Abstract:For the limitations of existing channel emulators on the channel numbers and scalability, a flexible emulator for multiple-input multiple-output (MIMO) channel is designed and implemented. By adopting an improved coordinate rotation digital computer (CORDIC) algorithm, the emulator can generate large number of accurate random channel fading in real-time, but consumes less hardware resources. On this basis, a flexible hardware framework based on the MIMO discrete channel model is also developed. With the advantages of field-programmable gate array ( FPGA) on parallel processing, the emulation framework is implemented on a FPGA platform. Hardware measurements are conducted with the parameters of extended vehicular A model (EVA) in 3GPP standard. The test results for both static and time-variant scenarios show that the output statistical properties of proposed emulator, i. e. , power delay profiles (PDP) and Doppler power spectrum densities (DPSD) are consistent well with the theoretical ones. Thus, it can be applied on the design verification, algorithm optimization, and performance evaluation of wireless communication devices.

Abstract:In the aerospace camera development process, in order to test and evaluate the imaging performance of its complete system, a set of test system for the imaging performance of aerospace cameras based on Gigabit Ethernet has been developed aiming at the characteristics of high resolution, high quantization digits and high frame rate. In the test system, FPGA is used to complete the Gigabit Ethernet protocol to achieve the high-speed transmission of image data, the image is displayed in real time by the host computer. Then image processing and calculation are performed, after theoretical analysis of the modulation transfer function, it is determined that the contrast method is used to calculate the modulation transfer function of the camera (MTF). The experiments have verified the feasibility of real-time display of image data and the correctness of image data transmission, at the same time, the MTF values of test images with different resolutions of 2 048× 2 049, 2 564× 2 162, and 2 045× 256 were obtained by using three different models of cameras, the standard deviations of the three sets of data were obtained and compared with traditional manual test methods. The experiment shows that the test device greatly improves the test efficiency and test accuracy, it realizes the real-time evaluation of the overall system imaging performance.

Abstract:Because of the measurement error, the measurement result is uncertain. Measurement uncertainty is an important indicator for evaluating the quality of measurement results. The ratio of the sampling length to the signal period is divided into integer and decimal parts, and the measurement errors of the DFT frequency and phase difference are deduced. The Class B uncertainty evaluation method is adopted to analyze the frequency measurement uncertainty caused by spectrum leakage, windowing, and fence effect, and the synthetic uncertainty of frequency measurement is obtained. Then the phase difference uncertainty is determined through the relationship between phase difference measurement error and frequency measurement error. Finally, through calculation and analysis, the relationship between the sampling length and measurement error as well as uncertainty of frequency and phase difference is verified. The analysis results show that the evaluation process is feasible and effective.

Abstract:Aiming at the problem that the simultaneous localization and mapping ( SLAM) algorithm has a large pose error and inconsistent map construction when a moving object appears in mobile robot’ s operating scene, an improved visual SLAM algorithm based on feature point motion vector is proposed. Firstly, the algorithm of motion points based on feature point motion vector is introduced. The motion vector can be calculated by combining the initial camera pose, and the Gaussian mixture model parameters of its angle are solved by using the expectation maximization method. And the motion point detection result of the previous frame is used to distinguish motion points in the current image. Secondly, the camera pose will be optimized based on results of the motion point detection. Then the image is pre-processed, and images with a number of motion points and higher similarity to the previous frame will be eliminated, which can improve the calculation efficiency of loop closure detection. Finally, the scene is described by using feature points after excluding dynamic features, and the similarity score calculation function of two images at a single node is improved. After loop closure confirmation, the correct loop is obtained. The datasets experimental results show that the improved algorithm has better robustness and higher accuracy in the pose estimation. And it can effectively detect the existence of loops in the scene and has a good mapping.

Abstract:Aiming at the problem of “ pseudo regression” in traditional machine learning, ignoring the long-term dependence between non-stationary sequences, the cointegration theory and vector error correction model were proposed to predict the performance degradation trend of rolling bearings, and then predicted the remaining useful life (RIL). Firstly, extracted the kurtosis value, peak-to-peak value and root mean square value from the vibration signal, and analyzed its stability. Then, tested the cointegration relationship of the time domain features and established vector error correction models based on the test results, and performed the residual analysis. The analysis results were close to the Gaussian white noise distribution, indicating that the models were good. Finally, used the models to predict the bearing performance degradation trend and applied RUL, and RMSE, MPE, MAPE and the unequal coefficients of Theil to quantitatively evaluate the prediction results. Experimental results show that the proposed vector error correction model has higher prediction accuracy than the differential average moving autoregressive-Kalman filter model, and simplifies the modeling process.

Abstract:Aiming at the air combat scenario where the defensive missile intercepts the attacking missile, this paper proposes an optimal cooperative guidance algorithm with two-to-one active defense. This algorithm adds a second defensive missile pair on the basis of the traditional one-to-one optimal guidance algorithm protect high-value target aircraft, and derive the optimal cooperative guidance law in two-to-one combat mode. The simulation results show that compared with the proportional guidance (PN) guidance law, by introducing an optimization index that minimizes the deviation of the relative end intercept angle, the relative distance between the two defensive missiles is kept within a very small range, which achieves a very good effect of coordinated interception, it reduces the energy consumption of defense missile control, thereby verifies the effectiveness and superiority of the optimal coordinated guidance rate.

Abstract:According to the problems that the fault features identification of rolling bearing vibration signal, a method for fault feature extraction was proposed base on the improved EEMD with multiple population differential evolution (MPDE) and adaptive resonance demodulation technique (ARDT). Firstly, in order to solve the problem that the EEMD􀆳s parameters selection depending on individuals’ experiences, an evaluation function for distribution characteristics of extreme value points was established. It was used to optimize white noise amplitude using MPDE. Then, EEMD adaptive decomposition was implemented. Secondly, effective signals of the decomposed IMF components were reconstructed using criteria for kurtosis and relativity. The signal de-noising process was realized. Finally, the center frequency and bandwidth of band-pass filter was adaptively determined based on ARDT, and the fault characteristic frequency was extracted using envelop demodulation analysis. A simulation signals and a rolling bearing test results show the validity of the proposed method.

Abstract:In order to achieve non-contact accurate measurement of the mechanical wear of the elevator traction sheave rope groove, a wear detection system for the elevator traction sheave rope grooves was developed based on monocular vision technology. The industrial camera was fixed on the inner side of the steel wire rope through a special fixture, and the images were collected by aligning each rope groove. OpenCV was used for image preprocessing and rope groove edge detection, least squares method was used for rope groove edge numerical fitting, and then according to the physical model structure, the calculated value was compensated for the occlusion to achieve real-time accurate measurement of the bottoming distance of the wire rope. The experimental results indicated that the RMSE of system measurement is less than 0. 07 mm, the system has higher accuracy and simplicity than traditional manual measurement methods, and can be used to accurately measure the amount of wear and sink of the sheave groove.

Abstract:The purpose of the most traditional intersection signal is to improve the traffic efficiency of vehicles, without considering the pedestrian factors at the intersection and vehicle exhaust emissions, it is easy to cause traffic accidents and aggravate environmental pollution. On the basis of the improved Webster delay model, proposing a multi-objective function with vehicle delay, pedestrian delay, parking rate, traffic capacity and exhaust emission as the optimization index. Using the improved PSO algorithm to find the optimal signal period for the established multi-objective function, and then using the flow ratio calculation formula to find the red and green phases Light time. The use of this timing scheme can reduce vehicle delays by 18. 7%, parking rate by 15. 9%, pedestrian delays by 17. 8% and exhaust emissions by 38% during peak periods of traffic flow. At peak periods, vehicle delays can be reduced by 11. 6%, parking rate by 6. 9%, capacity by 8%, pedestrian delays by 28. 6% and exhaust emissions by 5. 6%. Using this timing scheme of this subject can improve road capacity, reduce traffic accidents, reduce exhaust emissions and reduce road congestion to a certain extent.

Abstract:To solve the problem of network QoS degradation caused by the limited energy, fast movement and data of UAV Ad-Hoc network nodes, the improved firefly multipath routing algorithm (AOMDV-FMRA) was put forward. Firstly, in order to reduce the influence of speed on the path stability, a boundary evaluation factor is introduced into the route discovery process to adapt to the topology change. Then, the reverse route is selected according to the load information of the nodes on the path. Finally, the energy evaluation parameters are mapped to the firefly algorithm to process the collected path energy information as the basis for flow allocation. Experimental results show that, compared with AODV and AOMDV, in the environment of high node speed and large network traffic, the algorithm has good performance in aspects of end-to-end delay, network survival time, delivery success rate and routing cost.

Abstract:As an important special light source, superluminescent diode ( SLD) has been proposed to evaluate the reliability of SLD performance degradation based on graphic methods in view of the high requirement of the reliability and lifetime prediction accuracy of SLD. First, the accelerated stress method is used to obtain the short-term failure-free data of the SLD. Secondly, according to the degradation trajectory, a curve equation suitable for the degradation trajectory is fitted by using the least square method, and the pseudo failure life of the SLD is calculated according to the failure threshold. Finally, Minitab was used to identify the individual distribution of the pseudo failure life, and a log-normal distribution with high acceptance was used to obtain the average working time before failure at room temperature. Experimental results show that this method can accurately predict the reliability and life of SLD, and its accuracy is improved by 9. 09% compared with other methods.

Abstract:Aiming at the problem of how to ensure the measurement accuracy of the on-site mounted ultrasonic flow measurement device, a precise measurement method of geometric parameters of the on-site mounted probe based on articulated arm coordinate measuring machine(AACMM) is proposed. Firstly, the flow-rate calculation model of ultrasonic flow measurement device is analyzed, the results show that the measurement accuracy is based on the accurate measurement of the geometric parameters of the probes. On this basis, through special design of probe and comparison of internal and external test, feasibility of using external probe 3-D coordinates value measured by AACMM to calculate probe actual geometric parameters is verified. Finally, in three test cases of middle route of south-tonorth water transfer project, 3-D coordinates value of probe of plug-in four-path ultrasonic flow measurement device are accurately measured by AACMM, precise geometric parameters of the probe and corrected weight coefficients are obtained by analysis. Through analysis of the influence of different weight coefficients on the average flow velocity of ultrasonic flow measurement device, the results show that the method can improve the measurement accuracy of the on-site mounted ultrasonic flow measurement device by 0. 02% ~ 0. 58%, the accuracy of scribe localization is evaluated, and also the rationality and effectiveness of the method are verified.

Abstract:Chip testing is an effective method to prevent defective or faulty chips from reaching the market. During test application, largescale test patterns are transmitted to a system-on-chip through chip pins. Test application time mainly depends on the transmission time with the limited chip pins. Code-based compressions can reduce the transmission time with no information for the circuit under test, and save storage space for test patterns. Therefore, it is widely applied to compress the test set composed of test patterns. However, codebased compressions fail to fully exploit the characteristics of the test set, resulting in poor compression efficiency. To solve this problem, this study proposes a transform-decomposing method based on significant components of the test set, which makes the current coding compression effect significant. First, extract the significant components that best represent the characteristics of the test set, and then use them as vectors to construct a matrix. Via performing matrix transform in mathematics between the matrix and the test set, the test set can be decomposed into a primary component set (PCS) and a residual component set (RCS). Compared with the original test set, the RCS has better compressibility. As the PCS can be generated on-chip, it does not occupy the transmission time. The experimental results on ISCAS′89 benchmark circuits show that the highest compression ratio for the proposed method reaches 80. 53% on average. Compared with the state-of-the-art transform-composing method, the average compression ratios for different code-based compressions are increased, and that for the most improved code-based compression is increased by 5. 27%.

Abstract:In order to solve the problem that the Beidou multipath error caused by environmental complexity cannot be effectively eliminated, an adaptive unscented Kalman filter (UKF) method based on a new error model is proposed. Firstly, the method uses the measurement state amplification method to solve the problem that the measurement noise is colored noise, and then uses the improved Sage-Husa adaptive UKF to dynamically estimate the system noise and measurement noise, so as to solve the situation that the error offset effect is not obviously caused by the undetermined statistical characteristics of noise. Comparing this method with the traditional UKF, the experimental results show that the improved Sage-Husa adaptive UKF algorithm under colored observation noise can reduce the multipath error bynearly 60%. This method has strong applicability for the case where the noise generated by the multipath error is unknown to the Beidou positioning.

Abstract:Radar target track recognition can help a commander to estimate the intention and tasks of the adversary, and facilitate the decision making of the commander. However, little attention has been paid on the radar target track recognition problem. Given the radar target data such as time, distance, direction, and so on, a radar target track recognition based on spatial-temporal relationship of track points is proposed to solve the problem of target track recognition. The algorithm selects the attributes of radar track data and then extracts the space characteristics of the track points in the spatial domain. The algorithm constructs the recursive neural network to capture the characteristics of the track points in the time domain and realize the classification and recognition of the target track. We conduct experiments through simulations, and the simulation results show that the proposed algorithm TRST can effectively improve the target track recognition performance in terms of accuracy, precision, recall and F1-Score.

Abstract:Unmanned technology changes human lifestyle. The high-precision map with the lane line attribute plays a crucial role in the unmanned field. Proposing a detection method based on MultiRes + UNet network, aiming at the problem that the low accuracy and low efficiency of existing algorithms in the detection of composite lane lines. This method expands the convolution receptive field by dilated convolution to co-ordinate global information. The MultiRes block and Res path structure are used to reduce the difference between the encoder and decoder features, which greatly reduces the memory requirement. The experimental results show that the proposed algorithm improves the detection speed of the algorithm while ensuring the detection accuracy. When the pure lane, compound lane and shadow fouling lanes are obtained, the harmonic mean F1 scores are 0. 959, 0. 942, and 0. 891, the algorithm is high efficiency and high robustness.

Abstract:The increase of signal-to-noise ratio ( SNR) is an effective measure to improve the measurement accuracy of an ultrasonic flowmeter. However, the increase of transmission voltage for higher SNR could damage the ultrasonic sensor and the receiving circuit; it also introduces higher mutual interference between the transmitting and receiving circuits. Proposing an ultrasonic transceiver design for flow measurement of composite pipes: In the transmission circuit, a central-axis transformer instead of a damping resistor is used. In the receiving circuits, a limiting bridge circuit is employed for protection. The experimental results showed that the optimized transmission circuit proposed in this paper could increase the transmission voltage and reduce the discharge time. The output voltage is amplified by 13. 85 and 4. 5 times without and with loads, respectively. At the same time, the receiving circuit can effectively suppress the mutual interference and improves the power of the receiving branch. The transmission efficiency of the receiving circuits for signals below 5 V exceeds 90%. The circuit design was applied to a non-contact ultrasonic flowmeter, and the fluid in the composite pipeline was measured. The test results showed that the linearity and repeatability of the flowmeter had met the actual requirements of the flow measurement.

Abstract:Task scheduling is an important part of multi-core processor technology. List-based scheduling algorithms are widely concerned because of their low complexity and high efficiency, but the singleness of the task priority list method makes the algorithm not enough to search the solution space, and it is easy to fall into the local optimal. Therefore, proposes an iterative list scheduling algorithm (TPIA) based on task perturbation. The algorithm selects the task disturbance factor to iterate the scheduling list according to a certain disturbance strategy, greedily selects the iterated list, and generates a better scheduling list sequence to obtain better scheduling results. The thesis validates the algorithm through examples and a limited set of random DAG graphs. The results show that the algorithm can effectively improve the scheduling solution, and the scheduling performance can be improved by an average of 16. 51%. It is suitable for processing large-scale and high-entry complex DAG graphs. When the total number of tasks is low and the communication overhead is high, the performance is reduced. The task graphs with an average task node number of less than 130 are grouped and tested to obtain the corresponding upper CCR and its change trend.

Abstract:A wideband vibration harvester based on piecewise linear system is proposed, which is mainly composed of the structure of diamagnetically stabilized levitation, induced coils and elastic films. The floating magnet is freely levitated between two pyrolytic graphite sheets without any external energy input. When a horizontal vibration excitation was applied to the vibration energy harvester, the floating magnet would oscillate between two pyrolysis graphite sheets, the floating magnet impacted to the elastic film with the increase of vibration intensity, which would cause the system stiffness to be segmented, the magnetic flux through the induced coils would change during the movement of the floating magnet, and the external vibration was converted into electrical energy. When the horizontal excitation amplitude is 5 mm, the operating bandwidth of the system reaches 3. 2 Hz, the maximum output voltage reaches 78 mV, and the maximum output power is 56. 3 μW. Under an excitation acceleration of 3 m/ s 2 , a maximum output voltage of 44 mV with the maximum power of 18 μW is obtained by the harvester, and the operating bandwidth of the system reaches 2 Hz.

Abstract:A new improved level set algorithm is proposed to solve the problem of initial contour sensitivity in cervical image segmentation and the problem of unclear image gray level. Firstly, the image is denoised by anisotropic filtering algorithm. Then the region growth algorithm was used on the binary image to extract the rough cervical lesion area. Finally, a level set model based on the new symbolic pressure function is established to refine the initial segmentation. The algorithm can combine local information with global information and automatically allocate the proportion between them. The accuracy, sensitivity and specificity of this method can reach 81. 11%, 63. 97% and 78. 64% respectively, which are 30. 69%, 15. 15% and 4. 37% higher than the traditional level set algorithm. Therefore, this improved level set algorithm has some value and significance in practical application.

Abstract:To accurately and rapidly recognize and classify different kinds of sensing events in distributed fiber optic vibration sensing system, a time-frequency based hybrid feature extraction algorithm has been proposed. In the algorithm, a zero crossing rate based time domain feature vector and a wavelet packet energy based frequency domain feature vector are used as the feature description of the given sensing event. Then, the feature vectors are classified by radial basis function neural network classifier. A series of experimental results show that the vibrations can be accurately recognized from the noise with high efficiency. Specifically, the average identification rate of 94. 5% is achieved and the recognition response time can be limited in 0. 3 s.

Abstract:Traditional machine learning based methods use hand-crafted features in the inertial data to achieve the task of human activity recognition (HAR). However, as these features are normally without abstract high-level knowledge, the recognition rate is thus limited. Deep learning based methods, on the other hand, can avoid the aforementioned disadvantage by learning high-level features through labeled data. In this paper, the convolutional long short-term deep neural networks (CLDNN) is adopted for solving the HAR problem. This network has the features of both the convolutional neural network (CNN) and the recurrent neural network (RNN), which can extract features of different levels and can adopt the correlations in time sequences. Moreover, we use the GRU instead of LSTM as the gated cell of the RNN, which can make the network lighter. Through experiments adopting open source data, we can show that our method has 3% and 7% better recognition rate than CNN and RNN respectively, the training time and forward recognition time has decreased by 14% and 10% respectively, if replace the LSTM with GRU cell.

Abstract:In orthogonal frequency division multiplexing (OFDM) systems, the genetic algorithm-based partial transmit sequence (GAPTS) technology can effectively reduce the computational complexity of PTS, but it is not ideal in improving the peak-to-average power ratio (PAPR) performance. Therefore, proposing to embed a simulated annealing (SA) operator in the GA to construct a hybrid genetic simulated annealing (GSA) algorithm, and it is also applied to search for the optimal phase factor of the PTS. First, chromosome formed by encoding PTS phase factor, the chromosomes composed of random elements are used as an initial population of the GA, and the fitness value of each chromosome is evaluated. Then, the chromosomes are also selected according to their fitness values, and the mutation rules and the mating (or cross) rules of the chromosomes are established to perform the iterative evolution on the population. Finally, the chromosomes in population are updated with annealing temperatures to produce new next-generation populations. The simulation results show that compared with the GA-PTS, the proposed method not only decreases the computational burden, but also effectively reduces the PAPR value of the OFDM system.

Abstract:Ultrasonic ranging is a non-contact measurement method that is widely used in industry and transportation. In order to improve the ranging range of ultrasonic waves and solve the problem of crosstalk noise when the distance between multiple channels is relatively close, applies “ dual one-way pseudorange measurement technology” to the premise of changing the electroacoustic converter and transmitting and receiving hardware circuits. In the ultrasonic ranging system, the dual one-way measurement replaces the traditional reflective two-way ranging, and the clock between the ranging main end and the ranging slave is synchronized by the wireless serial communication to improve the ranging range, and the technology can also eliminate the system itself. Clock synchronization error. At the same time, the “time division multiplexing” technology is applied to perform time slot allocation through radio signal broadcasting, so that only one path in the ranging channel at any time is in a working state, and multi-channel close-range crosstalk-free measurement is realized. The experimental results show that the dual one-way measurement can more than double the reflection ranging range, and the detection accuracy is better than 1%. The reference of the time division multiplexing technique enables measurement of crosstalk without interference in ultrasonic multi-channel short-range channel spacing.

Abstract:This article proposed a synthetic aperture radar (SAR) target recognition method by decision level fusion of monogenic signal using random weighting. The sparse representation-based classification ( SRC) was employed to classify the multi-scale and multicomponent monogenic representations. For erro vectors, the random weight matrix was designed to perform the fusion, which includes a large volume of random weight vectors. The statistics of the fused reconstruction errors were analyzed to form the decision values, which reflect the correlations between the test sample and different classes. Finally,the target label was decided by comparison of the decision values. Extensive experiments were conducted on the MSTAR dataset to evaluate the proposed method, which was compared with some existing SAR target recognition methods. The results showed that the proposed method could effectively improve the overall performance.

Abstract:In order to reduce the limitation of the processing level on the performance of micro-sonic nozzles, three non-standard sonic nozzle structures including the shrink nozzle, the diffusion nozzle, and the conical Venturi nozzle are proposed. It was also analyzed respectively that how the geometry and shape of the nozzle affects flow characteristics. Numerical simulation was used to analyze the internal flow field of the nozzle whose diameter is selected from 0. 03mm to 0. 12mm. Not only the influence of the geometry of the constricted section, the throat and the diffusing section on the nozzle flow characteristics were analyzed, but also the flow characteristics of them were compared. It is found that the discharge coefficient and critical back pressure of the diffusion nozzle are superior, and the processing difficulty of this type of nozzle is the smallest. In addition, throat diameter and throat length have the greatest influence on nozzle flow characteristics. In addition, it is found that the throat diameter and the throat length have the greatest influence on the nozzle flow characteristics. The larger the throat diameter and the smaller the throat length, the larger the nozzle discharge coefficient and the critical back pressure ratio. These studies provide some reference for the development of subsequent micro-nozzles.