JOURNAL OF BEIJING INSTITUTE OF TECHNOLOGY

Radio Frequency Fingerprint-Based Satellite TT&C Ground Station Identification Method

Xiaogang Tang, Junhao Feng, Binquan Zhang, Hao Huan

2023, 32(1): 1-12. doi:10.15918/j.jbit1004-0579.2022.074

[Abstract](912) [FullText HTML](482) [PDF 4850KB](37)

Abstract:
This study presents a radio frequency (RF) fingerprint identification method combining a convolutional neural network (CNN) and gated recurrent unit (GRU) network to identify measurement and control signals. The proposed algorithm (CNN-GRU) uses a convolutional layer to extract the IQ-related learning timing features. A GRU network extracts timing features at a deeper level before outputting the final identification results. The number of parameters and the algorithm’s complexity are reduced by optimizing the convolutional layer structure and replacing multiple fully-connected layers with gated cyclic units. Simulation experiments show that the algorithm achieves an average identification accuracy of 84.74% at a –10 dB to 20 dB signal-to-noise ratio (SNR) with fewer parameters and less computation than a network model with the same identification rate in a software radio dataset containing multiple USRP X310s from the same manufacturer, with fewer parameters and less computation than a network model with the same identification rate. The algorithm is used to identify measurement and control signals and ensure the security of the measurement and control link with theoretical and engineering applications.

Multi-Scale Feature Extraction for Joint Classification of Hyperspectral and LiDAR Data

Yongqiang Xi, Zhen Ye

2023, 32(1): 13-22. doi:10.15918/j.jbit1004-0579.2022.120

[Abstract](165) [FullText HTML](83) [PDF 2905KB](23)

Abstract:
With the development of sensors, the application of multi-source remote sensing data has been widely concerned. Since hyperspectral image (HSI) contains rich spectral information while light detection and ranging (LiDAR) data contains elevation information, joint use of them for ground object classification can yield positive results, especially by building deep networks. Fortunately, multi-scale deep networks allow to expand the receptive fields of convolution without causing the computational and training problems associated with simply adding more network layers. In this work, a multi-scale feature fusion network is proposed for the joint classification of HSI and LiDAR data. First, we design a multi-scale spatial feature extraction module with cross-channel connections, by which spatial information of HSI data and elevation information of LiDAR data are extracted and fused. In addition, a multi-scale spectral feature extraction module is employed to extract the multi-scale spectral features of HSI data. Finally, joint multi-scale features are obtained by weighting and concatenation operations and then fed into the classifier. To verify the effectiveness of the proposed network, experiments are carried out on the MUUFL Gulfport and Trento datasets. The experimental results demonstrate that the classification performance of the proposed method is superior to that of other state-of-the-art methods.

Resonance and Bifurcation of Fractional Nonlinear Systems with Power Damping Term for Robot Grinding

Wei Shi, Qingxue Huang, Jinzhu Zhang, Tao Wang, Ziliang Li, Yanjie Zhang, Xiaoyan Xiong

2023, 32(1): 23-40. doi:10.15918/j.jbit1004-0579.2022.135

[Abstract](2407) [FullText HTML](1220) [PDF 2984KB](110)

Abstract:
A fractional nonlinear system with power damping term is introduced to study the forced vibration system in order to solve the resonance and bifurcation problems between grinding wheel and steel bar during robot grinding. The robot, grinding wheel and steel bar are reduced to a spring-damping second-order system model. The implicit function equations of vibration amplitude of the dynamic system with coulomb friction damping, linear damping, square damping and cubic damping are obtained by average method. The stability of the system is analyzed and explained, and the stability condition of the system is proposed. Then, the amplitude-frequency characteristic curves of the system under different fractional differential orders, nonlinear stiffness parameters, fractional differential term coefficients and external excitation amplitude are analyzed. It is shown that the fractional differential term in the dynamic system is the damping characteristic. Then the influence of four kinds of damping on the vibration amplitude of the system under the same parameter is investigated and it is proved that the cubic damping suppresses the vibration of the system to the maximum extent. Finally, based on the idea that the equilibrium point of the system is the constant part of the Fourier series expansion term, the bifurcation behavior caused by the change of damping parameters in linear damping, square damping and cubic damping systems with different values of fractional differential order is investigated.

GP Algorithm-Based Fourier Transform Infrared Spectrum Trend Term Removal Model

Bo Yan, Shuaihui Li, Hao Chen

2023, 32(1): 41-51. doi:10.15918/j.jbit1004-0579.2022.124

[Abstract](59) [FullText HTML](30) [PDF 2158KB](19)

Abstract:
Trend term removal is a key step in Fourier transform infrared spectroscopy(FTIR) data pre-processing. The most commonly used least squares (LS) method, although satisfying the real-time requirement, has many problems such as highly correlated initial values of the expression parameters, the need to pre-estimate the trend term shape, and poor fitting accuracy at low signal-to-noise ratios. In order to achieve real-time and robust trend term removal, a new trend term removal method using genetic programming (GP) in symbolic regression is constructed in this paper, and the FTIR simulation interference results and experimental measurement data for common volatile organic compounds (VOCs) gases are analyzed. The results show that the genetic programming algorithm can both reduce the initial value requirement and greatly improve the trend term accuracy by 20%–30% in three evaluation indicators, which is suitable for gas FTIR detection in complex scenarios.

Reliability Analysis of Satellite Turntable System under Multiple Operation Modes Based on Multi-Valued Decision Diagrams

Peng Zhang, Zhijie Zhou, Yao Ding, Dao Zhao, Yijun Zhang

2023, 32(1): 52-68. doi:10.15918/j.jbit1004-0579.2022.148

[Abstract](23) [FullText HTML](10) [PDF 1318KB](5)

Abstract:
As a payload support system deployed on satellites, the turntable system is often switched among different working modes during the on-orbit operation, which can experience great state changes. In each mode, the missions to be completed are different, consecutive and non-overlapping, from which the turntable system can be considered to be a phased-mission system(PMS). Reliability analysis for PMS has been widely studied. However, the system mode cycle characteristic has not been taken into account before. In this paper, reliability analysis method of the satellite turntable system is proposed considering its multiple operation modes and mode cycle characteristic. Firstly, the multi-valued decision diagrams (MDD) manipulation rules between two adjacent mission cycles are proposed. On this basis, MDD models for the turntable system in different states are established and the reliability is calculated using the continuous time Markov chains (CTMC) method. Finally, the comparative study is carried out to show the effectiveness of our proposed method.

Design and Optimization of Winograd Convolution on Array Accelerator

Ji Lai, Lixin Yang, Dejian Li, Chongfei Shen, Xi Feng, Jizeng Wei, Yu Liu

2023, 32(1): 69-81. doi:10.15918/j.jbit1004-0579.2022.094

[Abstract](21) [FullText HTML](10) [PDF 999KB](4)

Abstract:
With the rapid development and popularization of artificial intelligence technology, convolutional neural network(CNN) is applied in many fields, and begins to replace most traditional algorithms and gradually deploys to terminal devices. However, the huge data movement and computational complexity of CNN bring huge power consumption and performance challenges to the hardware, which hinders the application of CNN in embedded devices such as smartphones and smart cars. This paper implements a convolutional neural network accelerator based on Winograd convolution algorithm on field-programmable gate array (FPGA). Firstly, a convolution kernel decomposition method for Winograd convolution is proposed. The convolution kernel larger than 3×3 is divided into multiple 3×3 convolution kernels for convolution operation, and the unsynchronized long convolution operation is processed. Then, we design Winograd convolution array and use configurable multiplier to flexibly realize multiplication for data with different accuracy. Experimental results on VGG16 and AlexNet network show that our accelerator has the most energy efficient and 101 times that of the CPU, 5.8 times that of the GPU. At the same time, it has higher energy efficiency than other convolutional neural network accelerators.

MobileNetV3-CenterNet : A Target Recognition Method for Avoiding Missed Detection Effectively Based on a Lightweight Network

Yajing Li, Xiaoyan Xiong, Wenbin Xin, Jiahai Huang, Huimin Hao

2023, 32(1): 82-94. doi:10.15918/j.jbit1004-0579.2022.076

[Abstract](108) [FullText HTML](45) [PDF 4435KB](14)

Abstract:
To solve the problems in online target detection on the embedded platform, such as high missed detection rate, low accuracy, and slow speed, a lightweight target recognition method of MobileNetV3-CenterNet is proposed. This method combines the anchor-free Centernet network with the MobileNetV3 small network and is trained on the University at Albany Detection and Tracking (UA-DETRAC) and the Pattern Analysis, Statical Modeling and Computational Learning Visual Object Classes(PASCAL VOC) 07+12 standard datasets. While reducing the scale of the network model, the MobileNetV3-CenterNet model shows a good balance in the accuracy and speed of target recognition and effectively solves the problems of missing detection of dense and small targets in online detection. To verify the recognition performance of the model, it is tested on 2683 images of the UA-DETRAC and PASCAL VOC 07+12 datasets, and compared with the analysis results of CenterNet-Deep Layer Aggregation (DLA) 34, CenterNet-Residual Network (ResNet) 18, CenterNet-MobileNetV3-large, You Only Look Once vision 3(YOLOv3), MobileNetV2-YOLOv3, Single Shot Multibox Detector (SSD), MobileNetV2-SSD and Faster region convolutional neural network (RCNN) models. The results show that the MobileNetV3-CenterNet model accurately recognized the dense targets and small targets missed by other methods, and obtained a recognition accuracy of 99.4% with a running speed at 53 (on a server) and 14 (on an ipad) frame/s, respectively. The MobileNetV3-CenterNet lightweight target recognition method will provide effective technical support for the target recognition of deep learning networks in embedded platforms and online detection.

Salient Object Detection Based on a Novel Combination Framework Using the Perceptual Matching and Subjective-Objective Mapping Technologies

Jian Han, Jialu Li, Meng Liu, Zhe Ren, Zhimin Cao, Xingbin Liu

2023, 32(1): 95-106. doi:10.15918/j.jbit1004-0579.2022.078

[Abstract](20) [FullText HTML](9) [PDF 9508KB](3)

Abstract:
The integrity and fineness characterization of non-connected regions and contours is a major challenge for existing salient object detection. The key to address is how to make full use of the subjective and objective structural information obtained in different steps. Therefore, by simulating the human visual mechanism, this paper proposes a novel multi-decoder matching correction network and subjective structural loss. Specifically, the loss pays different attentions to the foreground, boundary, and background of ground truth map in a top-down structure. And the perceived saliency is mapped to the corresponding objective structure of the prediction map, which is extracted in a bottom-up manner. Thus, multi-level salient features can be effectively detected with the loss as constraint. And then, through the mapping of improved binary cross entropy loss, the differences between salient regions and objects are checked to pay attention to the error prone region to achieve excellent error sensitivity. Finally, through tracking the identifying feature horizontally and vertically, the subjective and objective interaction is maximized. Extensive experiments on five benchmark datasets demonstrate that compared with 12 state-of-the-art methods, the algorithm has higher recall and precision, less error and strong robustness and generalization ability, and can predict complete and refined saliency maps.

An Improved Deadbeat Predictive Current Control Method for SPMSM Drives with a Novel Adaptive Disturbance Observer

Shuo Zhang, Lingding Lei, Chengning Zhang, Tian Liu, Shuli Wang

2023, 32(1): 107-123. doi:10.15918/j.jbit1004-0579.2022.080

[Abstract](25) [FullText HTML](11) [PDF 9867KB](5)

Abstract:
To improve the dynamic performance of conventional deadbeat predictive current control (DPCC) under parameter mismatch, especially eliminate the current overshoot and oscillation during torque mutation, it is necessary to enhance the robustness of DPCC against various working conditions. However, the disturbance from parameter mismatch can deteriorate the dynamic performance. To deal with the above problem, firstly, traditional DPCC and the parameter sensitivity of DPCC are introduced and analyzed. Secondly, an extended state observer (ESO) combined with DPCC method is proposed, which can observe and suppress the disturbance due to various parameter mismatch. Thirdly, to improve the accuracy and stability of ESO, an adaptive extended state observer (AESO) using fuzzy controller based on ESO, is presented, and combined with DPCC method. The improved DPCC-AESO can switch the value of gain coefficients with fuzzy control, accelerating the current response speed and avoid the overshoot and oscillation, which improves the robustness and stability performance of SPMSM. Finally, the three methods, as well as conventional DPCC method, DPCC-ESO method, DPCC-AESO method, are comparatively analyzed in this paper. The effectiveness of the proposed two methods are verified by simulation and experimental results.

A New Two-Stage Tunable Space-Time Adaptive Detector

Xiaojing Su, Da Xu, Dongsheng Zhu

2023, 32(1): 124-130. doi:10.15918/j.jbit1004-0579.2022.107

[Abstract](27) [FullText HTML](11) [PDF 3779KB](6)

Abstract:
In order to improve the rejection capability of mismatched interferer signals, a new two-stage detector is proposed under homogeneous scenarios with unknown covariance matrix, which is obtained by cascading the adaptive matched filter (AMF) detector and the enhanced RAO (EnRAO) detector. The new detector has constant false alarm performance, and the closed-form expression of probability of false alarm and probability of detection is derived. The performance of the new detector is assessed, and analyzed in comparison with other detectors. The results show that, the proposed detector can provide enhanced rejection capability in the case of mismatch, but the performance of the detector is slightly lost under the condition of matching.

Current Issue2023 Vol. 32, No. 1

Current Issue
2023 Vol. 32, No. 1