To address the issues of low positioning accuracy and the risk of overfitting when the random forest (RF) algorithm is used for indoor visible light positioning, a sparrow search algorithm(SSA) optimized RF algorithm for indoor visible light positioning based on sine population mapping(SPM) and Cauchy distribution(hereinafter referred to as SCSSA-RF algorithm) is proposed. Firstly, this algorithm establishes a fingerprint database using the received signal strength values and position coordinates. Then, it uses the global search capability of SCSSA to optimize key parameters of the RF algorithm, inputs the data into the optimal model for training, and averages the prediction results from the decision trees to obtain the predicted value for the target location point. The experimental results show that SCSSA-RF algorithm converges faster than the unimproved SSA-RF algorithm, the average positioning error of SCSSA-RF algorithm is 0.08 meters, with errors mainly concentrated between 0.05 to 0.1 meters. At a positioning error of 0.2 meters, the prediction accuracy of SCSSA-RF algorithm reaches 93%.

Regarding the strong correlation between channels that may be caused by dense layouts of transmitters and receivers in indoor multiple-input multiple-output visible light communication systems, which can lead to inter-channel interference and significantly affect communication quality, a challenge that is addressed by proposing the application of lattice basis reduction techniques to indoor visible light communication systems.By implementing a reduction operation on the visible light channel to reduce the correlation between channels, and applying the reduced channel matrix, combined with the zero forcing criterion and the minimum mean square error criterion, precoding design is carried out at the transmitting end. The simulation experiment results show that the application of lattice reduction technology can effectively improve the orthogonality of indoor visible light channels, and the more transceivers there are, the more obvious the improvement effect on channel correlation. The application of zero forcing criterion and minimum mean square error criterion can effectively improve the system's error rate performance.

Radio frequency(RF) wireless communication systems face issues of spectrum scarcity and access congestion. Integrating visible light communication (VLC) with RF can complementarily achieve advantages of high data rates and broad coverage. For indoor hybrid VLC/RF heterogeneous network scenarios, a joint optimization method that combines cell scaling power con trol and user association has been proposed to maximize the total system transmission rate under power and quality of service (QoS) constraints for users. By modeling this problem as a multi-variable coupled mixed-integer nonlinear programming (MINLP) optimization problem, it is effectively solved using the BARON optimization solver. The simulation results show that the integrated cell scaling in VLC/RF heterogeneous networks can achieve superior power allocation and user association, thereby enhancing the overall system transmission rate.

In order to improve the covert transmission performance of visible light mobile communication systems, a random transmission power strategy was adopted to increase the detection uncertainty of monitors. An effective covert rate maximization problem under joint constraint conditions was constructed. Based on graphical methods, the optimal transmission power allocation ratio of light emitting diode (LED) transmitters and the optimal drone position were obtained, and simulation experiments were designed. The simulation results have verified that the proposed scheme can significantly improve the effective concealment rate of covert users in visible light mobile communication based on non orthogonal multiple access.

To enhance the robustness and positioning accuracy of the Elman indoor visible light position sensing model, a bio-inspired multi-feature fusion learning method for indoor visible light position sensing is proposed. This method first preprocesses the acquired visible light images to ensure the accuracy of feature extraction. Then, by fusing features from different levels of a pre-trained neural network model, it constructs a position-sensing feature library, thereby enhancing feature representation capa bility and richness, which improves the model's position sensing precision. Finally, the dung beetle optimization(DBO) algorithm is employed to optimize the topology and weight parameters of the Elman neural network, addressing issues where traditional Elman neural networks easily fall into local optima in indoor position sensing, accelerating convergence speed, and enhancing generalization performance. The experimental results show that within a 3D space of 4 m×3.5 m×3 m, the proposed algorithm achieves an average positioning error of 0.21 m, with 91.3% probability of average positioning error is less than 0.4 m, improving positioning accuracy by 22.3% compared to the Elman algorithm.

To improve the spectral efficiency of the system and solve the problem of single signal mode in traditional orthogonal frequency division multiplexing(OFDM) systems, a layered hybrid visible light communication system based on binary on-off keying(OOK) and adaptive constellation optimization(ACO-OFDM) is designed. The transmitting end of the system uses OOK signals to limit the OFDM signal, generating ACO-OFDM and non adaptive constellation optimization (NACO)-OFDM signals. Add the OOK signal as a direct current (DC) bias to the NACO-OFDM signal to ensure that the output is a unipolar real signal. The receiving end adopts signal reconstruction technology, omitting the noise estimation step and reducing the receiving complexity to about 1/5 of traditional receivers. The simulation results show that the system can effectively demodulate OFDM and OOK signals, and at the maximum transmission layer, its spectral efficiency is about twice that of conventional single-layer systems.

To improve the transmission rate and communication quality of indoor visible light communication(VLC), a VLC channel estimation and equalization method based on orthogonal time-frequency space(OTFS) modulation and residual convolutional neural network(ResCNN) is proposed. First, an indoor VLC system based on OTFS modulation is constructed, and a time-frequency domain channel estimation model based on training sequences is proposed. Then, ResCNN learns the mapping relationship between the least squares(LS) channel estimation and the optimized channel to solve the inter-symbol interference problem and improve the accuracy of channel estimation. Finally, the visible light channel is estimated and equalized using ResCNN. The experimental results show that when the signal transmission distance is 1 m and the transmission rate is 512 Mb/s to 1.5 Gb/s, the bit error rate estimated by the proposed method is lower than 3.8×10-3, effectively improving the transmission rate and communication quality of indoor short-distance VLC.

In order to overcome the impact of multipath effects and inter user interference on communication reliability in indoor multi-user scenarios faced by existing non orthogonal multiple access (NOMA) models in visible light communication(VLC) systems, improve spectral efficiency and communication rate, a NOMA-VLC system combining deep neural network (DNN) channel estimation and virtual time reversal(VTR) channel equalization is proposed. By analyzing the characteristics of multi-user NOMA-VLC channels, DNN is used for accurate channel estimation, and VTR technology is utilized to achieve channel equalization, focus energy, suppress multipath effects, enhance communication reliability and user fairness. The simulation results show that in a two user scenario, the system performance improved by 5.1 dB and 4.9 dB respectively at a bit error rate of 10-3, with performance advantages of 2 dB and 2.4 dB compared to other algorithms.

When orthogonal frequency division multiplexing(OFDM) systems are applied to visible light communication(VLC), they are prone to issues such as multipath interference, frequency offset noise, and a high peak-to-average power ratio(PAPR). A filter bank multi-carrier (FBMC) based VLC system is proposed. Firstly, by employing offset quadrature amplitude modulation (OQAM) and a composite filter consisting of inverse fast fourier transform(IFFT) and polyphase network(PPN), the system effectively improves bit error rate performance and reduces out-of-band leakage. Then, it is validated through simulation using the Monte Carlo method. The simulation and experimental results show that under conditions where the number of subcarriers is 128, the modulation scheme is 16QAM, and the system bit error rate is 10-3, the proposed system requires an improvement of approximately 5 dB in signal-to-noise ratio (SNR) compared to the OFDM system. Meanwhile, at a normalized frequency of 0.1, the out-of-band leakage power of the proposed system is reduced by 40 dB relative to the OFDM system, with the system's modulation error ratio(MER) reaching 19.6 dB.

Aiming at the issues of long communication delays and offline processing caused by data preparation and training in machine learning-based vortex beam orbital angular momentum-shift keying(OAM-SK) communication schemes, this paper proposes an OAM-SK communication scheme for underwater vortex beams based on spiral spectrum detection. This scheme enables the recognition of the mode of the vortex beam by directly performing spiral spectrum calculations on the optical field of the received vortex beam at the receiver end, thereby achieving information transmission. The simulation results show that although the proposed scheme has slightly lower reliability, it significantly enhances the effectiveness of the communication system.

To address the issue of low positioning accuracy caused by incomplete image capture of light emitting diodes(LEDs), a novel indoor visible light positioning algorithm based on visual error compensation is proposed. This algorithm consists of three components: a pixel coordinate fitting and restoration algorithm, a cyclic coordinate optimization algorithm, and a dual-lamp positioning algorithm that compensates for height differences. These methods are designed to handle rectangular luminaire images captured under various occlusion conditions, reduce the impact of image noise on positioning results, and improve positioning accuracy in scenarios with tilted ceilings. The experimental results show that compared to positioning accuracy under unobstructed conditions, the positioning error using the pixel coordinate fitting and restoration algorithm is less than 1.5 cm. The cyclic coordinate optimization algorithm can compensate for the effects of image noise on positioning, reducing the positioning error by 60%. Additionally, the dual-lamp positioning algorithm that compensates for height differences reduces the average positioning error by 65%.

To monitor the water level of the well field in the in-situ leaching uranium mine in real time, a real-time monitoring sensor for the water level of in-situ leaching uranium mining based on Fabry-Perot(F-P) cavity is designed. The cavity length of t he sensor is controlled by an electric precision displacement platform and a custom fixture. The sensing head is fabricated by bonding and coupling silicon wafers, bases, and ceramic ferrules. Field verification tests are conducted in a certain in-situ leaching uranium mine in Inner Mongolia. The results indicate that the sensor's pressure measurement range is 0 to 1.5 MPa, with a pressure sensitivity of 1.68 m/MPa and linearity of 0.998, achieving online monitoring of water level parameters in production wells.

To address the limitations of traditional array-based fiber probes in measuring gas-holding rate in gas-liquid two-phase flows, such as insufficient spatial coverage and flow field disturbance, a new method for gas-holding rate imaging using single-mode fiber probes has been proposed. This method involves precise measurements at selected key positions using single-mode fiber probes and employs Kriging interpolation algorithm to estimate the gas-holding rate in unmeasured regions, thereby achieving imaging. The experimental results show that under gas flow rates of 0.2 L/min, 0.3 L/min, 0.4 L/min, and 0.5 L/min, this method can generate high-quality gas-holding rate images, successfully visualizing the distribution of gas-holding rate within the measurement region.

To mitigate the influence of temperature fluctuations on biological parameters, this paper proposes a multi-parameter sensing structure based on ring-core photonic crystal fiber (PCF) detection for measuring both the concentration of the breast cancer susceptibility gene (BRCA) and temperature. The proposed structure features an outer surface coated with a gold film, which creates a surface plasmon resonance sensing channel specifically designed for detecting BRCA concentration. Simultaneously, a thermosensitive liquid is filled into the central air holes of the PCF, forming a directional coupling sensing channel dedicated to precise temperature measurement. This design enables the correction of potential impacts of temperature changes on BRCA concentration measurements. The simulation results indicate that the concentration and temperature sensing channels of the sensor structure operate independently. Specifically, for the concentration sensing channel, when the BRCA concentration ranges from 0 to 7.69 mM, the concentration sensitivity can reach 19.5 nm/mM. For the temperature sensing channel, within a temperature range of 20 to 40℃, the temperature sensitivity is -6.4 nm/℃.

In the dense wavelength division multiplexing system, when adjusting the process using a narrowband filter, it is necessary to measure the thickness, thickness distribution, and residual stress of a single layer of film. To solve the complexity and repeated positioning errors introduced by the need to use multiple systems for measurement, a measurement system based on ellipsometry spectrum and the optical lever method is introduced, which successfully achieved precise measurement of these parameters on a single integrated device. The experimental results show that the thickness deviation measured by the designed measurement system is less than 0.21%, the thickness distribution deviation is less than 0.025%, the curvature and the residual stress deviation is within ±1°. The system can meet the actual measurement needs.

To enhance the performance of future elastic optical networks, a modulation format recognition method based on amplitude density features is proposed. This method uses amplitude density features as input to an improved MobileNetV2 model to identify modulation formats and introduces a normalized attention mechanism (NAM) for accurate identification of transmission signal modulation formats. The feasibility of this approach is verified in 28 GBaud quadrature phase shift keying(QPSK), 8-level quadrature amplitude modulation (8QAM), 16QAM, 32QAM, 64QAM, and 128QAM transmission systems. The experimental results show that the minimum optical signal-to-noise ratio (OSNR) required for achieving 100% recognition accuracy for each modulation format is below its corresponding 20% forward error correction (FEC) threshold. Furthermore, within a broad OSNR range, a recognition accuracy of 99.62% is achieved. In optical networks with residual dispersion, this approach still maintains high recognition performance.

To enhance the information access capacity and signal reception and transmission quality, the remote base station units in heterogeneous passive optical access networks are increasingly becoming miniaturized and densely distributed. Photonic wireless power over fiber(PWoF) technology, which transmits energy via optical fiber to the remote unit, is expected to be the preferred solution for powering the access end. Integrating PWoF technology into passive optical access networks can transmit information and energy simultaneously on a single optical fiber, with certain practical application value. This paper reviews different wavelength-division multiplexing and space-division multiplexing structures of PWoF schemes in passive optical access networks, compares and analyzes the application features, functional realization, and research progress of each technology scheme, and finally discusses the challenges faced by PWoF in its evolution process and prospects.

In order to improve the contrast of the half adder, a high contrast Terahertz wave half adder design scheme based on mode interference is proposed. In this scheme, a linear waveguide is introduced into the complete two-dimensional square lattice tellurium dielectric pillar photonic crystal, and the logic operation function is realized. A comprehensive performance analysis of the half adder is conducted by combining the plane wave expansion method (PWM) and the finite difference time-domain method (FDTD) in Rsoft software, and optimization is carried out for the position of the incident waveguide and the offset of the dielectric column. The simulation results show that the designed half adder can perform two input half adder logic operations in the 2.85 THz band. The contrast between the sum end and carry end of the half adder can reach 21.37 dB and 5.96 dB, respectively, with an overall response time of 25.2 ps. Theoretically, it can achieve a data transmission rate of 39.68 Gb/s. The structure of the half adder is simple and compact, with a size of only 0.84 mm×0.72 mm.