needed. In this paper, the least square method is used to fit the experimental data，the polynomial regression method is used to build the RGB to CIEXYZ color conversion model of the display, and the accuracy of the model is analyzed. The experimental results show that the accuracy of the cubic polynomials curve model is the highest, the maximum color difference is 5.2862, and the average color difference is 2.6510.

Various approaches have been put forward to reduce impairments affected by turbulence fading in free-space optical(FSO) communication systems. In this paper ,We study the ASEP (average slot error probability )and APEP(average packet error probability) performance of PPMs approximately in optimal detection threshold while channel state information(CSI) is best estimated in receiver. In our framework , We mainly focus the weak turbulence with log normal fading and these two probability(ASEP and APEP) are calculated by the method proposed by Holzman. Furthermore, ASEP and APEP are simulated to study how the performance metrics are affected by the atmospheric conditions and other PPMs parameters such as modulation order M , numbers of slots m, pulses q parameters and numbers for information bits N.

In this paper, we propose an optimized approach to generation algorithm of optical true time delay pool (OTTD-P), by building the forbidden wavelength spacing matrixes (FWSMs), for the centralized control of beamforming (CCBF) system. In this optimized generation algorithm, we simplify the complex processes in the traversal process by FWSMs, so only a simple 0/1 decision is needed in the traversal cycle. And the design of OTTD-P correspondingly changes into a compatibility problem of the wavelength sequence heads. Moreover, a proof-of-concept CCBF system with 8-element phased array antennas is demonstrated to verify the feasibility of the optimized generation algorithm, in which 12 basic beam directions based on a 29.5 GHz millimeter wave signal are formed to provide a 360-degree beam steering. The time complexity of processes in the traversal algorithm is about 7×10²³ steps which is a product of 2376 steps in the verification process and 3×10²⁰ steps in the traversal process. But the time complexity is only about 2×10¹⁰ steps in the optimized generation algorithm. Finally, based on the optimized generation algorithm, a compression of needed optical wavelengths can be achieved and the rate of compressed-to-uncompressed (CTU) optical wavelengths decreases from 95% to 19% with the increasing beam directions.

In this paper, to mitigate frequency selective power fading induced by multipath effect, a clustering algorithm based on k-means is proposed for OFDM-VLC system. The simulation results show that, aided by the k-means clustering algorithm, the signal-to-noise ratio (SNR) is improved by more than 7 dB compared to the scheme without k-means clustering algorithm at BER of 3.8×10^-3, the hard decision forward error correction (HD-FEC) limit.

Human activities are highly predictable, so are the network traffic. It appears in a tidal phenomenon in traffic. Current resource allocation methods cannot adapt to tidal traffic, resulting in low resource utilization. To predicted tidal traffic, we have established a neural network model optimized by adaptive artificial fish swarm algorithm. Then we propose a novel multi-path pre-reserved resource allocation strategy to increase resource utilization. The results confirm the effectiveness of our method.

In this paper, a detection method of INS horizontal and azimuth reference datum deviation is designed, by using an electronic level meter and two electronic theodolite, which refers to water platform and azimuth reference mirror. The method has good results in practicability, under the condition of the laboratory.

In order to detect weak light in telecommunication wavelength, InGaAs/InP single-photon avalanche diode (SPAD) is receiving increasing interest. The operation conditions which contains excess bias and operating temperature will effectively influence the performance of SPAD. After analyzing the structure of SPAD, the integrated model of intrinsic parameters of SPAD including photon detection efficiency (PDE) and dark count rate (DCR) is built, which is based on the excess bias and operating temperature. Moreover, the integrated model is compared with the measured data, and it is verified to be able to predict the PDE and DCR accurately. Then, combining with the bit error rate (BER) model of photon-counting based optical wireless communication, the optimal operation conditions of SPAD are investigated. The simulation results show that with the increasing of excess bias from 0-V to 10-V, BER can be effectively reduced. And keeping the excess bias at 5-V can make the BER reach its bottom, when excess bias is limited by the gating circuits of f5-V. In addition, cooling the SPAD to operate at around 245-K could let the system work at the optimal BER.

In this paper, we propose a two-stage bit mapping based irregular QC-LDPC coded BICM scheme to further improve the BER performance of BICM system. Firstly, with the aid of EXIT chart, the irregular code ensemble with optimal convergence threshold can be found, then PEG algorithm is used to construct the required irregular QC-LDPC code with possibly enlarged girth from the ensemble. Secondly, a two-stage bit mapping design method is introduced to find the optimal bit mapping distribution and the near optimal bit mapping table, which can further improve the convergence threshold of the BICM system. Simulation results show that, when applying the proposed scheme to BTB coherent optical PM-64QAM system, additional 0.13dB net coding gain can be achieved compared with the traditional BICM scheme at the BER of 10^-5.

Flexi-grid elastic optical network (EON) provides a promising infrastructure for datacenter (DC) interconnection with its flexible spectrum allocation and high-capacity transmission. Since failures in inter-datacenter elastic optical networks (IDC-EONs) will result in vast amounts of request interruptions and data loss, much recent research has focused on the survivability of IDC-EONs. In this paper, we investigate how to improve network survivability and provide 100% request protection under single-datacenter and single-link failures (single-DLFs) scenario by leveraging multiple DCs that can provide the same services. A shared backup path protection based on minimum spectrum occupancy combination strategy (SBPP-MSOCT) is proposed, which combines the existing spectrum occupancy of the network and select working and protection paths with the least spectrum consumption. Numerical results demonstrate that SBPP-MSOCT can effectively enhance network survivability while reducing spectrum resource consumption and request blocking probability compared to reference algorithms.

Mode division multiplexing (MDM) has been widely investigated to enhance the capacities of passive optical networks (PONs), and weakly coupled transmission schemes are highly preferred to reduce the cost and complexity. In this paper, we have proposed an evolution scheme of the cascaded mode division multiplexing PON (MDM-PON) and conventional Gigabit PON (GPON), which supports bidirectional 1310/1490-nm transmission. This scheme is enabled by wavelength-insensitive weakly coupled (WIWC) few-mode fiber (FMF) and optical components. The bidirectional MDM-GPON transmission scheme with 2.5 Gb/s on-off-keying (OOK) modulation and simple direct detection over 10-km FMF and 10-km single-mode fiber (SMF) are experimentally demonstrated. The proposed MDM-GPON evolution scheme can achieve maximum compatibility by maintaining both the optical distribution network (ODN) and the optical network units (ONUs).

This paper deals with infrared and visible image fusion problems by maximum a posteriori probability (MAP) estimation. We use imaging mode to construct the conditional probability distribution, assume the fusion image approximate the visible image, and treat sparse property of image gradient as fusion image prior probability distribution. According Bayesian theorem, the fusion image’s posterior probability distribution is deduced. The fusion results are obtained by maxim the posterior probability distribution. In experiments, we conduct subjective and objective evaluation. The comparisons show that the MAP-based image fusion method has better performance in both subjective and objective evaluations. The MAP-based image fusion method can be applied to image interpretation, detection and recognition tasks.

In this paper, a multi-passband microwave photonic filter (MPF) based on multiple dispersive devices has been proposed and experimentally demonstrated. The Mach-Zehnder interferometer (MZI) divides the broadband light source (BBS) into multiple optical taps, and with the combination of different dispersion mediums such as chirped fiber Bragg grating (CFBG) and single mode fiber (SMF) to delay the optical tap, a MPF with multiple passbands can be simply achieved. The number of the passbands can be easily controlled by changing the number of the dispersion medium. In the experiment, the frequency response result of the four passbands is obtained by accessing two CFBGs and two SMFs. In addition, by adjusting the wavelength interval of the interference spectrum with a variable optical delay line (VODL), all passbands of the filter can be simultaneously tuned. The filter has broad application prospects in the fields of modern wireless and satellite communication, optoelectronic oscillator and optical sensing.

Free-space optical communications (FSO) suffers from irradiance fluctuation caused by atmospheric turbulence, which results in optical power loss and consequently decreased signal-to-noise ratio (SNR). The error performances of the FSO based on On-Off Keying (OOK), Differential Phase Shift Keying (DPSK), and Binary Phase Shift Keying (BPSK) schemes in a turbulent atmosphere are presented. The received irradiance after propagating the atmosphere is modeled using the gamma-gamma distribution to evaluate the system error performance in turbulence regimes from weak to strong. The results show that, to obtain a BER of 10^-6 at weak turbulence regime, ~15 dB and ~18 dB SNRs are required for BPSK and DPSK, respectively. However, for OOK with a fixed threshold of 0.5 under the same channel condition, OOK reaches an error floor greater than 10^-3. The values of SNR required to achieve the same BER increase as the turbulence strength increase to moderate and strong regimes.

We design a highly stable DFB laser source, with the 15-minute stability of the output optical power up to ± 0.0012 dB and the 8-hour stability up to ± 0.006 dB at 1310 nm.

We propose an adaptive resource allocation framework for on-demand communications in a software-defined mobile fronthaul (MFH) network that supports dynamic processing resource sharing. Our theoretical and experimental studies point to the feasibility of secure bidirectional transmission with guaranteed bit error rate (BER) service using adaptive modulation and coding.

Short term traffic flow prediction is of great significance for easing traffic congestion and maximizing road carrying capacity. This paper proposes an effective algorithm for traffic flow prediction. Firstly, the algorithm analyzes the characteristics of daily traffic flow. According to the difference, the daily traffic flows are divided into workday type, and holiday type, and each type of data is integrated to predict the corresponding day type traffic flow. Then based on phase space reconstruction, a chaotic local prediction algorithm is proposed. The algorithm uses Euclidean distance to select phase space reference neighborhood successively, and support vector machine is used to establish the mapping relationship between neighboring points. This algorithm is used to predict the data of an intersection in Guangzhou, and satisfactory prediction accuracy has been achieved.

Implement of Kalman scheme for RSOP equalization is investigated from hardware perspective and simulation results show that Kalman scheme is superior to CMA in terms of hardware resources utilization and recovery capacity.

Conventional space-time trellis code cannot be directly applied to the intensity modulation/direct detection atmospheric optical transmission systems due to the complex and negative signal problem. An asymmetrically clipped space-time trellis code scheme is proposed, which combines the Hermitian symmetry of the inverse fast Fourier transform and the N/4 points to N points signal mapping to make the processed bipolar signal sequence satisfies the odd harmonic symmetry. The conversion of complex signal to positive real signal can then be achieved by clipping the bipolar signal sequence. Monte Carlo method is utilized to verify the feasibility of the proposed scheme. The results show that the proposed method can effectively resist the atmosphere scintillation.

The ergodic channel capacity of MIMO WOC systems with combined effects is investigated under correlated channel. The results indicate that spatial correlation has the most serious influence on ergodic channel capacity, followed by the pointing error and finally the atmospheric turbulence.

Transmitter IQ skew severely degrades the performance of coherent optical communication systems, especially for high baud-rate and higher order quadrature amplitude modulation (QAM) systems. Utilizing Nyquist pulse shaping can increase the spectral efficiency of WDM systems, while sharp roll-off of Nyquist pulse shaping further reduces the tolerance to residual IQ skew. Thus, there is a tradeoff between WDM system spectral efficiency and its tolerance to residual IQ skew. In this talk, we review the impact of Nyquist pulse shaping on the tolerance to residual IQ skew. The roll-off factor, channel spacing, receiver bandwidth and equalizer length are optimized given certain residual IQ skew. We further discussed the potential application scenarios and the methodologies to apply this technique. The results serve as guideline for high baud-rate coherent long-haul optical communication systems.

The interaction is widespread between light and particulate matter. We propose a growth theory of particulate matter by nonequilibrium statistical concepts and methods. The theory connects macroscopic properties with microscopic physics mechanism. In this paper, the stochastic equation for particulate matter growth is given, and the fluctuation growth coefficient and transport growth coefficient are obtained. The cluster radius distribution function is derived.

The application of asymmetrically clipped optical-orthogonal frequency division multiplexing (ACO-OFDM) in free-space optical (FSO) communication system can increase the system channel capacity, while the use of channel estimation technology can ensure that ACO-OFDM has better performance. Since the channel estimation performance is directly affected by the selection of pilot patterns, especially when using compressed sensing (CS), it has been a major task in OFDM transmission schemes. In order to improve the channel reconstruction accuracy, a hybrid optimization algorithm based on genetic algorithm (GA) and particle swarm optimization (PSO) algorithm is proposed for two types of pilot design criterion. Simulation results show that compared with other three algorithms, the proposed algorithm has faster convergence speed and lower convergence value. With the pilots designed by the proposed algorithm, the system can achieve higher channel reconstruction accuracy.

Considering the weather’s influence to atmospheric transmittance, the paper modifies the attenuation matrix in the basement of original MIMO FSOC system channel model. And then, based on the modified channel model, the capacity in different given conditions is shown through numerical simulation. The paper simulates the capacity in different transmitting antenna’s height as well as in different weather. Considering that in most conditions, the optical signal doesn’t transfer vertically, the paper stimulates the capacity versus system slant angle. The paper also simulates difference value of capacity. At last, the paper designs a software to evaluate whether the given conditions are fit for communication or not.

Indoor visible light positioning (VLP) system has drawn more and more attention in recent years. In this paper, we propose a new positioning scheme, which utilizes a close-range signal selection method and weighted mean algorithm based on received signal strength ratio (RSSR). The new scheme can effectively improve the positioning accuracy at the corners and edges of the room, that is, it can obviously reduce the impact of reflections on indoor VLP system. Simulation results show that compared with the RSSR method, the proposed scheme has improved the positioning error by at least about 38 cm, and can achieve an average positioning error of 0.0806 m in the case of 25 LEDs. Then, the positioning errors are further analyzed for several different layouts of LEDs.

The scheme of frequency tripled and quintupled phase shift keying (PSK) signal without phase multiplication based on two parallel phase modulators (PM) is proposed. To eliminate the impact of the reduction of the vector signals’ Euclidean distance induced by the phase-precoding, a novel phase precoding method is proposed and applied to the original PSK signal. A phase-precoded PSK signal with the same amplitude, same carrier frequency but opposite phase compared with the original PSK signal is generated and modulates the optical carrier via one PM. The original PSK signal modulate the same optical carrier via the other PM. A pair of proper harmonics from the output of the PMs is selected by the wavelength selective switch (WSS). Then the optical millimeter wave (MMW) signal is injected into the fiber. After certain distance of the fiber link transmission, the desired MMW PSK signal is generated after the Photodiode (PD) detection. To verify the feasibility of the proposed scheme, a radio over fiber (RoF) system with 1 Gbit⁄s rate data has been implemented and the tripled and quintupled frequency of QPSK signal has been generated by numerical simulation. The error vector magnitude (EVM) of the 60-GHz and 100-GHz QPSK signal has been measured versus the transmission distance of the fiber link and the launched power into PD with the optical signal noise ratio (OSNR) of 15 dB and 18 dB respectively. And the result shows that the proposed MMW PSK signal generation scheme could realize tripled and quintupled frequency upconversion of the QPSK signal and reliable transmission at least 80 km when the launched power into PD is higher than -5 dBm.

In order to achieve higher data transmission efficiency, a novel design of 64 state trellis coded modulation (TCM) based on three dimension 32QAM signal is presented in this paper. The simulation results show that when the BER is 10^-3, the Eb/N0 of the 64-state 32QAM-TCM scheme is 2.3dB smaller than that of the uncoded-16QAM scheme. The proposed scheme introduces significant performance improvement of bit error ratio over two dimension uncoded-16QAM scheme.

We propose a low-complexity linear interpolation based phase noise suppression method in Nyquist-WDM-PON. The numerical results show that it can achieve better performance as well as lower computation complexity than conventional methods.

Mobility is one of important issues for line-of-sight visible light communication (VLC). For a high-speed VLC system, the field of view (FOV) of the transceivers is very narrow for high optical gain at the receiver so that precise alignment between the transceivers is required. To enhance the mobility performance, in this paper, a simple and effective scheme of tracking light source is designed and experimentally demonstrated for practical VLC systems. To detect the target light source, a beacon light encoded with a low-frequency label signal is used along with the information light at the transmitter, whilst a high-speed camera with wide FOV is used for location detection by differentiating received images at the receiver. The detected signal of the light source location is then digitally processed to control a two-axis gimbal for auto-alignment between two transceivers. To mitigate the effects of ambient and interference lights, adaptive threshold and dynamic search window are used. A field-programmable gate array (FPGA)-based on-off keying (OOK) transceiver with the adaptive tracking function is developed for evaluation of the tracking performance. Experimental results show that with the proposed scheme 25 Mbps OOK transmission is successfully demonstrated at a bit error rate of < 10^-3 over a 3 m VLC link with the transmitter horizontally moving at a speed of 1m/s, which corresponds a tracking speed of >18°/s. The angular accuracy and latency of tracking the LED transmitter are < 0.02° and < 21 ms, respectively.

This work presents an investigation of multipath interference (MPI) of operator's existing PON network. The reflectance of connectors within the ODN link are measured, and the MPI penalty of the link is calculated for some candidate modulation schemes. The result will be great helpful to the ongoing development next-generation FTTx standards.

We propose and experimentally demonstrate a 5 × 100-Gb/s DFT-spread OFDM-WDM-PON employing probabilistic shaping (PS) 16QAM modulation format with direct detection. Each optical carrier at the optical line terminal (OLT) is a 25-Gbaud PS-16QAM OFDM according to the near-Gaussian distribution. The spectral efficiency of the OFDM-WDM-PON can be improved under the same average power and received power sensitivity by PS. The experimental investigation shows that the joint equalization of DFT-spread and PS signal can further increase the tolerance to fiber nonlinear effects in OFDM-WDM-PON.

Network failures are unavoidable and can easily cause huge losses. The occurrence of failures typically results in a number of changes that have to be made to recovery and keep operating the network in a normal manner. Restoration is a common method of network failure recovery. However, the traditional methods of Path Restoration and Link Restoration will be effective only when there are resources that can satisfy the condition in the network. And the resource utilization is not high enough. We propose a network failure recovery method based on reinforcement learning, integrated Path Restoration and Link Restoration. The protection channel of the damaged service flow and the channel of the normal service flow share the bandwidth resource. A simulation is designed to evaluate the performance of the proposed algorithm. Simulation shows that whether there is only one input service flow or multiple input service flows, when the final switched flow cannot find a suitable path in the idle resource, the traffic of the best situation will be the minimal of all possible cases. The scheme of this paper can effectively improve the success rate of network failure recovery with high utilization of physical resources. It is more extensive than traditional methods.

In this paper, firstly, we employed the a modified coupled-mode theory to investigate the backscattering power characteristics of Rayleigh backscattering (RB) and stimulated Brillouin scattering (SBS), synchronously and precisely. We discovered how the RB and SBS power changed with pump powers and fiber lengths. The scattering power curves with different pump powers could be divided into three regions. The first was called “the linear region of RB”, in which the pump power was mainly converted into RB and the RB power grew linearly with pump powers. The second region was called “the stimulation region of Brillouin scattering”, in which the power was mainly converted into SBS. In “the gain-saturated region”, when the pump power was much higher than the SBS threshold, both the RB and SBS power were gain-saturated. And we got the power curves with different fiber lengths. The power curves were helpful to set up proper launch powers in practical engineering applications, such as optical timedome reflectometry systems and distributed fiber sensors. Finally, we verified our theoretical results by experiments.

The number of optical true time delay (OTTD) is too many while operating in wideband and wide-angle condition for optically phased array radar. To solve the problem, a beam-forming method for optically phased array radar with sparse array is proposed. Firstly, the influence factor for main-lobe deviation is discussed. Secondly, a configuration of sparse antenna array is presented and then, on the basis of this, a beam-forming method based on Compressed Sensing theory is put forward. With this method, just a small amount of OTTD is required for wideband and wide-angle scanning. Finally, the effectiveness of this algorithm is validated by the simulative results.

For systems based on array source, owing to discrete single sources in the array, variation in power intensity can usually be detected as the receiving end moves. Although uniform illumination in lighting system has been widely investigated, its implemented light sources are commonly LEDs with Lambertian distribution. We further study an optical design applied for optical wireless communication system with VCSEL in this article. The optical system mainly consists of collimating lens and micro lens array to homogenize power intensity of the communication system based on VCSEL array source which has a Gaussian distribution. Our proposed scheme was verified with commercial optical design software Zemax. The simulation result shows that illumination uniformity can reach up to 91% after passing through this optical system, which can fulfill the requirements of high power uniformity in practical applications. The homogenization of power intensity solves the issue of signal power variation at the moving receiver end in optical wireless communication system, which guarantees stable communication link and robust system performance.

In this paper, we propose a multi-level non-binary quasi-cyclic low density parity check (NB-QC-LDPC) coded modulation (CM) scheme to reduce the decoding complexity while maintaining and/or exceeding the super BER performance of the traditional NB-QC-LDPC CM scheme. Firstly, local optimal searching combined with the classical superposition construction is proposed to construct the required NB-QC-LDPC codes with possibly enlarged girth and less number of the shortest cycles. Secondly, by analyzing the changing trend of the BER performance and decoding complexity of the NB-QC-LDPC code defined over 𝐺𝐹(𝑄) incident to 𝑄. We find that the decoding complexity increases proportionally with the increase of 𝑄, but the best BER performance usually corresponds to a smaller 𝑄. Finally, by replacing the high-order NB-QC-LDPC code based CM scheme to the multi-level low-order NB-QC-LDPC code based CM scheme, the decoding complexity can be effectively reduced. Simulation results show that the designed two-level 𝐺𝐹(4)/𝐺𝐹(8) QC-LDPC code based 16QAM/64QAM CM system can perform almost the same or even better BER performance with/than the corresponding traditional 𝐺𝐹(16) / 𝐺𝐹(64) QC-LDPC code based 16QAM/64QAM CM system.

We propose and experimentally demonstrate a 5 × 50-Gb/s wavelength-division-multiplexing passive optical network (WDM-PON) based on probabilistic shaping (PS) 4-level pulse amplitude modulation (PAM4) with direct detection. Each optical carrier at the optical line terminal (OLT) carries a 25-Gbaud PS-PAM4 signal according to the near-Gaussian distribution. The spectral efficiency (SE) of the system can be enhanced under the same average power and received power sensitivity (RPS) by employing PS-PAM4 at the optical network units (ONU). A blind feedback pre-equalization (BFP) based on the joint constant modulus algorithm (CMA) and decision-directed least-mean-squares (DD-LMS) is proposed to accurately compensate the signal high frequency loss of the electro-optical components such as DAC, optical modulators and electrical drivers at the OLT. The experimental results show that the proposed WDM-PON over 20 km single mode fiber (SMF) transmission are with 1.44 dB superior receiver sensitivity by employing PS-PAM4 and additional 0.78 dB device linear impairment effect improvement by BFP.

This work introduces a 2×2 silicon thermo-optical switch based on photonic crystal waveguides. Photonic crystal waveguides (PCWs) were used to replace conventional channel waveguides in the phase shifters of a Mach-Zehnder interferometer (MZI). This device was fabricated on silicon-on-insulator (SOI) platform. The active length of the PCWs-based phase shifter was only 50μm. Because of the slow light effect of PCWs, the heat needed to generate a phase shift of π reduced dramatically. The experiment results showed that the power needed for phase shifter of π is 6 times less than that of conventional MZI based optical switches.

We demonstrated Erbium-doped mode-locked fiber lasers by using a chirped fiber Bragg grating (CFBG) as a dispersion managing element. Both conventional soliton and M-shape soliton pulses were achieved when the laser operates in the large anomalous and normal regime by controlling the direction of the CFBG, respectively.

The 5G bearer network consists of three parts: pre-transmission, intermediate transmission, and backhaul transmission. The integrated service access node bearer devices can aggregate the pre-transmitted traffic to the wireless devices of the node, and can also upload the intermediate/backhaul services to the upper-layer bearer devices. The integrated service access node requires a rich access service types and high requirements for bandwidth and delay. Packet enhanced OTN (POTN) can well meet above requirements. IP Radio Access Network (IPRAN) has been widely used in 4G bearer network. In order to achieve intelligent upgrade of the network, the POTN+IPRAN scheme came into being. For each operator, it is necessary to reduce the construction cost of the bearer network while meeting the needs of new services and new scenarios in the future. In this paper, we design a cost-minimized algorithm for POTN device in POTN+IPRAN 5G bearer network (CM_POTN). We set two constraints, they are the number of centralized BBUs on each integrated service access node and the number of integrated service access nodes on each integrated service access ring. For the purpose of minimizing the cost of POTN devices, the algorithm attempts to calculate the minimum cost of the main component modules in the POTN devices within the two constraints. Moreover, a simulation is designed to evaluate the performance of the proposed algorithm, and results show that our proposed CM_POTN algorithm can achieve the minimum cost of the POTN devices within the given constraints.

We proposed a new method using DNN to compensate the damage caused by PDL. The scenario this experiment used is 16-QAM signals transmitted at 2x28 GBaud rate in a polarization-division multiplexing system. A DNN is proposed to compensate for the damage caused by PDL. The result shows that, BER of the system is reduced to 1e-3, and the damage caused by PDL can be minimized in the polarization multiplexed optical fiber communication system within 600 km.

The adaptive coupling of making laser beam into single-mode fiber(SMF) plays an important role in free space optical communication. It’s the key point to the coupling of spatial light and single-mode fiber that finding the best position in the transverse plane of fiber precisely. An adaptive coupling system is proposed, which mainly comprises two-dimensional piezoelectric nanopositioning stage, driver, controller, photoelectric detector, coupling lens which is a closed loop control system. The adaptive coupling system uses optimized raster scanning algorithm to confirm the best position accurately. Theories of optimized raster scanning algorithm are expounded. Analysis results of experiment show that the adaptive coupling system proposed can search the best position automatically based on the power of the light which is coupled into the single-mode fiber. At the same time, it can correct the numerical deviation which is caused by atmospheric turbulence, and maintain high coupling efficiency which can improve the performance of the receiver.

Space division multiplexing elastic optical networks (SDM-EONs) are advocated as a promising solution to cater to the significant increasing of network bandwidth demands. In this paper, a novel crosstalk-aware and flexible grouping routing, spectrum and core assignment (CaFG RSCA) algorithm based on mixed super-channel is proposed to suppress the spectrum fragments and improve the network performance in SDM-EONs. The algorithm focuses on the issues of spectrum fragmentation and allocation efficiency with consideration on the inter-core crosstalk in multi-core fiber (MCF). Simulation results confirm that the proposed algorithm gains effective reduction in spectrum fragments and performs better than other congeneric algorithms in terms of the service blocking probability and spectrum resource utilization.

A remote meter reading system was designed based on LoRa wireless transceiver and pic24 series microcontroller. Compared with current domestic meter reading system, this system has the advantage of long transmission distance and simple routing structure. The design of the system includes the model selection of hardware and software framework and the communication protocols. The network of the system is composed of two types of devices (collector and concentrator) with high modularization, which makes it possible to be used in wide range of situations. The key of software design is a task scheduler used time-based improved cooperative scheduling mechanism. The system is capable to manage the data of users' meters in a certain scope with high efficiency in real time, improving the work efficiency and accuracy.

We propose and demonstrate a microwave photonic communication system with tunable wireless multi-waveband based on pulse shaping, direct detection, and multi-wavelength optical comb. Duo to the parameter configuration and RF- photonics modulation, the number of the flat channels output from a multi-wavelength optical comb can be up to 21. In the near terminal unit of the communication system, four adjacent optical sources output from the multi-wavelength optical comb can be obtained. One of the four optical sources is modulated by a RF vector signal with the frequency of 5 GHz. A single sideband optical signal can be obtained from the modulated signal by pulse shaping via a Gaussian optical filter. The single sideband optical signal and the other three optical sources are coupled into a standard single-mode fiber with 50 km, and then reach the remote terminal unit of the communication system, in which the wireless carriers with the frequency of 5 GHz, 25 GHz, 45 GHz, and 65 GHz can be obtained via a photoelectric direct detector. These multi- carriers can be employed for wireless transmission between the mobile terminal and the remote terminal unit; in addition, three pure optical sources utilized for the uplink can be obtained in the meantime.

A machine learning equalization technique based on KNN for 56Gpbs PAM4-GPONis proposed by nonlinear classification characteristics of KNN. Simulation results show that the proposed method can effectively optimizes the performance of equalization and increase bandwidth of the GPON network.

A GFDM-OFDM hybrid modulation scheme is experimentally demonstrated in a 60-Gb/s IM-DD system. 1.9-dB and 3.1- dB receiver sensitivity improvements are achieved compared to the conventional GFDM and OFDM signals in the back-to-back case, respectively.

We propose a multilevel brightness modulation (MBM) scheme for optical camera communications, where a LED array and a k-means clustering algorithm are used to encode and decode the data, respectively. The experiment results show that the MBM scheme with a focal length of 18 mm offers longer transmission distance compared with the conventional 4-level undersampled pulse amplitude modulation with subcarrier modulation (UPAMSM) scheme.

For the first time, a channel estimation method for hybrid ACO-OFDM (HACO-OFDM) based visible light communications systems is proposed. The channel characteristics carried on the ACO-OFDM subcarriers are estimated by the least square (LS) method, and the channel response on the PAM-DMT subcarriers can be obtained by linear interpolation of which on the ACO-OFDM subcarriers. While considering the multipath channel in the paper, the simulation results show that the proposed scheme is able to estimate the channel effectively for HACOOFDM system. After equalization based on the results of the channel estimation, the average BER on the ACOOFDM branch will be lower than10^-3 when SNR>18.2dB, the average BER on the pulse-amplitude-modulated discrete multitone (PAM-DMT) branch will be lower than 10^-3 when SNR>19.4dB.

We compare the bit error rate (BER) performances of discrete Fourier transform (DFT), discrete cosine transform (DCT) and discrete wavelet modulation (DWT) based orthogonal frequency division multiplexing (OFDM) schemes for visible light communications (VLC). The experiment results show that DWT-OFDM-VLC system with low peak to average power ratio (PAPR) offers better BER performance compared to DFT-OFDM-VLC and DCT-OFDMVLC systems.

In this paper, we experimentally demonstrated a free space optical communication system transmission based on continuous wide-spectrum laser at 10Gbit/s over 1km simulated atmospheric channel. The wide-spectrum laser is part of supercontinuum generated by using a self-made picosecond mode-locked laser based on Raman gain solution effect to pump a dispersion compensation fiber which dispersion coefficient are D=-135ps/nm ⋅ km@1550nm. A 10Gbit/s NRZ signal was then modulated on the wide-spectrum laser and transmitted into a tunable simulated atmospheric turbulence pool which can simulate atmospheric turbulent intensity up to 0.59cm of coherent length. Scintillation intensities and bit error rates of transmission based on wide-spectrum laser and narrow linewidth laser were compared. The results show that communication sensitivity reached -10.98dBm at FEC limit. which has 1.53dB improvement comparing with FSO communication system using narrow linewidth laser under weak turbulent condition.

A Mach-Zehnder interferometer (MZI) based on core-offset splicing technique for sensing applications is demonstrated. The interferometer is composed of a section of single-mode fiber sandwiched by two core-offset spliced standard optical fibers. The refractive index (RI) of the external media around the middle section of the MZI could be detected by investigating the peak wavelength of the MZI spectrum. Experimental results show that the proposed sensor exhibits a sensitivity to external refractive index of 17361 nm per refractive index unit and the temperature sensitivity of only 33 pm/°C. The present device has great potential in biochemical and medical sensing due to the advantages including easy fabrication, high sensitivity and excellent compactness.

Collimation is important for long-range transmission in a Wireless Optical Communication (WOC) system. We developed a novel method to control the size of lens used in collimation of VCSEL based WOC systems. In our design example, the divergence angle is well-controlled to be 0.106 mrad achieved by two lenses with each focal length > 60mm and the system total length < 80 mm. Also, we proposed a universal framework based on light field representation for modeling similar optical systems.

Transmitter in-phase/quadrature (IQ) mismatch is difficult to suppress by using traditional methods in coherent optical orthogonal frequency division multiplexing (CO-OFDM) system with higher data rate, as a result an effective compensation method to solve the problem is urgently needed. In this paper, a novel training symbol structure to compensate IQ mismatch is proposed. Based on this structure we deduce the compensation method in frequency domain, and by utilizing the compensation method we can achieve independent compensation of IQ mismatch as well as channel distortion and promote the system performance significantly with the data rate increasing by 30Gb/s compared with the traditional methods. Numerical simulation demonstrates that the proposed method has better BER performance compared with the traditional method at the data rate of 50Gb/s after 640km transmission. When the phase and amplitude mismatch are 15° and 3dB respectively, the OSNR penalty of proposed method can decrease by 6dB while BER reaches to 1×10^-3. What’s more, the tolerance of amplitude mismatch and phase mismatch can be improved about 3.5dB and 20° by the proposed method, respectively.

In Elastic Optical Networks (EONs), elastic operations have been constrained by the assumption that an optical signal has to occupy a dedicated frequency range, with no sharing of spectrum resources with other independent optical signals. Recently, the emergence of signal overlap techniques can break this traditional constraint in theory. In this study, we fully combine the characteristics of the application signal overlap technology, summarize two factors that affect the ultimate transmission distance of an optical signal as the modulation format of the optical signal and the interval between the center frequencies of the two optical signals, and make a reasonable assumption about their correspondence. On this basis, we propose a Routing, Modulation level and Spectrum Assignment (RMSA) algorithm based on Signal overlap for EONs. The paper simulates the two strategies of the algorithm and compares them with the basic algorithm without signal overlap. The results show that it can significantly reduce the bandwidth blocking probability, especially under heavy traffic load scenario.

Chromatic dispersion (CD) is one of the main limitations for high speed intensity modulation and direct-detection transmission system. Electrical dispersion compensation is a cheaper and easier way to compensate CD. In this paper, we compare the performance of chromatic dispersion pre- and post-compensation for optical single side-band (SSB) 4-level pulse amplitude modulation (PAM4) signal with Kramers-Kronig (KK) receiver by simulation and experiment. We put the pre-compensation after Hilbert transform at the transmitter side and the CD post-compensation after KK receiver at the receiver side. The results show that CD tolerance of the dual-drivel Mach-Zehnder modulator (DDMZM) based SSB signal is related to the optical modulation index (OMI) of the modulator in CD pre- and post-compensation scenarios. When the optical signal-to-noise ratio (OSNR) is relatively large, CD post-compensation is more sensitive to OMI than precompensation with KK receiver. We also explore the impact of signal peak-to-average power ratio (PAPR) on the precompensation system by simulation.

A waveform generation scheme based on complex filter and frequency-time mapping is proposed. In the experiment, a programmable processor was used for the complex filter. The shape of the spectrum is controlled by the amplitude frequency response, and the second order phase is controlled by the phase frequency response to realize frequency-time mapping. In experiment, rectangular pulse, sawtooth pulse and comb lines are realized by adjusting the complex filter. The influence of dispersion and spectral 3dB bandwidth on the time domain waveform is also analyzed. The proposed waveform generation is dynamically adjustable due to the dispersion is continuously tunable.

A novel parallelized few-mode receiver for mode-division multiplexed transmission systems has been proposed and demonstrated based on the orthogonality among spatial eigenmodes. The receiver includes only one free-space 90-degree optical hybrid without the need for mode demultiplexing. A 2× 20 Gb/s QPSK 2-mode back-to-back (BTB) transmission system was demonstrated experimentally.

In this paper, a novel blind modulation format identification method is proposed based on fourth-power characteristic value, the effectiveness of this method is demonstrated and discussed in a 20GBaud long-haul coherent transmission system. It has been proved the usage of FFT peak value and variance is able to achieve accurate estimation among DP-QPSK/16QAM/32QAM/64QAM.

A digital-controllable broadband analog optical chaos source is proposed. An analog-digital hybrid electro-optic feedback loop is introduced to enhance the robustness of the semiconductor laser based chaos system. The scheme could generate a broadband optical chaotic signal of high dynamical complexity. Robust digital synchronization strategies can be used for the proposed optical chaos sources. Compared with the existing schemes, the adoption of a semiconductor laser with self-feedback reduces the requirement on the processing speed of digital components. Meanwhile, the hybrid system structure could be simplified markedly. The proposed chaotic source has potential in long-haul secure fiber communication, chaotic radar, random number generation, and so on..

A high resolution fiber optic spectrometer system was set up, which consists of a tunable laser, a fiber optic power meter, GPIB-USB data acquisition cards, and a computer control platform. The wavelength absolute accuracy of the tunable laser (Santec TSL-550) is±2.5pm, and the wavelength setting resolution can be 0.1pm, which makes the system features high resolution. However, the resolution of the laboratory fiber optic spectrometer (YOKOGAWA AQ6317C) can only be set to 20pm. The extinction ratio (ER), free spectral range (FSR) and the trough position of the spectrum of the microfiber knot resonator (MKR) is measured with our system and AQ6317C fiber optic spectrometer. The experimental results of the two methods show that the ER and FSR are offset by 0.675dB and 0.003nm, respectively, and the standard deviations are 0.17% and 0.43%, respectively. From the results, and the stability of MKR is considered, we conclude that the resolution of our system can be better than 1nm due to the resolution of the TSL-550 light source, and the overall standard deviation is also smaller, therefore, the repeatability and stability of our testing system are better than the AQ6317C fiber optic spectrometer, which can better meet the actual measurement requirements.

A secure scheme based on Rubik’s Cube (RC) transformation algorithm with low calculation complexity is proposed and experimentally demonstrated in the orthogonal frequency division multiplexing passive optical network system (OFDM-PONs). The RC transformation algorithm is adopted to generate encryption key stream for symbol substitution, and the chaos is utilized to control the RC transformation. An experiment with 7.64Gbps 16-quadratureamplitude- modulation (QAM) encrypted OFDM data are successfully transmitted over 25km standard signal mode fiber (SSMF). The experimental results indicate that the proposed scheme is an effective and promising method for physical-layer secure optical communication to meet the demands for low implementation complexity and high security performance.

We have proposed a frequency reshaping and compensation scheme for FTN (Faster-Than-Nyquist) CAP (Carrierless Amplitude and Phase modulation) signal based on machine learning for the first time in this paper. The cascaded post equalizer consists of 2 stages. The first stage digital signal processing (DSP) is a Deep Neural Network (DNN), then LMS post-equalization is conducted as the second stage offline DSP. Through this method, we successfully demonstrated a data rate of 1.12Gbit/s FTN CAP 9QAM modulation over 1meter free space transmission with bit error rate (BER) below 7% FEC threshold of 3.8×10^-3. Compared to the traditional FTN CAP modulation without DNN, the proposed method would increase the data rate by approximately 100 Mbps, leading to an improvement of system capacity of 9.8%. The experimental results clearly validate the proposed cascaded two stage DNN-LMS (C2S DNNLMS) can be a promising solution for future high speed VLC system.

The shipboard navigation system on the spaceflight ship is made up of strapdown inertial navigation system (SINS) and GNSS. The experiments of the evaluation of SINS’s attitude error and the differential service functions of StarFire star-station were carried out by using the method of data evaluation to analyze the data accuracy after start-up of SINS, the system positioning accuracy after differential service interruption and the data convergence time after differential service recovery respectively. Based on the above data analysis results, a rough conclusion of data availability during start-up and running process of navigation equipment is proposed.

As an inherent background sound field in the ocean, Ocean ambient noise interact with the seabed frequently in the process of communication, and it is bound to carry the information of the seabed acoustic parameters. Therefore continuous monitoring of the seabed characteristics on a large scale for a long time can be used for the inversion of the bottom acoustic parameters, with the advantages of low cost, high efficiency and environmental friendly. It has become a hot topic of domestic and foreign acoustic and seismic researches in recent years. The work of this paper is to explore the technology of the broadband vertical coherence of ocean ambient noise to retrieve the parameters of the marine environment.

Failures in optical transport networks usually result in lots of services being interrupted and a huge economic loss. If the failures can be predicted in advance, some actions can be conducted to avoid the above adverse consequences. Deep learning is a good technology of artificial intelligence, which can be used in many scenarios to replace humans’ activities. Event prediction is a typical scenario, where deep learning can be used based on a large dataset. Therefore, deep learning can be used in optical transport networks for failure prediction. However, dataset construction is an important problem for deep learning in optical transport networks, because there may be not enough data in reality. This paper proposes a deep-learning-based failure prediction (DLFP) algorithm that constructs available dataset based on data-augmentation for data training. DLFP algorithm is composed of alarm compression, data augmentation, and fully-connected back-propagation neural network (FCNN) algorithm. Besides, a benchmark algorithm (BA) without data augmentation is introduced. A training model is constructed based on massive real performance data and related alarm data within one month, which are collected from national backbone synchronous digital hierarchy (SDH) network with 274 nodes and 487 links in China. Then the training model is used with test dataset to verify the performance in terms of prediction accuracy. Evaluation results show that the proposed algorithm is able to reach better performance for failure prediction compared with the benchmark without data augmentation.

The space target detection system can realize the accurate location and tracking of the space target. The current situation of the measurement of the space target by the optical system is introduced in detail, and the problems in this field existing in our country is analyzed at the same time. Subsequently, the idea of measurement of space target by shipborne large theodolite is put forward, of which the advantages and problems are analyzed by experiment. Finally, the improved method is discussed.

A 4×8 overlapping amplitude and pulse position modulation scheme under non-Kolmogorov turbulent Gamma-Gamma Channel Free-Space Optical Communication system is proposed and simulated. Then the bit-error-rate performance is simulated for further research and application.

We propose and demonstrate an integrated optical filter with variable bandwidth and tunable central wavelength. It consists of two stages of second-order microring resonators with different radiuses. The two stages are serially cascaded, with each stage having a flat-top spectrum. The free spectral range (FSR) of the cascaded structure is expanded via the Vernier effect. The central wavelength of the proposed optical filter is tuned by synchronous tuning of the two stages. The 3dB bandwidth is adjusted via intentional misalignment of the passbands of the two stages. We demonstrate a prototype of such an optical filter on the silicon-on-insulator platform. The radiuses of ring cavities in the two stages are 7 μm and 10 μm, respectively. All the four ring cavities can be thermally tuned independently. The FSR of the fabricated device is around 90 nm. We show the tuning of the central wavelength from 1460 nm to 1550 nm. We adjust the 3dB bandwidth from 37.5 GHz to 100 GHz with a step of 12.5 GHz, with the insertion loss varying from -5.4 dB to -7.9 dB. Keywords: Integrated optics devices, wavelength filtering devices, coupled resonators

A novel photonics channelization spectrum stitching technique in a dual-OFC-based photonic channelizer for receiving a wide-band signal is proposed. By studying the slow and rapid variant characteristics of the channel differences separately, all the channels’ responses are precisely estimated and stitched in the frequency domain. A proof-of-concept experiment is performed and signals with 3 GHz bandwidth are recovered with an SNR loss of 1.01 dB.

A millimeter-wave optoelectronic oscillator employing self-regenerative frequency dividing and phase-locking techniques is proposed. The frequency division of millimeter-wave signal is achieved effectively via self-regenerative frequency divider breaking the frequency limitation of commercial frequency dividers. In virtue of the frequency conversion pair, the phase-locking technique is effectively utilized to stabilize the millimeter-wave optoelectronic oscillator by a commercial analog phase shifter in relative low frequency band. Finally, a 40-GHz millimeter-wave signal is generated with the single-sideband phase noise about -116 dBc/Hz at 10-kHz frequency offset. Besides, the frequency stability of the proposed millimeter-wave optoelectronic oscillator is greatly improved from 1.2×10^-6 to 2.96×10^-13 at 1024-s averaging time in a lab room without any thermal control.

Optical vortex beams carrying orbital angular momentum are being widely investigated for boosting the information capacity of communication systems by virtue of their unbounded state space for spatial mode division multiplexing or high-dimensional encoding. Vortex mode sorters are a critical component in such systems and a spiral transformation scheme working in the paraxial regime for vortex mode sorting has been proposed very recently to achieve high modal resolution with unity efficiency. Here we modify the spiral transformation scheme by developing a phase corrector in the nonparaxial regime, which is confirmed to have a more accurate phase-correcting function in implementing the spiral transformation and eventually lead to a better mode separation, especially in the case of a compact vortex mode sorter. Based on the above principle, a home-made compact vortex mode sorter is demonstrated by integrating the two phase elements implementing the spiral transformation onto the opposite sides of a thin quartz plate as flat diffractive optical elements, which achieves high-resolution and high-efficiency vortex mode sorting as expected. The modified optical spiral transformation and the compact design of vortex mode sorter based on this scheme therefore provide a much better tool for the effective mode separation, which can enable new applications in both classical and quantum information systems based on vortex modes.

We report a low coherence supercontinuum source with high-repetition rate and compare its spectral width and coherence under conditions of different pump pulse. The repetition rate of supercontinuum source is 4GHz and the spectrum width of supercontinuum is nearly 400nm. The supercontinuum spectral width and coherence under different pump pulse duration were compared, which indicates that with the same pump power, as the pump pulse duration increases, the width and coherence of the supercontinuum both decreases. We also compare the coherence of the supercontinuum with different pump power, as the pump power increases, the coherence of the supercontinuum deteriorates. The coherence of supercontinuum is related to pulse width and power of injected pulse.

We report an optical vector network analysis (OVNA) based on optical suppressed carrier double-sideband (DSB) modulation and the Pound Drever Hall (PDH) technique. In this novel scheme, the optical carrier suppressed DSB modulation signal propagates through the high Q optical device, and then the double frequency of the driven radio frequency signal is detected, by which the frequency responses of the device can be accurately achieved. Comparing with the common DSB-based OVNA, by biasing the modulator at the minimum transmission point (MITP), the accuracy improvement can be realized since the errors caused by the even-order sidebands are eliminated. Moreover, the high stability of the proposed OVNA can also be achieved by using the PDH technique. In the proof-of-concept experiment, the magnitude and phase responses of the Fabry-Perot (FP) interferometer are realized with high accuracy when the modulation index is small. There is no repeated frequency response even if the test time is up to 30 minutes. The proposed scheme provides a novel strategy for high-accuracy frequency responses measurement, which can be potentially used in high Q optical devices characterization.

We demonstrate a broadband grating couplers which can be fabricated on SOI by MPW process. By using chirped grating coupler, broad-bandwidth of 78 nm with 1 dB insertion and 130 nm with 3 dB insertion are achieved.

A remarkable self-sustained thermo-optomechanical oscillator has been observed in various optical micro-cavities, which caused by competition among the thermal expansion, the thermo-optic effect, and Kerr effects as we scan the probing laser across a cavity resonance at various tuning rates. Oscillation periods in the thermo-optomechanical oscillator are considered to be related to the heat dissipate rate from the micro-cavity resonator bulk to the environment, and it is possible that the thermo dissipate rate can be measured by detecting oscillation periods. Although a nonlinear relationship between the heat dissipate rate and oscillation periods is exhibited, an artificial neural network is applied to identify the heat dissipate rate. Numerical results demonstrate that the method can be used to measure the heat dissipate rate effectively in the thermo-optomechanical oscillator based on a CaF₂ whispering-gallery-mode resonator.

We propose a novel microwave photonic phase shifter (MWP-PS) architecture which is suitable for on-chip integration. The MWP-PS consists of a laser source, two identical phase modulators (PMs), a tunable optical coupler (TOC), an optical filter and a photodetector (PD). By introducing a microwave signal and its reversed counterpart to the two PMs, respectively, and using the optical filter to remove one of the generated sidebands, a phase shift with its value tuned by the coupling coefficient of the TOC will be introduced to the microwave signal after the PD. A proof-of-concept experiment is carried out. An MWP-PS with a 360° phase shift tuning range and an operational frequency range of 8-40 GHz is achieved.

We propose a novel filterless approach to generate high-quality optical frequency 16-tupling millimeter-wave based on only two MZMs. OSSR higher than 30.34 dB and RFSSR not less than 23.79 dB are achieved in this scheme.

We report a novel optical microcavity with micro-nano size based on the Fabry-Perot microcavity, which is a roof structure with two top reflectors that are symmetrical structure and proper inclined angle. The special design of the top reflectors increases the one-period-length for the mode light ray inside the microcavity and confines light in a smaller area effectively. The resonance principle of the microcavity is analyzed and the properties are numerically studied by the finite element method (FEM). Compared with the Fabry-Perot microcavity, the result show that the proposed microcavity has outstanding advantages such as high quality factor, narrow spectrum, and small effective module volume, which has a great potential in photonic devices and optoelectronic devices.

A multi-layer polarization rotator is proposed and demonstrated based on inverse design. Simulation results show that the device successfully rotates the polarization state of the input light from TE to TM with a coupling length of only 2.8 um, and the insertion loss and extinction ratio are 3.29 dB and 25 dB respectively at the wavelength of 1550 nm. Besides, the device has a fairly-wide operating bandwidth of 64 nm from 1506 nm to 1570 nm.

In this article, we propose a novel method using machine learning, especially for artificial neural networks (ANNs) to achieve variability analysis and performance optimization of the plasmonic refractive index sensor (RIS). A Fano resonance (FR) based RIS which consisted of two plasmonic waveguides end-coupled to each other by an asymmetrical square resonator is taken as an illustration to demonstrate the effectiveness of the ANNs. The results reveal that the ANNs can be used in fast and accurate variability analysis because the predicted transmission spectrums and transmittances generated by ANNs are approximate to the actual simulated results. In addition, the ANNs can effectively solve the performance optimization and inverse design problems for the RIS by predicting the structure parameters for RIS accurately. Obviously, our proposed method has potential applications in optical sensing, device design, optical interconnects and so on.

An optically tunable radio frequency (RF) downconversion scheme is proposed based on an optoelectronic oscillator (OEO) incorporating a tunable microwave photonic filter. The local oscillation (LO) is generated in the OEO, whose frequency is varied through simply tuning the frequency difference between the optical carrier and the reflection notch of a phase-shifted fiber Bragg grating (PS-FBG). The LO and the input RF signal are combined and added to the OEO loop by a single phase modulator. The RF modulation sidebands and one of the LO modulation sidebands are extracted out of the OEO loop by the PS-FBG and sent to a photo-detector to achieve RF downconversion. In the experiment, optically tunable LOs in the frequency range of 6 GHz to 15 GHz are generated, and RF signals in the frequency range of 7 GHz to 16 GHz are successfully down-converted to intermediate frequency band around 1 GHz. The proposed scheme has the potential to cover a frequency range beyond 40 GHz.

In this work, we propose a novel high-precision frequency estimation method based on harmonic expansion technique and a simple FFT-based algorithm. By increasing the information content through harmonic expansion in spectral domain instead of sampling time length in time domain, the proposed method can greatly improve the frequency estimation precision, needless to introduce other complex algorithms. The harmonic expansion process is to synthetize multiple harmonic components of the fundamental frequency of the input signal, which are detected to perform high-precision frequency estimation. The proposed method is analyzed in theory and numerical analysis, and demonstrated in experiment. The harmonic expansion in the experiment is achieved by microwave photonics technology through optical comb generation by electro-optical modulation. The signal optical comb containing wideband optical harmonic components are downconverted into low frequency band in electrical domain through optical harmonic sampling. Through digital signal processing on the 2th ~ 12th harmonic components with the FFT-based algorithm, the frequency estimation precision of a single RF tone is improved by about dozens of times as compared with the measurement value of the 1th fundamental frequency. This method is also compatible with other existing FFT-based high-precision frequency estimation algorithms and has the potential for a variety of application scenarios, such as Radar/LIDAR, spectrum sensing, vibration measurement and electronic reconnaissance.

In this paper, we present a numerically study of the nanoscale refractive index sensor based on periodically modulated graphene ribbon arrays with finite-difference time-domain method (FDTD). The results show that the resonance wavelengths of the graphene metamaterials structure have an approximate linear correlation with the refractive index of the dielectric substrate, which can be used as the refractive index sensor exhibits a higher sensitive of 3069.1 nm/RIU. Furthermore, the resonant wavelength of the structure tends to exhibit a clearly blue shift with the Fermi-level (E_f) of graphene ribbons increasing. And the shift of the resonant peak ups to 4906.3 nm per unit Fermi level. Besides, the influence of ribbon-width and Fermi level of the graphene on the performance of sensor are investigated in details. We believe that the work is useful for the design and application of the plasmonic refractive index sensors based on a grapheme ribbon array.

We propose a near half-wavelength one-dimensional (1-D) optical phased array (OPA) antenna based on a superlattice structure design approach which overcomes conventional crosstalk problems and offers high resolution broadband beam steering while preserving a small footprint size. The performance of the OPA have been optimized by investigating the parameters of the strip gratings, including grating depth, grating period and number of antennas. Results show that the proposed OPA can steer 130° in the longitudinal axis with a divergence beam width of 2.52° at the main lobe for 33 grating elements. This provides 52 resolution points which achieves a 44% improvement over recent 1-D superlattice gratings array antenna system.

A novel broadband microwave frequency divider enabled by an optoelectronic oscillator (OEO) is proposed, which features large bandwidth, low noise and compact structure. The Mach-Zehnder modulator in the optoelectronic oscillator operates at the minimum transmission point to perform the optical carrier-suppressed double sideband modulation and avoid the free-running oscillation of the OEO. In a proof-of-concept experiment, the proposed frequency divider successfully divides the microwave frequency from 12 GHz to 6 GHz.

A fiber-chip spot size converting edge coupler has been presented on silicon-on-insulator (SOI) platform. The design is optimized for the wavelength of 1.55 μm using a fully-vectorial three-dimensional eigenmode expansion method. The proposed coupler combines an inverse taper edge coupler and a mode expander on a 220-nm-thick SOI platform. Results show that our proposed coupler could achieve an output profile of 10.4 μm×220 nm and 90% coupling efficiency operating for both transverse-electric and transverse-magnetic polarizations.

We present an integrated continuously tunable microwave photonic phase shifter based on the strain effects in silicon-oninsulator microring resonators, which eliminates the use of microheaters in conventional phase shifter tuning schemes. We demonstrate the tunability of the radiofrequency phase shift by mechanically inducing a strain-dependent variation in the ring resonator circumference, which causes a subsequent shift in the resonant wavelength and optical phase of the resonator. By applying a small strain, large changes in the radiofrequency phase shifting is demonstrated. As an application example, the proposed strain based microwave photonic phase shifter reveals a phase-shifting of a full range.

A photonic approach for simultaneous frequency down-conversion, self-interference cancellation, and image rejection for in-band full-duplex radio-over-fiber systems is proposed based on a dual-polarization quadrature phase-shift keying (DP-QPSK) modulator. The upper dual-parallel Mach–Zehnder modulator (DP-MZM) of the DP-QPSK modulator realizes the radio frequency (RF) self-interference cancellation and generates the upper and lower first-order optical sidebands of the desired RF signal, whereas the lower DP-MZM generates two optical sidebands of the local oscillator. The optical signals from the two DP-MZMs are combined, and the upper and lower optical sidebands are separated by a wavelength division multiplexer (WDM). The two outputs from the WDM are detected at two photodetectors for frequency down-conversion. By combining the two intermediate-frequency (IF) signals at a 90°hybrid coupler, the proposed system can also achieve image rejection. In addition, by changing the bias point of the lower DP-MZM, second-harmonic frequency down-conversion can be implemented. A simulation is performed. A QPSK-modulated RF signal from 5 to 44 GHz is successfully down-converted to 1-GHz IF signal with self-interference and image frequency cancelled, and the EVM of the received IF signal is lower than 15%.

A new window of optical communication is emerging around 2 μm. It is important to design and experimentally demonstrate the photonic devices and components that can support the optical communication in this spectral region by providing the functionalities of switching and routing. The silicon photonics platform for realizing the photonic devices and components will be preferred around 2 μm, like other optical communication windows of 1310 nm and 1550 nm, due to availability of cost effective and high yield CMOS fabrication technology. Photonic switches that are non-volatile in nature and consume lesser power while having ultra-low footprint are likely to be in great demand for future optical communication around 2 μm. Here, we report an ultra-compact 1×1 photonic switch operating at 2.1 μm using nonvolatile phase change material Ge₂Sb₂Te₅ embedded in silicon-on-insulator platform. Embedding of Ge₂Sb₂Te₅in silicon-on-insulator waveguide is done in two different ways to evaluate and compare the switching performance. The emphasis has been on optimization of position and dimensions of Ge₂Sb₂Te₅ in partially and fully etched silicon waveguide. We obtained an extinction ratio of 34.04 dB with low insertion loss of 0.49 dB in ON state with Ge₂Sb₂Te₅ of volume 920 nm× 240 nm × 800 nm (length × height × width) embedded into partially etched silicon waveguide. When Ge₂Sb₂Te₅ is embedded in fully etched silicon waveguide, maximum extinction ratio of ~14dB at the expense of insertion loss of 1.36 dB with Ge₂Sb₂Te₅ of volume 1020 nm× 240 nm × 800 nm.

Optical frequency combs (OFCs) have discrete optical spectra with equidistant spacing and equal intensities over a broad spectral range. In recent year, OFCs have been widely researched in many areas. However, the traditional methods of generating optical frequency comb are suffered from the long-term instability. In this article, a feedback system of the stability improvement of optical frequency comb (OFC) generating system based on a single integrated polarization multiplexing dual-drive Mach-Zehnder modulator (PM-DMZM) is proposed. The effect of the fluctuation of the halfwave voltage on the amplitude of the OFC spectra is analyzed. The proportion between the optical power of OFC and the optical carrier power of OFC is used as a feedback parameter to guarantee the simplicity and effectiveness. The stability of the proposed OFC system based on PM-DMZM is also analyzed by the numerical simulation. The results show that the flatness of 7-line OFC is improved for about 3 dB when only the bias voltage is regarded as the controlled parameter. In addition, the flatness of OFC can be improved for about 10 dB when the bias voltage and the amplitude of the modulated RF signal are all used as the controlled parameters simultaneously.

We numerically present the generation of five types of pulse states using nonlinear Schrödinger equations in an all-normal-dispersion (ANDi) mode-locked fiber laser based on a nonlinear amplifying loop mirror (NALM). Using a two-dimensional pump power space of the two amplifiers, we investigate the pulse characteristics among typical dissipative soliton resonance (DSR), dissipative soliton (DS), transition state (TS) between DS and DSR, noise-like pulse (NLP) and unstable pulsation state. Different from previous research results, we firstly find that asymmetric coupling ratio of the NALM causes dependent and non- monotonous variation of the pulse peak power and width of DSR. The location of incident power and the saturable power of NALM (when the former is 2 times higher, higher, equal, lower than the later) determine the pulse states (NLP, DSR, unstable, DS or TS) in the cavity. The pump-power intercoupling-induced frequency-selective NALM transmission is demonstrated by the simulated pulse evolution over round-trips.

We present thermally tunable silicon coupled resonators optical waveguide (CROW) with an ultralow tuning power of 1.26mW per free spectral range by exploiting thermal isolation trenches close to the ring waveguides. When the micro-rings space is 300μm, the thermal crosstalk is 1.05 pm.

In this paper, we studied the silicon nitride waveguide phase modulator using graphene as a heating electrode. We demonstrate the advantages of graphene heater waveguide phase shifter by comparative simulation, The effect of the length and the width of the heating electrode on the thermal phase modulating efficiency was also been studied. Finally, by covering the graphene heating electrode directly on the waveguide, we obtained a low-power thermal-adjustable phase shifter with a π-phase shift modulation power of 96.8mW.

Owing to the promising application prospect of mode division multiplexing technology, reconfigurable optical mode multiplexers/demultiplexers, which enable convenient construction of flexible and complex on-chip optical networks, have attracted increasing attention in the academic community. Here, we propose and experimentally demonstrate a reconfigurable two-mode multiplexer/demultiplexer based on novel three-waveguide-coupling structures. The fabricated device can multiplex two input signals to fundamental and first-order quasi-transverse electric modes with large operation bandwidth. Static response spectra show that the optical crosstalk at the output ports of the device are less than -14.3 dB and -13.7 dB over the entire C band (> 40 nm), respectively. The highspeed data transmission ability of 40 Gbps for each multiplexing channel are also demonstrated successfully. The presented device is believed to be a potential candidate for future on-chip optical network with large-scale integration, flexible functionality, and low cost.

We propose and demonstrate a reconfigurable data exchange device for wavelength-division multiplexing (WDM) and mode-division multiplexing (MDM) compatible optical networks, which is capable of exchanging data information between a fundamental mode (TE₀) signal and an arbitrary high-order mode (TE_x) signal with the same wavelength. We experimentally demonstrate exchange for a TE₀ mode signal and a TE₁ mode signal. The insertion loss is about 7.5 dB including terminal coupling loss, the extinction ratio is more than 20 dB within the C band. The crosstalk for the TE₁-to- TE₀ and TE₀-to-TE₁ mode conversions is measured to be less than -22 dB and -16 dB, respectively. Clear and open eye diagrams of two different wavelength signals at 10 Gb/s are obtained. The proposed scheme is expected to be used for on-chip WDM-MDM compatible networks in future due to its scalability, compact size and reconfigurable functionality.

Weak fiber Bragg grating array has been theoretically analyzed by the transfer matrix method and coupled-mode theory. Fig.1 gives the analysis diagram of weak fiber Bragg grating array Through the mathematical operation, the simulation model of the Weak Fiber Bragg Grating Array is obtained. Numerical predictions of the effects of distributed weak Bragg gratings on the reflection spectra of such a grating are calculated. This fiber with several random distributed weak Bragg gratings has been fabricated. In this paper, the number of peaks in the reflection spectrum increases significantly as the number of gratings increases, and the intensity of reflection increases.These predictions are compared with experimentally measured spectra of the random distributed weak Bragg gratings, the reflection spectra of the weak grating array are measured with an AQ6370 optical spectrum analyzer with a 0.02 nm resolution, A good agreement between the theoretical predictions and the experimental results was obtained. From comparing the experimental results to the simulation results, we figured that transmission peaks have good agreements; the central wavelength of reflection spectra of the weak Bragg grating array about simulation result is 1551.05 nm. We found that when the position and size are changed simultaneously, the reflection spectra of the weak grating Bragg array can be predicted. the number of gratings has a significant effect on the number of reflection peaks, and the larger the number of gratings, the greater the number of reflection peaks. We think it maybe because the light propagates back and forth in the grating and so on.

Ultrafast fiber laser has been widely used for spreading the extensive industrial applications and exploring the optics nonlinear dynamics. Here, we report a bidirectional fiber laser passively mode-locked by nonlinear polarization rotation (NPR) technique, supporting the emission of multi-state solitons. For the first time to the best of our knowledge, a Φ-shape auxiliary cavity based on chirped fiber Bragg grating (CFBG) is proposed to simultaneously introduce large anomalous and normal dispersion into the lasing oscillator. As a result, different dispersion distributions are achieved intra-cavity, respectively corresponding to counter propagating directions. Thus, conventional soliton (CS) and dissipative soliton (DS) are respectively generated in clockwise and counter-clockwise propagating directions. The results could benefit multifunctional ultrafast fiber laser system, which is potentially set for many practical applications as well as the study of soliton dynamics.

With the development communication, polarization filter, as the key components of communication, becomes more and more important. For satisfying various requirements, the paper studies different photonics crystal fiber (PCF) polarizing devices based surface plasmon resonance (SPR). Besides, we have also demonstrated their excellent polarization performance from two aspects of theory and experiment. On one hand, the paper theoretically proposed dual-wavelength single polarizing filter with the broadband of low crosstalk. It could have single polarizing filtering in the dual-wavelength of 1310 and 1550 nm and the polarization direction of optical output is independent of the wavelength of incident light. The polarization direction of emergent light through filter can always keep the core mode in x-polarized direction (x-PCM) in a wide range of wavelength, which increases the difference of confinement loss of core mode in y-polarized direction (y-PCM) and x-PCM, and reduces mutual crosstalk. On the other hand, considering the new communication waveband emerging, we present a V-type birefringence microstructure optical fiber with embedded dual copper wires that were made by the stack-drawing method and the dual copper wires were aligned in the vertical direction symmetrically. Experimental results showed that the fiber can be easily excited strong SPR in the S and L waveband. Then the core mode is completely coupled into the copper wires and generates the SPR in the y-polarized direction, to achieve the purpose of directional filtering.

Decomposing the tangential force which impacts on the fiber into radial forces and utilizing optical fiber line force model, one novel temperature insensitivity fiber Bragg grating (FBG) torsion measurement model is proposed in this paper. Basic principle of this torsion measurement model is that center wavelength shift of FBG is analyzed to obtain the torsion angler of optical fiber. In order to get the angle of torsion per unit length resolution of this measurement model, simulation experiment is carried out. Results of the simulation experiment show that torsion angle resolution of this measurement model changes from -1.76·10^-3 rad to 2.36·10^-2 rad with detection range ±0.5rad. The demand of the torsion measurement model is just one embedded FBG which decreased the torsion detection expense greatly.

Optical fiber is sensitivity to strain and bend change, so it could be used as strain sensor, curvature sensor and even weight sensor. An optical fiber weight sensor based on single-mode-multimode-single-mode (SMS) fiber structure is presented in this paper. The sensing structure is formed by cascading one segment of multimode fiber (MMF) with two segments of single mode fiber (SMF), namely SMS structure. Modal interference would occur in MMF among multiple core modes. When the axial strain and curvature on the sensing fiber structure is changed by the weight hanging at the end of the cantilever beam, the refractive index and length all change. In experiment, the weight is increased via adding different mass of weights. Afterwards, the phase difference among different core modes would be changed. Therefore, the output interference spectrum would shift accordingly. Plot the weight and wavelength shift in one graph and then it could be shown that in the range of 0 to 55g, there are three peaks shifting linearly. The weight sensitivity is up to -11.3 pm/g. This weight sensor has the advantages of low cost, simple fabrication process, electromagnetic immunity and capability of remote operation.

The hollow-core optical fiber gas laser (HOFGLAS) is emerging as a promising technology in the generation of mid-infrared radiation, which combines the advantages of both fiber and molecular gas lasers. Whilst the experimental study is of rapid development, the mechanism of lasing and explanations on emission power characteristics are yet to be discovered further. We introduce a typical four-energy level system of acetylene molecules and hence build up a rate-equation model to simulate the output power of different transitions and their relative strengths. The numerical values of slope efficiency and power ratio coincide with prior laboratory measurements. The pump threshold is numerically estimated as well.

A method of dual-parameter control chaos in a dual-ring erbium-doped fiber laser is studied via stimulating dual-pump. And a chaos-control physical model is presented via considering the dual-parameter control technology and the laser structure characteristics. Dual-signal of periodic mutation as control signal is used to modulate to stimulate the two-pump to obtain chaos-control or pressure chaotic oscillation using same amplitude, same frequency, different amplitude or different frequency. A chaotic dual-ring erbium-doped fiber laser can be controlled to single-cycle, two-cycle, three-cycle, four-cycle and other multi-cycle. Low-cycle orbits are controlled to show in laser under the case of same amplitude and same frequency. High-cycle orbits are controlled to show in laser under the case of different amplitude or different frequency. Multi-cycle belt under the case of same amplitude and different frequency, controlled-frequency-locking region under the case of same amplitude and low frequency and controlled-semi-frequency-locking region under the case of same amplitude and high frequency are found. The results are helpful to the research of laser chaos and application of fiber laser.

In this paper, we use a pulse trigger to manipulate soliton-fission in supercontinuum generation process by modulation instability (MI) in ~picosecond pump scheme and we demonstrate the prior performance of generated supercontinuum compared to non-trigger situation. A distributed feedback (DFB) laser is modulated into a pulse source with a high repetition rate of 4GHz by pulse amplitude modulation (PAM) and an adiabatic soliton compression (ASC) method is used to compress pulse width from 8.2ps to 3.3ps. It is observed that threshold of SC generation is relatively lower when the trigger is used. As pump light coupled with pulse trigger launch into a 400m high non-linear dispersion-shifted fiber (HNL-DSF), the generated supercontinuum is nearly 100nm wider in redshifted side and exhibit better stability than untriggered one.

In this paper, by shortening the cavity length, we have realized all-fiber Tm-doped fiber laser with pulse width less than 600 fs based on nonlinear polarization rotation effect. Under the condition of pump power of 0.281 W, the traditional soliton limit pulse is realized by adjusting the polarization controllers. The pulse width is 560fs and consistent with the theoretical calculation of the limit pulse width 569fs, repetition frequency is 55.25 MHz, the central wavelength is 1978.34 nm and the 3dB spectral width is 7.25 nm. The traditional soliton index is helpful to improve the application of ultrafast Tm-doped fiber laser in mid-infrared light source and medical operation.

50 Gbps PAM4 transmissions at 850 nm and 908 nm, and 100 Gbps SWDM4 transmissions over OM3/OM4/OM5 fibers were carried out. OM5 fiber can support 50 Gbps PAM4 transmission links of 500 m at 850 nm and 300 m at 908 nm. Another OM5 fiber can support an over 400 m link of 100 Gbps SWDM4 transmission, while the right-tilted OM3/OM4 fibers can only support 100 m and 150 m links, respectively, with similar EMB at 850 nm as the OM5 fiber.

A photonic-assisted dual-band coherent radar transmitter system with a large frequency tunable range is proposed and demonstrated. This dual-band transmitter is composed of a triple-loop optoelectronic oscillator (OEO) link, a low frequency band subsystem (LFBS) and a high frequency band subsystem (HFBS). The triple-loop OEO link is developed for the generation of an ultralow phase noise microwave signal with a large tunable range, microwave photonic down-converting is used in the LFBS to change band range, microwave photonic frequency multiplying is applied in the HFBS to achieve the bandwidth extension. The band ranges of the proposed dual-band transmitter can cover from S to Ka six bands in all. Performances in the time and the frequency domains of the dual-band microwave signals are also investigated.

Ytterbium doped microstructured optical fibers have many advantages for fiber amplifiers. Due to air-clad structures, high-power microstructured optical fiber laser systems require stricter end-facet treatment than that of conventional fiber laser systems. Here, we fabricated microlenses in ytterbium doped double-cladding microstructured optical fibers by a CO₂ laser. The surface power density of the MOF tip can be reduced by 2 quantity levels theoretically. The theoretical and experimental investigation is carried out to study effect of minor and major axis semidiameter of ellipsoidal microlenses on coupling efficiency. The experimental coupling efficiency was over 88%. Results indicated that this microlens combines the end cap scheme and pump coupling. The proposed fabrication method is helpful for high mechanical stability, well reproducibility and miniaturization of the high-power rare earth-doped microstructured optical fiber laser systems.

In order to monitor the motor vehicle exhaust effectively, a motor vehicle exhaust telemetry device is set up based on differential optical absorption spectroscopy (DOAS) in this paper. The absorption spectrum of NO and 1,3-butadiene are analyzed quickly, meanwhile, qualitative and quantitative analysis are made simultaneously by the Least Square Method. Moreover, the correlation coefficient of absorbance and concentration reaching more than 0.999, which is obtained by non-linear analysis through polynomial fitting method. Additionally, the minimum detection limits of NO and 1,3-butadiene are 3ppm and 0.5ppm respectively. In order to verify the performance of the system, a set of tests were performed for a long time. So the experimental results show that the relative error of NO is not more than 2%, and the relative error of 1,3-butadiene is less than 2%. In addition, the calculation results of the experiment demonstrate that the repeatability error of NO is better than 1%, and the repeatability errors of 1,3-butadiene is lower than 2%. All the results meet the requirement of the application.

Fiber bragg grating has become one of the widely used optical measurement technologies in the field of aviation structure monitoring because of its light weight, high sensitivity and immunity to electromagnetic interference. A strain transfer model is established for the surface attached fiber grating, the function between real strain and sensing strain is obtained. Basing on finite element simulation, strain field distribution of the sensing structure is analyzed and the distribution curve of strain transform rate is drawn. The strain calibration system of fiber grating is built to verify the accuracy of the model. The average error of this model is superior to 5.5% in the measurement range of 0-3000με. The study improves the measurement precision of surface bonded fiber grating sensor, which also provides academic bases and references for the aircraft structural health monitoring.

A 100 watt and ytterbium doped fiber laser based on MOPA structure is reported in this paper. The laser uses a modulated semiconductor laser as the seed source with pulse width of 10-350 ns and repetition frequency continuously adjustable in the range of 90 kHz-1100 kHz. Using 30/250μm large mode area fiber, the stable nanosecond pulse width is obtained by setting the optimum fiber length of 6 m. The second stage main amplifier uses 6 30W pumps, and the reverse-pump mode. When the repetition rate is 90 K Hz, the average power output of 100 W is obtained by the main power amplifier (MOPA). At the highest output power, due to gain shaping mechanism, the pulse width is reduced from 375 ns to 350 ns. The corresponding peak power is 10 kW, the single pulse energy is 1.11 mJ, and the output signal-to-noise ratio reaches 36 dB at the highest output power. And at the moment, the SRS effect happens just at the center wavelength 60NM. In this paper, a tunable 100-watt nanosecond fiber laser with two-stage amplification based on semiconductor modulation technology is presented. A compact 100W prototype has been completed.

Compared with the traditional electromagnetic displacement sensor, the optical sensor has the advantages of reduced size and anti-electromagnetic interference. The increasing demand for high-resolution displacement sensors in today's production and research environments has led to the rapid development of variable spacing grating displacement sensors. The principle of variable spacing grating displacement sensor is verified by experiments, and the reliability of the sensor is proved. Based on the theoretical analysis of the linear density formula of variable spacing grating, we proposes a new displacement sensor optimization method. The method is demonstrated by a series of experiments. The main component of the displacement sensor adopts a variable spacing blazed grating in order to concentrate on the first energy level by the energy of the grating diffraction to improve diffraction efficiency. This paper analyzes the correlation between the resolution of the displacement sensor and the resolution of the grating and the resolution of the spectrometer and by experiment we further prove that the smaller the grating spacing is, the stronger the resolution of the grating. Theoretical analysis shows that by increasing the grating spacing variation, the linear density formulas of the variable spacing gratings can be optimized to achieve a 50mm range 0.02mm resolution

Weakly-coupled mode division multiplexing (MDM) over few-mode fiber (FMF) for short-reach transmission has attracted great interest, which can avoid multiple-input-multiple-output digital signal processing (MIMO-DSP) by greatly suppressing modal crosstalk. In this paper, step-index FMF supporting 4 linearity polarization (LP) modes for MIMO-free transmission is designed and fabricated for the first time. Modal crosstalk of the fiber is suppressed by increasing the mode effective refractive index differences. The same fabrication method as standard single-mode fiber is adopted so that it is practical and cost-effective. The mode multiplexer/demultiplexers (MUX/DEMUX) consists of cascaded mode-selective couplers (MSCs), which are designed and fabricated by tapering the proposed FMF with single-mode fiber (SMF). The mode MUX and DEMUX achieve very low modal crosstalk not only for the multiplexing/demultiplexing but also for the coupling to/from the FMF. Based on the fabricated FMF and mode MUX/DEMUX, we successfully demonstrate the first simultaneous 4-modes (LP₀₁, LP₁₁, LP₂₁ & LP₃₁) 10-km FMF transmission with 10-Gb/s intensity modulation and MIMO-free direct detection (IM/DD).

We propose and experimentally demonstrate a technique using a composite-double-probe-pulse (CDPP) to eliminate the effect of polarization fading for phase-sensitive optical time-domain reflectometry (Φ-OTDR) based on ultra-weak FBG (UWFBG) array. The CDPP is composed of two optical pulses whose spatial interval is equal to twice the spatial interval of adjacent UWFBGs in the UWFBG array. One optical pulse is a long optical pulse, and the other optical pulse is composed of two continuous short optical pulses, whose polarization states are orthogonal to each other. The width of the short pulse is equal to half of the width of the normal pulse and their frequencies are different from the long pulse. By using such a method to perform the sensing for the UWFBG array, distributed quantitative measurement can be realized with only direct detection scheme and the influence of polarization fading in the demodulation of signal is thoroughly eliminated.

In this paper we report the development and field test of a high sensitivity 3C fiber laser geophone for seismic acquisition application. When the fiber laser geophone senses the seismic signal, its output wavelength is proportionate to the seismic signal. By detecting the wavelength shifts using interferometric demodulation method, the micro-seismic signal can be detected. We are presenting field test results for the 3C fiber laser geophone array system and comparing its performance with regular exploration geophones. The detailed information of the acquired seismic signals, e.g. waveform, frequency spectrum, and wavelet are analyzed for assessing their performance. The 3C fiber laser geophone has advantages of wide bandwidth and good high-frequency response.

We demonstrated tunable Q-switched and mode-locked Ho3+/Pr3+-codoped ZBLAN fiber laser in mid-infrared region using single-walled carbon nanotubes. Both stable Q-switched pulses from 2837.6 to 2892.6 nm and modelocked pulses from 2836.2 to 2872.6 nm were achieved.

In our paper, we selected several commercial optical transceivers, which consist of single-channel transceiver modules, parallel transmitting and receiving modules, and EPON OLT and ONU modules, to do the displacement damage (DDD) testing via 10MeV proton radiation method. The changing of current and receiver sensitivity of optical transceivers is discussed and analyzed. Based on the DDD testing exposed to total of 10¹¹ 10MeV protons at a dose rate of about 10⁷ protons/cm²/s, the performance of transceivers are discussed and analyzed, and all the modules can be worked well after annealing. Finally, based on the theoretical analyzed and testing results, this paper provides several design suggestions to improve the reliability for optical transceivers, which can be referenced by satellite system designation for various space missions.