本期目录

2020年 27卷 4期 刊出日期：2020-08-31



Artificial Intelligence

Real-time prediction of the motion tendency of human lower limbs during gait

陈法权 樊军

中国邮电高校学报(英文), 2020, 27 (4). doi： 10.19682/j.cnki.1005-8885.2020.0031

摘要 ( 306 ) PDF (1655 KB)( 102 )

参考文献 | 相关文章 | 计量指标

Aiming at the problem of hysteresis in the human motion intention recognition algorithm based on kinematic sensors, a real-time prediction method about human lower limb motion tendency is proposed. It could be used to control exoskeleton robots, intelligent prosthes and other equipments in advance to eliminate the hysteresis of equipment movement. Firstly, the angle signals of ankle, knee and hip are segmented by the extreme points. Secondly, the multi-dimensional temporal association rules algorithm is used to analyze the angle signals to find out the relationships between signal patterns in adjacent time segments. Finally, the signal patterns at the next moment are predicted through the association rules algorithm, so as to predict the motion tendency of human lower limbs. Experimental results show that the proposed scheme achieves an average prediction accuracy of 78.3% for each signal segment, and can predict the subsequent motion of human lower limbs in average 92.24 ms.

Generative adversarial networks for hyperspectral image spatial super-resolution

蒋伊琳 邵然 唐三强

中国邮电高校学报(英文), 2020, 27 (4). doi： 10.19682/j.cnki.1005-8885.2020.0032

摘要 ( 321 ) PDF (8026 KB)( 156 )

参考文献 | 相关文章 | 计量指标

It is becoming increasingly easier to obtain more abundant supplies for hyperspectral images ( HSIs). Despite this, achieving high resolution is still critical. In this paper, a method named hyperspectral images super-resolution generative adversarial network ( HSI-RGAN ) is proposed to enhance the spatial resolution of HSI without decreasing its spectral resolution. Different from existing methods with the same purpose, which are based on convolutional neural networks ( CNNs) and driven by a pixel-level loss function, the new generative adversarial network (GAN) has a redesigned framework and a targeted loss function. Specifically, the discriminator uses the structure of the relativistic discriminator, which provides feedback on how much the generated HSI looks like the ground truth. The generator achieves more authentic details and textures by removing the place of the pooling layer and the batch normalization layer and presenting smaller filter size and two-step upsampling layers. Furthermore, the loss function is improved to specially take spectral distinctions into account to avoid artifacts and minimize potential spectral distortion, which may be introduced by neural networks. Furthermore, pre-training with the visual geometry group (VGG) network helps the entire model to initialize more easily. Benefiting from these changes, the proposed method obtains significant advantages compared to the original GAN. Experimental results also reveal that the proposed method performs better than several state-of-the-art methods.

Research on equipment fault diagnosis method based on random stochastic adaptive particle swarm optimization

杨健健 张强 王晓林 杜毅博 王超 吴淼

中国邮电高校学报(英文), 2020, 27 (4). doi： 10.19682/j.cnki.1005-8885.2020.0033

摘要 ( 295 ) PDF (5294 KB)( 98 )

参考文献 | 相关文章 | 计量指标

The traditional fault diagnosis method of industrial equipment has low accuracy and poor applicability. This paper proposes a equipment fault diagnosis method based on random stochastic adaptive particle swarm optimization (RSAPSO). The entire model is validated by using the data of healthy bearings collected by Case Western Reserve University. Different gradient descent algorithms and standard particle swarm optimization (PSO) algorithms in a back propagation (BP) network are compared experimentally. The results show that the RSAPSO algorithm has a higher accuracy of weight threshold updating than the gradient descent algorithm and does not easily fall into a local optimum. Compared with PSO, it has a faster optimization speed and higher accuracy. Finally, the RSAPSO algorithm is validated with the data of bearings collected from the laboratory rotating machinery test bench and motor data collected from the tower reflux pump. The average recognition rate of the four kinds of bearing data constructed is 97.5% , and the average recognition rate of the two kinds of motor data reaches 100% , which prove the universality of the method.

Independently recurrent neural network for remaining useful life estimation

王锴烨 崔绍华 许方敏 赵成林

中国邮电高校学报(英文), 2020, 27 (4). doi： 10.19682/j.cnki.1005-8885.2020.0034

摘要 ( 304 ) PDF (1309 KB)( 78 )

参考文献 | 相关文章 | 计量指标

In the industrial fields, the mechanical equipment will inevitably wear out in the process of operation. With the accumulation of losses, the probability of equipment failure is increasing. Therefore, if the remaining useful life (RUL) of the equipment can be accurately predicted, the equipment can be maintained in time to avoid the downtime caused by equipment failure and greatly improve the production efficiency of enterprises. This paper aims to use independently recurrent neural network (IndRNN) to learn health degradation of turbofan engine and make accurate predictions of its RUL, which not only effectively solves the problem of gradient explosion and vanishing, but also increases the interpretability of neural networks. IndRNN can be used to process longer time series which matches the scene with high frequency sampling sensor in industrial practical applications. The results demonstrate that IndRNN for RUL estimation significantly outperforms traditional approaches, as well as convolutional neural network (CNN) and long short-term memory network (LSTM) for RUL estimation.

Enhancement network: confining deep convolutional features based on hand-crafted rules

李艳华 肖文光

中国邮电高校学报(英文), 2020, 27 (4). doi： 10.19682/j.cnki.1005-8885.2020.0035

摘要 ( 301 ) PDF (2844 KB)( 93 )

参考文献 | 相关文章 | 计量指标

Despite convolutional neural network ( CNN) is mature in many domains, the understanding of the directions where the parameters of the CNNs are learned towards, falls behind, and researches on the functions that the convolutional networks (ConvNets) learns are difficult to be explored. A method is proposed to guide ConvNets to learn towards the expected direction. First, for the sake of facilitating network converging, a novel feature enhancement framework, namely enhancement network (EN), is devised to learn parameters according to certain rules. Second, two types of hand-crafted rules, namely feature-sharpening (FS) and feature-amplifying (FA) are proposed to enable effective ENs, meanwhile are embedded into the CNN for the end-to-end learning. Specifically, the former is a tool sharpening convolutional features and the latter is the one amplifying convolutional features linearly. Both tools aim at the same spot achieving a stronger inductive bias and more straightforward loss functions. Finally, the experiments are conducted on the mixed National Institute of Standards and Technology (MNIST) and cooperative institute for Alaska research 10 (CIFAR10) dataset. Experimental results demonstrate that ENs make a faster convergence by formulating hand-crafted rules.

Wireless

Reduced-state noncoherent MLSD method for CPM in aeronautical telemetry

周游 段瑞枫 蒋伯峰

中国邮电高校学报(英文), 2020, 27 (4). doi： 10.19682/j.cnki.1005-8885.2020.0036

摘要 ( 287 ) PDF (858 KB)( 123 )

参考文献 | 相关文章 | 计量指标

Due to high spectral efficiency and power efficiency, the continuous phase modulation ( CPM) technique with constant envelop is widely used in range telemetry. How to improve the bit error rate (BER) performance of CPM and keep a reasonable computational complexity is the key of the entire telemetry system and the focus of research and engineering design. In this paper, a reduced-state noncoherent maximum likelihood sequence detection (MLSD) method for CPM is proposed. In the proposed method, the criterion of noncoherent MLSD is derived for CPM when the carrier phase is unknown. A novel Viterbi algorithm (VA) with modified state vector is designed to simplify the implementation of noncoherent MLSD. Both analysis and numerical results show that the proposed method reduces the computational complexity significantly and does not need accurate carrier phase recovery, which overcomes the shortage of traditional MLSD method. Additionally, the proposed method exceeds the traditional MLSD method when carrier phase deviation exists.

UWB positioning enhancement using Markov chain in indoor NLOS environment

李绍贤 吕凡 王成瑞 侯延昭

中国邮电高校学报(英文), 2020, 27 (4). doi： 10.19682/j.cnki.1005-8885.2020.0027

摘要 ( 387 ) PDF (1262 KB)( 101 )

参考文献 | 相关文章 | 计量指标

The accuracy of the positioning system in indoor environment is often affected by none-line-of-sight ( NLOS) propagation. In order to improve the positioning accuracy in indoor NLOS environment, a method used ultra-wide-band ( UWB ) technology, which based on time of arrival ( TOA) principle, combining Markov chain and fingerprint matching was proposed. First, the TOA algorithm is used to locate the target tag. Then the Markov chain is used to identify if blocking happened and revise the position result. And the fingerprint matching is used to further improve the position accuracy. Finally, an experiment system was built to test the accuracy of the proposed method and the traditional Kalman filter method. The experimental results show that, compared with the traditional Kalman filter method, the proposed method can improve the positioning accuracy in indoor NLOS environment.

VMScan: an out-of-VM malware scanner

林杰 刘川意 方滨兴

中国邮电高校学报(英文), 2020, 27 (4). doi： 10.19682/j.cnki.1005-8885.2020.0037

摘要 ( 378 ) PDF (1974 KB)( 76 )

参考文献 | 相关文章 | 计量指标

The harm caused by malware in cloud computing environment is more and more serious. Traditional anti-virus software is in danger of being attacked when it is deployed in virtual machine on a large scale, and it tends not to be accepted by tenants in terms of performance. In this paper, a method of scanning malicious programs outside the virtual machine is proposed, and the prototype is implemented. This method transforms the memory of the virtual machine to the host machine so that the latter can access it. The user space and kernel space of virtual machine memory are analyzed via semantics, and suspicious processes are scanned by signature database. Experimental results show that malicious programs can be effectively scanned outside the virtual machine, and the performance impact on the virtual machine is low, meeting the needs of tenants.

Continual auxiliary leakage-resilient attribute-based broadcast encryption with constant size ciphertexts

商玉洁 张乐友 高小旭

中国邮电高校学报(英文), 2020, 27 (4). doi： 10.19682/j.cnki.1005-8885.2020.0038

摘要 ( 261 ) PDF (699 KB)( 85 )

参考文献 | 相关文章 | 计量指标

Attribute-based broadcast encryption ( ABBE) under continual auxiliary leakage-resilient ( CALR) model can enhance the security of the shared data in broadcasting system since CALR model brings the possibility of new leakage-resilient (LR) guarantees. However, there are many shortcomings in the existing works, such as relying on the strong assumptions, low computational efficiency and large size of ciphertexts, etc. How to solve the trade-off between security and efficiency is a challenging problem at present. To solve these problems, this paper gives an ABBE scheme resisting continual auxiliary leakage ( CAL ) attack. ABBE scheme achieves constant size ciphertexts, and the computational complexity of decryption only depends on the number of receivers instead of the maximum number of receivers of the system. Additionally, it achieves adaptive security in the standard model where the security is reduced to the general subgroup decision (GSD) assumptions (or called static assumptions in the subgroup). Furthermore, it can tolerate leakage on the master secret key and private key with continual auxiliary inputs. Performance analysis shows that the proposed scheme is more efficient and practical than the available schemes.

Others

Hybrid time slot allocation algorithm based on LoRa Internet of things

姚引娣 王磊 贺军瑾

中国邮电高校学报(英文), 2020, 27 (4). doi： 10.19682/j.cnki.1005-8885.2020.0039

摘要 ( 404 ) PDF (904 KB)( 81 )

参考文献 | 相关文章 | 计量指标

To improve the efficiency and stability of data transmission in the long-range (LoRa) Internet of things (IoT),a hybrid time slot allocation algorithm is proposed, which implements a priority mechanism with high-priority nodes sending data in fixed time slots and low-priority nodes using the carrier sense multiple access (CSMA) algorithm to compete for shared time slots to transmit data. To improve network efficiency, a gateway is used to adjust the time slot allocation policy according to network status and balance the number of fixed and shared time slots. And more, a retransmission time slot is added to the time slot allocation algorithm, which redesigns the time frame structure, and adopts a retransmission mechanism to improve communication reliability. Simulation and measurement results show that the packet loss rate and transmission delay of the proposed hybrid algorithm are smaller than those of the fixed slot allocation algorithm, making the proposed algorithm more suitable for LoRa IoT.

Monocular camera and 3D lidar joint calibration

郑鑫 吴晓军

中国邮电高校学报(英文), 2020, 27 (4). doi： 10.19682/j.cnki.1005-8885.2020.0040

摘要 ( 343 ) PDF (5385 KB)( 106 )

参考文献 | 相关文章 | 计量指标

Joint calibration of sensors is an important prerequisite in intelligent driving scene retrieval and recognition. A simple and efficient solution is proposed for solving the problem of automatic joint calibration registration between the monocular camera and the 16-line lidar. The study is divided into two parts: single-sensor independent calibration and multi-sensor joint registration, in which the selected objective world is used. The system associates the lidar coordinates with the camera coordinates. The lidar and the camera are used to obtain the normal vectors of the calibration plate and the point cloud data representing the calibration plate by the appropriate algorithm. Iterated closest points (ICP) is the method used for the iterative refinement of the registration.