In order to improve the transmission rate and spectrum utilization of visible light communication (VLC)systems, a field programmable gate array(FPGA)based on faster-than-Nyquist(FTN)-VLC system is designed. The system mainly consists of FPGA, digital-to-analog(D/A)converter, low-pass filter, avalanche photodiode, current-to-voltage(I/V)converter, and voltage comparator. Both software and hardware of the system are simulated and tested. The simulation and test results indicate that when the transmission distance is 0.5 m, the data transmission rate of the system can reach 10 Mb/s, and the spectrum utilization is improved by 18.25% compared to the Nyquist optical communication system under the same conditions.

Aiming at the problem that the fixed threshold decision scheme leads to the deterioration of the performance of the underwater wireless optical communication system in the turbulent channel environment, a dynamic threshold decision scheme is proposed. Based on the slow fading characteristic of the underwater turbulent channel, the data is sent by frame at the sending end, and the decision threshold is updated every frame at the receiving end according to the amplitude of the leading code of the receiving frame. This paper uses the Monte Carlo method and the Gamma-Gamma turbulence channel model to simulate the underwater wireless optical communication system. The simulation results show that in the case of weak turbulence channel environment and when the system bit error rate is 10-6, the dynamic threshold decision scheme improves the bit error performance by 5.4 dB compared with the fixed threshold decision scheme. In the case of medium and strong turbulence channel environment,the dynamic threshold decision scheme can effectively suppress the occurrence of bit error rate threshold phenomenon.

Due to the strong time-varying nature of the ultraviolet communication network channel, corresponding adaptive routing is needed to solve the problems of high network transmission delay and uneven energy consumption of nodes dhring the net-working process. Based on ant colony optimization(ACO) algorithm, an improved ACO algorithm for ultraviolet commmnication networks is proposed. The algorithm introduces the network node energy into the state transition probability formula of ACO algorithm, and analyzes the delay performance of the algorithm under different transmitting angles, transmitting power and data transmission rates by Matlab simulation software. The simmlation results show that compared with the ACO algorithm, when the transmit-receive elevation is 50o, the delay of the proposed algorithm is reduced by 1 s, the convergence speed is increased by 28%, the average residual energy of the convergence path is also significantly increased, and the life cycle of the network is effectively extended.

In order to seek the optimal light source optimization scheme in different indoor environments, a deep learning based indoor visible light comminication light source configuration strategy is proposed. Introducing indoor obstacles, natural light, and human movement interference separately, with indoor signal-to-noise ratio uniformity as the objective function, using the flower pollination algorithm (FPA) to optimize the light source power and half power angle under different room conditions. Using the obtained room state and light source configuration as the training set, a convolutional neural network (CNN) is used for training.The resulting computational model can predict the optimal light source settings in different room states, achieving dynamic adjustment of light source configuration. The simulation results show that the qualification rate of the light source parameters of this strategy reaches 88%, and the predicted results meet the requirements in terms of room signal power and lighting intensity.

In order to further reduce the bit error rate of mlti-user mlti-cell visible light comnnication systems using non-or-thogonal multiple access/orthogonal frequency division multiple access(NOMA/OFDMA) technology, a joint power and subcarrier allocation algorithm is proposed. This algorithm aims to minimize the system's symbol error rate by achieving the joint optimal allocation of carrier and user power across sub-cells. Firstly, the symbol error rate expression of the system is theoretically derived. Then, Matlab software is used to simulate the visible light communication system model, and the optimal allocation results are obtained through genetic algorithms. Finally, the proposed algorithm is compared with existing algorithms. The simulation results show that the symbol error rate performance of the proposed algorith is superior.

Aiming at the problem of low positioning accuracy and poor stability caused by the environmental interference of the positioning technology based on received signal strength (RSS), an radial basis function (RBF) neural network indoor visible light positioning algorithm based on multi-source information fusion is proposed. By fusing the color moment feature of the image with the RSS moment feature, a fingerprint database is constructed, and the RBF neural network is used for prediction to achieve complementary advantages between the image and RSS. Finally, the positioning algorithm is verified. The experimental results show that the optimized multi-source information fusion positioning algorithm improves the positioning accuracy by 9.4% compared with the single RSS positioning algorithm.

Due to the large number of hyperparameters, it is difficult to obtain the optimal system model for the bidirectional long short-term memory (Bi-LSTM) neural network. At the same time, considering the possibility of premature convergence in the Grey Wolf optimizer (GWO) algorithm, a single-lamp localization method using the GWO combined with particle swarm optimization (GWO-PSO) algorithm to optimize the Bi-LSTM neural network is proposed. By optimizing hyperparameters such as the learning rate and the number of hidden neurons in the network, the stability and positioning accuracy of the system are improved. Finally, the weighted K-nearest neighbors (WKNN) algorithm is used to optimize points with large errors to obtain more accurate positioning locations. The simulation results show that in an indoor environment of 3 mx3.6 mx3 m, the average positioning error of the proposed localization method is 3.57 cm, with 90% of the positioning errors within 6 cm.

Aiming at the the problem of complex deployment, poor robustness, and low positioning accuracy in visible light indoor positioning systems based on received signal strength (RSS), a visible light indoor positioning perception model based on convolutional neural network(CNN) visual imaging is proposed, and the layout method of light sources is studied. Firstly, a visible light visual imaging positioning perception model is designed and built based on ambient light and ordinary light-emitting diode(LED) light sources. Then, the deep features of the image are extracted through the pre-trained CNN model. Based on this,the positioning accuracy model of the indoor positioning perception model based on CNN visual imaging is optimized by studying the relationship between positioning accuracy and the number of light sources, as well as the spacing between light sources indifferent light source layout models. The experimental results show that compared with the indoor positioning perception model based on RSS, when the positioning errors are less than 2.1 cm and 3.9 cm respectively, the confidence probabilities of the proposed model are increased by 10% and 6.7% respectively. At the same time, compared with the rectangular layout method and the triangular layout method, the positioning accuracy of the cross layout method is improved by 9.5% and 16% respectively.

Indoor visible light positioning (VIP) is a research hotspot in the field of indoor positioning. In recent years, more and more artificial intelligence methods are used to optimize and improve the positioning accuracy and efficiency of VIP. In order to show the latest research progress of VIP intelligent optimization methods, the relevant concepts and application scenarios of VIP are first summarized, and VIP applications under three scenarios of severe electromagnetic, high capacity of hot spots and blocked wireless signal transmission are introduced in detail. The adaptability of VIP positioning methods is compared with that of traditional indoor positioning (IP). In this paper, the VIP intelligent optimization methods proposed in recent years are reviewed in detail. Finally, the general problems and challenges in this field are summarized.

Aiming at the problems of high computational complexity and easy overlap of adjacent constellation points in multi-input multi-output visible light communication systems, a superposed probabilistically shaped (PS) 20-level quadrature amplitude modulation(20QAM) constellation scheme based on nonlinear coding is proposed. Firstly, based on the proposed nonlinear coding equation, a 4-level signal with equal probability is encoded into a PS 5-level signal. Then, the PS 5-level signal is mapped into a trapezoid-shaped PS 5QAM signal to maximize the Euclidean distance of the superposed constellation. Finally, the PS 5QAM signal is superposed with an equal-probability 4QAM signal to obtain a PS 20QAM signal. The experimental results prove that the proposed superposed 20QAM constellation scheme has strong robustness to nonlinear distortion of the system and better bit error rate performance than superposed 16QAM and superposed 24QAM constellation schemes.

In order to solve the problems of low reliability and user fairness in indoor non-orthogonal multiple access (NOMA) visible light communication(VLC) channel estimation methods, a method of using the Gray Wolf optimizer(GWO) algorithm to compensate and restore signals in deep neural network (DNN) in NOMA-VLC systems, that is GWO-DNN channel estimation method. This method introduces nonlinear convergence factors and two-dimensional chaotic mapping, effectively improving system reliability and multi-user transmission fairness. Experimental results show that, at a bit error rate (BER) of 104, the proposed method can achieve a maximum signal-to-noise ratio (SNR) gain of 4.3 dB compared to the minimum mean square error (MMSE) method, and the SNR difference between the two users is reduced from 3.2 dB to 0.9 dB. Under different orders of quadrature amplitude modulation, the performance of the proposed method is better than the MMSE method, and the higher the modulation order, the more significant the performance improvement.

Addressing the shortcomings of traditional multi-core fiber beam splitters and multi-core fiber connections, such as dif-ficulties in disassembly and the need for additional test beam splitters in testing solutions, a four-core fiber channel beam splitter is prepared using a double-clad fiber bundle tapering method. To facilitate device testing and connection, LC/PC-type active connectors are prepared at the ends of the single-mode fiber and the four-core fiber, thus forming a single modular four-core fiber channel beam splitter. Meanwhile, a mirror-based testing method is proposed for this modular device. This testing method requires only one LC/PC active connector-type mirror to easily complete the testing of the device's insertion loss and inter-core crosstalk. The test results indicate that the prepared four-core fiber channel beam splitter exhibits excellent performance, with a maximum insertion loss of 0.84 dB and a maximum inter-core crosstalk of -46.32 dB.

Aiming at the problem of anomaly detection of white-box communication network equipment, a white-box optical network fault detection scheme based on blockchain is proposed. Firstly, a data collection oriented blockchain architecture and a data collection and storage scheme based on blockchain are designed. Then, a data anomaly detection algorithm based on autoencoder is designed. Finally, the autoencoder model is deployed to the blockchain consensus device to form a complete blockchain-based anomaly detection mechanism. The experimental results show that the proposed scheme can quickly realize the on-link storage of data collected by different devices, accurately and quickly identify the abnormal data of devices, and realize the real-time anomaly detection of network devices.

A fiber optic grating high-frequency seismic detector based on a bearing type lever mechanism is designed to address the relatively weak research on high-frequency signal waves at present. With the lever mechanism and bearing fulcrum as the innovation point, the static structure analysis, modal analysis and harmonic response analysis of the detector model are carried out by using simulation software. The simulation results show that the simulation sensitivity of the geophone is 64.588 pm/g, and the simulation natural frequency is 1 638.5 Hz. By adjusting the length ratio between the power arm and the resistance arm of the lever amplification mechanism and the cutting radius of the cycloidal hinge, the simulation natural frequency of the detector is 1 204.4 Hz and the simulation sensitivity is 75.56 pm /g.

In order to effectively remove noise from the signal, a denoising method for submarine cable vibration signals based on variational mode decomposition (VMD) and permutation entropy (PE) is proposed. This method first uses VMD to decompose the signal into several intrinsic mode functions from high frequency to low frequency, and then calculates the PE values of each mode component to quantitatively characterize its degree of randomness. Next, wavelet thresholding is applied to the intrinsic mode function (IMF) components with more noisy information, and the denoised submarine cable vibration signal is obtained through signal reconstruction. In order to verify the effectiveness of this method, a simulation experimental system for monitoring the vibration signals of submarine cables based on Brillouin optical time-domain analysis(BOTDA) is established. The vibration signals of submarine cables under three working conditions of erosion, friction, and anchoring are obtained through experiments.The experimental results show that the proposed method can effectively remove high-frequency noise in submarine cable vibration signals. The signal-to-noise ratios of the three types of signals, namely erosion, friction, and anchoring, are increased to 30.257 8 dB, 29.635 4 dB, and 38.786 2 dB, respectively, with a low standard deviation.

Aiming at the problems of slow training speed and weak generalization ability which leads to low positioning accuracy in visible light indoor positioning technology based on neural network, a visible light positioning method based on beetle antennae search(BAS) algorithm is proposed to optimize neural network. A measured model of 0.8 m×0.8 m×0.8 m is built. In this method, the connection weight matrix of the neural network is optimized by using the BAS algorithm and the indoor wireless channel parameters are fitted to achieve indoor location. The simulation results show that the average positioning error of the simulation method is less than 3.42 cm, which is 40% higher than the neural network training speed. In the experiment, the average error measured in the plane of height h=0.25 m is less than 4.0 cm.

In order to improve the transmission performance of visible light communication link, a hybrid modulation system of carrier free amplitude phase(CAP) and amplitude differential pulse position modulation(ADPPM) is proposed. The system combines spectrum efficient multidimensional CAP modulation with power efficient ADPPM, and uses the time slot pulse of ADPPM to carry the CAP signal for information transmission. Under the Gaussian white noise channel model, the analytical expression of frame error rate of CAP-ADPPM hybrid modulation system is derived and verified by simulation. The influence of the order and amplitude of the hybrid modulation system on the frame error performance of the system is analyzed. The experimental results show that when the spectral efficiency is the same, the reliability of CAP-ADPPM hybrid modulation system is better than that of traditional single modulation system, and can provide a variety of parameter selection schemes, which provides a more flexible design choice for visible light communication.

In order to further reduce the volume of the transmitter and receiver of the system, realize high-speed communication, and improve the anti-interference ability of the system, a set of visible light character transmission system based on singlechip microcomputer is designed with STC12 microcontroller as the main control chip. By adopting the line-of-sight link and optical amplitude modulation-direct detection technology, the light-emitting diode(LED) is modulated at high speed, and the anti- environmental noise capability of the system is tested. The experimental results show that the system can realize high-speed data communication of Chinese and English characters within 3 m, with an accuracy of more than 90%. It has good anti-interference performance and easy to carry.

Aiming at the problem that the indoor white light visible light communication system basically focuses on the research of direct link and ignores the practical application value of diffuse link, by establishing a closed experimental space environment, the visible light communication experimental system is designed to measure the spatial distribution law of diffuse reflection signal. The feasibility of the application of diffuse reflection signal in practical environment is explored, and the white light visible light communication experiment based on diffuse reflection is carried out. The experimental results show that diffuse reflection provides a solution to the problems such as signal interruption caused by blocking and shadow in the direct link of visible light communication and poor communication quality in indoor corners.

The traditional visible light communication(VLC) dimming scheme uses redundant compensation symbols to achieve dimming control, resulting in low transmission efficiency. In order to improve the transmission efficiency of VLC system, a dimming control scheme based on shaping polarization code is proposed. The probability of "1" in code word is changed by preencoder, so the dimming control function can be realized without adding redundant compensation symbol. The experimental results show that the shaping polarization code scheme has higher transmission efficiency and better error correction performance compared with the compensating polarization code scheme. In the two typical dimming scenarios of 25% and 12.5%, the encoding gain can reach at least 1.9 dB and 3.3 dB respectively.

Aiming at the performance optimization problem of multi-pulse position modulation(MPPM) method applied to visible light communication, an MPPM transmission design scheme based on autoencoder model is proposed. The scheme uses fully connected network(DNN) and convolutional neural network(CNN) to build autoencoder models respectively. The encoder side realizes the generation of MPPM source symbols by adopting multi-stage training strategy and self-defined loss function regulation, the decoding end completes the functions of channel and MPPM signal feature learning through a network constructed by a fully connected layer or a one-dimensional convolution layer. The simulation results show that the bit error performance of MPPM transmission system based on self encoder can reach and slightly better than the performance of traditional maximum likelihood sequence detection.

The vast majority of existing wireless solutions applied to various industrial environments are based on traditional radio frequency technology paradigm, which has many potential risks such as electromagnetic interference and industrial data leakage. Many research institutions are actively introducing various wireless optical technologies into industrial intelligence scenarios. In order to clearly describe the research progress of wireless optical technology in industrial scene application, this paper combs the aspects of industrial wireless optical multi carrier transmission, multiple input multiple output(MIMO) transmission, positioning and prototype communication system experiment, and gives the main technical challenges and potential solutions for the further development of industrial wireless optical technology preliminarily.

Aiming at the limited modulation bandwidth of light emitting diode(LED) in visible light communication(VLC) system, a hybrid scheme combining multiband carrierless amplitude-phase(mCAP) modulation with pairwise coding(PWC) is proposed. In this scheme, PWC technology pairs and interleaves high-frequency sub-bands with low-frequency sub-bands to effectively overcome the signal to noise ratio(SNR) imbalance problem. Validated by the experiment, the optimal rotation angle corresponding to PWC coding is given when the number of subbands is 2, 3, 4 and 5 respectively, and the maximum modulation order that can be transmitted under the specified bandwidth is analyzed. The experimental results show that compared with the traditional mCAP-VLC system, the mCAP-VLC system based on PWC can transmit higher modulation order.

Aiming at the problem that the indoor visible light communication(VLC) light-emitting diode(LED) layout cannot guarantee the quality of lighting and communication at the same time, an optimization scheme of LED layout is proposed. First, different indoor VLC system models are introduced. Then, based on the VLC channel model, the signal to interference to noise ratio coverage(SINRC) is determined as the optimization target, which is defined as the ratio of signal to interference plus noise ratio(SINR) greater than the threshold in the case of ensuring indoor lighting quality. Finally, a differential evolution algorithm(DEA) is used to determine the optimal deployment quantity and position of LED in different environments. In addition, the random rotation of the user is considered and its effect on indoor SINRC is investigated by simulation. The simulation results show that the LED layout optimization scheme based on DEA has good performance.

Visible light communication(VLC) physical layer security research plays an important role in the commercialization and popularization of VLC. Based on a multi-input multiple output(MIMO) indoor VLC system, the lower limit expressions of the se-curity capacity in the case of eavesdroppers colluding and non-colluding are derived and compared. Then, an optimal beam forming scheme is proposed to maximize the security rate in the case of non-colluding. Finally, numerical simulation is carried out. The simulation results show that the confidentiality capacity in the case of non-collusion is better than that in the case of collusion, and the number of eavesdroppers only has an effect on the confidentiality capacity in the case of collusion.

Aiming at the problem that the centralized deployment mode of traditional supplier information base is difficult to ensure the authenticity of data and system security, a heterogeneous supplier electronic file storage system of blockchain is constructed. Firstly, the distributed storage of heterogeneous data is carried out according to the blockchain and interPlanetary file system(IPFS), the blockchain saves small files, and the IPFS network splits and stores heterogeneous large files, so as to improve the storage speed of files. Then, the file access control strategy of blockchain smart contract is adopted to ensure the security and transparency of supplier file management and meet the file security access in multiple scenarios. The simulation results show that the system can not only ensure efficient throughput performance, but also effectively ensure the information security and privacy on the chain.

Because the accuracy and stability of traditional humidity sensors are easily affected by the environment, an optical fiber humidity sensor based on optical vernier effect sensitization is proposed. Two cascaded optical Fabry-Perot(F-P) interfer-ometers are obtained by successively welding a small hollow core capillary and single-mode fiber at the tail end of single-mode fiber. Adjusting the cavity length makes the spectrum range of the two interference spectrum close to each other, thus obtaining higher humidity sensitivity through the optical cursor effect. The sensing of humidity is realized by polyvinyl alcohol gel film coated on the end of the optical fiber sensing head. The experimental results show that the humidity sensitivity of the sensor based on optical vernier effect is 4.05 times higher than that of the single F-P optical fiber sensor when the relative humidity(RH) varies from 40% to 65%.

The current research work focuses on how to improve the sensitivity and linear response of Fabry Perot(F-P) sensor, while the directional response of the sensor is less studied in partial discharge location. The directional sensitivity of F-P acoustic sensor based on micro electro mechanical system(MEMS) is experimentally studied. The thickness of diaphragm is 5 μm, and the acoustic signals from discharge sources with different incident angles and linear distances are measured and analyzed. The experimental results show that the sensor has 5.90 dB amplitude fluctuation and high signal-to-noise ratio in the incident range of ±60°, which makes it capable of detecting weak partial discharge signals beyond 3 m.

In order to further increase the measurement range of optical fiber stress sensor, an optical fiber Mach-Zehnder interference(MZI) stress sensor based on the interference of vortex beam and plane wave beam is designed, and a new stress demodulation model is proposed. The principal component analysis method is used to extract the main features of the fork interference pattern set, and the relationship between the phase difference between the two beams and the correlation coefficient of the corresponding interference pattern for the main features is obtained. According to the stress phase difference relationship, the relationship between the stress and the correlation coefficient is obtained, a stress sensing experimental system based on the interference between vortex beam and plane wave beam is built, and the stress sensing response characteristics of the system are studied. The experimental results show that the stress response sensitivity of the sensor is 0.25694 rad/MPa, and the maximum stress measurement range is 3392 με。

In order to improve the sensitivity of biological parameter detection and avoid electromagnetic interference, a photonic crystal fiber sensor based on surface plasmon resonance is proposed for the detection of carcinoembryonic antigen(CEA) solution. The sensor adopts a three-layer arrangement structure of air holes with different diameters. The gold film is plated on the surface of the fiber core as a metal layer, and a specific adaptation layer based on nucleic acid aptamer is added between the gold film and the CEA solution. The full vector finite element method is used to numerical simulate and emulate the sensor. The simulation results show that the sensor has obvious surface plasmon resonance effect in the refractive index range of 1.33~1.41. The sensitivity of the structure can reach 10000 nm/RIU and the linearity is 0.94919.

In order to realize the wavelength division multiplexing function of light wave and improve the radiation efficiency of light wave, a two-dimensional photonic crystal wavelength division multiplexer based on double ring resonator cavity structure is designed. Firstly, the structural parameters of photonic crystal are designed. Through the coupling of resonator cavity and microcavity, the light waves with different frequencies are selected, and the photonic crystal filter model is designed. Then, according to the mode coupling theory, the optimal coupling conditions between the backbone waveguide and the resonator cavity are determined, and the wavelength division multiplexer is designed. Finally, the scattering dielectric column is added and the relative position between the two resonators is adjusted to optimize the wavelength division multiplexer to improve the transmittance of light wave. The experimental results show that the multiplexer structure realizes the improvement of 1.562 μm、1.544 μm、1.451 μm、1.436 μm、1.366 μm and 1.322 μm, the maximum radiation efficiency of light wave can reach 96% and the device size is only 13.44 μm× 16 μm。

Aiming at the slow convergence speed of gradient descent(GD) algorithm, a finite impulse response(FIR) dispersion equalizer based on GD optimization algorithm is proposed. The momentum gradient descent(MGD) algorithm is used to optimize the gradient exponentially weighted average(CMGD algorithm for short) and correct the deviation at each update iteration. The single carrier coherent optical transmission system is built by optisystem and simulated with MATLAB. The simulation results show that when the learning rate is 0.1 and the momentum coefficient is 0.98, the convergence value of CMGD algorithm reaches 0.015461. Under different SNR conditions, the equalization effect of the equalizer designed by CMGD algorithm is better than that of GD-FIR and classical time-domain FIR, and the hardware implementation complexity is reduced.