Loading...
首页
期刊介绍
期刊订阅
联系我们
快速检索
引用检索
图表检索
高级检索
最新录用
|
预出版
|
当期目录
|
过刊浏览
|
阅读排行
|
下载排行
|
引用排行
|
百年期刊
ISSN 1000-0585
CN 11-1848/P
Started in 1982
About the Journal
»
About Journal
»
Editorial Board
»
Indexed in
»
Rewarded
Authors
»
Online Submission
»
Guidelines for Authors
»
Templates
»
Copyright Agreement
Reviewers
»
Guidelines for Reviewers
»
Online Peer Review
Office
»
Editor-in-chief
»
Office Work
»
Production Centre
Table of Content
, Volume 61 Issue 11
Previous Issue
Next Issue
For Selected:
View Abstracts
Download Citations
EndNote
Reference Manager
ProCite
BibTeX
RefWorks
Toggle Thumbnails
VULNERABILITY ANALUSIS AND RISK ASSESSMENT
Select
exLCL for defense against spectre attacks
WANG Shaoqing, ZHAO Youjian, L�Zhiyuan
Journal of Tsinghua University(Science and Technology). 2021,
61
(11): 1221-1227. DOI: 10.16511/j.cnki.qhdxxb.2020.22.040
Abstract
HTML
PDF
(1172KB) (
373
)
The newly discovered spectre attack poses severe challenges to computer security. The attacker leaks secret data by exploiting the indelible micro-architecture (such as cache) state changes left by speculative execution commands combined with the cache side channels. This paper first describes the instruction execution process of the spectre attack, presents a stage model for the attack, and identifies the competition conditions when a vulnerability can be exploited. Then, a defense entitled exLCL (extended L1 cache latency) is presented for preventing an attacker from meeting the competition conditions. Simulations based on gem5 show the effectiveness and feasibility of the exLCL defense which has simpler logic than existing defenses.
Figures and Tables
|
References
|
Related Articles
|
Metrics
Select
Efficient semantic-based multiuser search scheme
CAO Laicheng, WU Qirui, WANG Yafei, WU Rong, GUO Xian
Journal of Tsinghua University(Science and Technology). 2021,
61
(11): 1228-1233. DOI: 10.16511/j.cnki.qhdxxb.2020.22.45
Abstract
HTML
PDF
(1860KB) (
187
)
Current cloud storage searchable encryption schemes have mostly been developed for multiuser environments with multi-keyword searches, but research on these schemes have not considered the influence of the semantics on the search results. This paper presents an efficient semantic-based multiuser search scheme using two cloud servers to ensure efficient searches. One cloud server matches the trapdoor and generates results that match the semantics while the other searches for the most relevant files and returns them to the user. This coordination reduces the server load by sending the semantic tree search to the cloud server which uses the trapdoor strategy. The semantic tree is encrypted to ensure the key information security. Tests show that this scheme is effective, resists the collusion attack, provides semantic privacy, and is more efficient than current related schemes.
Figures and Tables
|
References
|
Related Articles
|
Metrics
Select
Attack detection and security state estimation of cyber-physical systems
GAO Yang, REN Wang, WU Renpu, WANG Weiping, YI Shengwei, HAN Baijing
Journal of Tsinghua University(Science and Technology). 2021,
61
(11): 1234-1239. DOI: 10.16511/j.cnki.qhdxxb.2021.21.008
Abstract
HTML
PDF
(1777KB) (
268
)
In the Industry 4.0 context, cyber-physical systems (CPS) need effective security and control capabilities. This study analyzes attack detection and security state estimation based on a cyber-physical system model attacked by an actuator. The attack detection uses a finite time attack detector that ensures that existing attacks can be accurately detected within a preset time limit. An observer is then designed to estimate the system state. A theoretical analysis shows that the observer ensures that the system can immediately adjust when attacked and that the system will reach a safe, stable state. Simulations verify the effectiveness of this method.
Figures and Tables
|
References
|
Related Articles
|
Metrics
Select
Data generation and annotation method for source code defect detection
GUAN Zhibin, WANG Xiaomeng, XIN Wei, WANG Jiajie
Journal of Tsinghua University(Science and Technology). 2021,
61
(11): 1240-1245. DOI: 10.16511/j.cnki.qhdxxb.2021.21.005
Abstract
HTML
PDF
(2493KB) (
340
)
Existing deep learning based source code vulnerability detection methods use training and test data sets that are mostly derived from test source codes for academic research only which do not provide sufficient support for training of deep learning models. This paper presents a data generation and annotation method for source code defect detection. This method extracts the source code control flow relationships and uses trained deep learning models and commercial tools to complete the slice data annotation of the source code. The public data sets SARD, NVD and the open-source code Ffmpeg are utilized to verify the system performance. The results show that this method can generate a source code defect dataset for deep learning to support deep learning-based source code vulnerability detection methods.
Figures and Tables
|
References
|
Related Articles
|
Metrics
Select
Secure and energy efficient offloading of mobile edge computing in the Internet of vehicles
SONG Yubo, JING Xingyu, YAN Feng, HU Aiqun
Journal of Tsinghua University(Science and Technology). 2021,
61
(11): 1246-1253. DOI: 10.16511/j.cnki.qhdxxb.2021.25.005
Abstract
HTML
PDF
(1649KB) (
224
)
Task offloading negotiations between vehicles and multiple edge servers in mobile edge computing environments of the internet of vehicles have security problems and excessive system energy consumption. This paper presents a security offloading mechanism based on cooperation between the edge server task offload strategy and the vehicle server. The strategy has a secure switching interaction protocol when the vehicle is moving with a task assignment negotiation algorithm and constraints based on the edge server coverage. Simulations show that the scheme effectively guarantees security during communications while reducing the offloading energy consumption by 58% and the offloading time by 17% compared with an existing scheme.
Figures and Tables
|
References
|
Related Articles
|
Metrics
Select
Browser user tracking based on cross-domain resource access
SONG Yubo, WU Tianqi, HU Aiqun, GAO Shang
Journal of Tsinghua University(Science and Technology). 2021,
61
(11): 1254-1259. DOI: 10.16511/j.cnki.qhdxxb.2021.25.003
Abstract
HTML
PDF
(2713KB) (
91
)
In recent years, click fraud has caused huge economic losses to advertisers. Many advertisers have then used "user profiles" to identify users to eliminate click fraud. However, attackers can easily construct unique virtual operating environments to confuse the identification algorithms. This paper introduces a localization scheme to detect click fraud sources based on cross-domain resource access. This scheme extracts features from a ping response delay series to fingerprint users. Tests show that the delay features collected by this method are stable with a fingerprint localization accuracy of up to 98%.
Figures and Tables
|
References
|
Related Articles
|
Metrics
Select
Security analysis of 5G authentication and key agreement protocol
JIA Fan, YAN Yan, YUAN Kaiguo, ZHAO Lujing
Journal of Tsinghua University(Science and Technology). 2021,
61
(11): 1260-1266. DOI: 10.16511/j.cnki.qhdxxb.2021.26.001
Abstract
HTML
PDF
(1352KB) (
312
)
5G networks are developing rapidly. The security of the authentication and key agreement protocols is the core issue of 5G security. The TAMARIN prover is used here to analyze the EAP-AKA' protocol for 5G networks. The protocol specifications are analyzed to identify the security requirements as a confidential attribute and an authentication attribute. A model is then established according to the TAMARIN standard to verify that these security attributes are satisfied. The verification results show a single shot consistency violation between SEAF and AUSF regarding SNID and a violation of the forward secrecy of the anchor key K
SEAF
, which may lead to the network experiencing replay attacks, authentication synchronization failure attacks, and Anchor key K
SEAF
leak attacks. Security hardening methods are then presented for these attacks with theoretical and experimental verification.
Figures and Tables
|
References
|
Related Articles
|
Metrics
Select
Source code defect detection using deep convolutional neural networks
WANG Xiaomeng, GUAN Zhibin, XIN Wei, WANG Jiajie
Journal of Tsinghua University(Science and Technology). 2021,
61
(11): 1267-1272. DOI: 10.16511/j.cnki.qhdxxb.2020.26.042
Abstract
HTML
PDF
(1152KB) (
334
)
Deep learning-based source code defect detection looks at the source code as text data. The defect detection then uses a one-dimensional convolutional network to learn the single spatial characteristics of the code or uses the sequential characteristics of LSTM and BiLSTM which do not take various features of the source code into account. This article uses the multi-channel learning strategy of convolutional neural networks for image classification to identify multi-class source code defects by deep convolutional neural networks. First, a word embedding algorithm such as word2vec or fasttext is used to construct the fusion features with the deep convolutional neural network then used to identify the defect patterns contained in the source code defect data set to form a source code defect classifier. The classifier is then used to recognize defect codes and their corresponding CWE type. The method was evaluated on the SARD dataset and open source software. The results show that this method is superior to existing methods with a model evaluation parameter accuracy of 95.3%, a recall rate of 84.7%, and F
1
of 89.7%.
Figures and Tables
|
References
|
Related Articles
|
Metrics
MECHANICAL ENGINEERING
Select
Impact toughness variations of multi-pass, multi-layer CrMoV submerged arc welding weld metal
TIAN Hongwei, LU Li, ZHAO Chongyang, SHI Xiangwen, LI Kejian
Journal of Tsinghua University(Science and Technology). 2021,
61
(11): 1273-1280. DOI: 10.16511/j.cnki.qhdxxb.2021.22.004
Abstract
HTML
PDF
(24731KB) (
120
)
After post-welding heat treatment, the impact toughness of CrMoV weld metal from multi-pass, multi-layer submerged arc welding fluctuated greatly at room temperature. The microstructures and crack propagation paths of various samples were characterized using optical and scanning electron microscopes with the fracture surface morphology observed using a scanning electron microscope. In addition, the carbides in the matrix were analyzed using an Auger electron spectroscope. The results show that the reheated coarse-grained zone of the weld metal is the weakest microstructure component for the impact toughness and that the necklace-like distributed Fe
3
C carbides and nonmetallic inclusions can lead to crack initiation. The location of the V-shaped notch in the weld metal is the key factor determining the impact absorbed energy. A larger distance between the V-shaped notch and the reheated coarse-grained zone in the direction along the crack propagation path increases the weld metal impact toughness.
Figures and Tables
|
References
|
Related Articles
|
Metrics
Select
Digital product twin modeling of massive dynamic data based on a time-series database
ZHENG Menglei, TIAN Ling
Journal of Tsinghua University(Science and Technology). 2021,
61
(11): 1281-1288. DOI: 10.16511/j.cnki.qhdxxb.2021.26.006
Abstract
HTML
PDF
(1298KB) (
320
)
The integrated modeling of massive dynamic data sets is a key basic technology that needs to be solved for digital product twins. The data modeling target is identified by analyzing the characteristics of various data types for the digital product twin. The features of various data storage modes were studied here to develop a digital product twin modeling method for massive dynamic data sets based on a time-series database. The structure, attributes and scale characteristics of the dynamic product data are used to improve the database performance. A case analysis of rolling bearing vibration data sets is used to show how this method works with massive dynamic data sets.
Figures and Tables
|
References
|
Related Articles
|
Metrics
Select
Reliability analysis of the roller base of a counter-roller spinning machine based on the maximum entropy method
ZHENG Hongwei, MENG Guangwei, LI Feng, HUANG Tao, GUO Yaming, XUAN Mohan
Journal of Tsinghua University(Science and Technology). 2021,
61
(11): 1289-1294. DOI: 10.16511/j.cnki.qhdxxb.2020.26.031
Abstract
HTML
PDF
(10157KB) (
100
)
The maximum entropy method was used to develop a reliability analysis model for the roller base of a counter-roller spinning machine. The maximum entropy method was used to define the probability density expression for the performance function with the fourth-order central moments then computed using a Taylor series expansion and a finite element analysis. The fourth-order origin moments were then calculated by a standardized transformation. MATLAB was then used to solve the constrained expression for the moments to obtain the maximum entropy probability density function parameters. The parameters were fit to get the probability density curve with the failure probability of the structural function then found by integration. The error relative to a Monte Carlo simulation (MCS) was 1.61%, which shows that this method is accurate. The roller base was then optimized which reduced the failure probability after optimization by 11.42%.
Figures and Tables
|
References
|
Related Articles
|
Metrics
NUCLEAR ENERGY AND NEW ENERGY
Select
Nuclear power to improve the endurance of unmanned aerial vehicles
ZHU Bingyu, DING Qingqing
Journal of Tsinghua University(Science and Technology). 2021,
61
(11): 1295-1300. DOI: 10.16511/j.cnki.qhdxxb.2020.26.030
Abstract
HTML
PDF
(2156KB) (
173
)
Endurance is an important indicator of aircraft performance. Nuclear energy has extremely high energy densities and is expected to significantly improve the endurance of unmanned aerial vehicles. This paper takes a certain type of UAV in the world as a prototype. And through analysis and comparison of nuclear thermal propulsion core schemes, a core scheme suitable for the core module of the nuclear power system of the UAV is designed to replace the original power system engine, the core module combustion chamber of the engine. Then this paper analyses the feasibility of the core module of the engine to be equipped with nuclear power, and the improvement of the endurance of the UAV driven by nuclear power. The results show that the nuclear power reactor core weighs 1 504.75 kg, requires a 3% enrichment degree of UC 119 kg, and the fuel working temperature is lower than 2 500 K, which meets the reactor physics and thermal engineering requirements. The core lifetime is nearly 170 days, which can significantly improve the endurance of unmanned aerial vehicles.
Figures and Tables
|
References
|
Related Articles
|
Metrics
Select
Core temperature distributions in HTR-10 operating at very high temperatures
SUN Shiyan, ZHANG Youjie, ZHENG Yanhua, XIA Bing
Journal of Tsinghua University(Science and Technology). 2021,
61
(11): 1301-1307. DOI: 10.16511/j.cnki.qhdxxb.2020.22.036
Abstract
HTML
PDF
(4237KB) (
352
)
The 10 MW high temperature gas-cooled reactor test module (HTR-10) has a relatively large fuel temperature margin when operating at full power and steady state for the rated condition; thus, the coolant outlet temperature can be further increased above 700℃, which is useful for pebble-bed very high temperature gas-cooled reactors (HTGR) to operate with high coolant temperature at the core outlet. Several steady-state very high temperature operating conditions were designed for HTR-10 in this study based on a heat balance between the core heating power and the heat carried by the coolant. An improved thermal hydraulic model was used to analyze the core temperature distribution at each condition for the initial core to determine the temperature distribution characteristics, and the maximum core temperature for the nonuniform fuel distribution was discussed. For the coolant outlet temperature of 1 000℃, the predicted maximum core temperature is still below the fuel temperature limit, even when the most serious nonuniform distribution of the fuel and graphite balls occurs.
Figures and Tables
|
References
|
Related Articles
|
Metrics
Select
Radionuclide identification method based on a gamma-spectra template library simulated by radial basis function neural networks
DU Xiaochuang, TU Hongbing, LI Ke, ZHANG Jie, WANG Kang, LIU Hemin, LIANG Manchun, WANG Xiangwei
Journal of Tsinghua University(Science and Technology). 2021,
61
(11): 1308-1315. DOI: 10.16511/j.cnki.qhdxxb.2020.22.033
Abstract
HTML
PDF
(4987KB) (
153
)
The traditional gamma ray spectrum analysis methods are usually computationally complex and time-consuming. This paper simulated 26 radionuclide spectra using Geant4 to develop a gamma-spectra template library by fitting the spectra with radial basis function neural networks. Unknown gamma ray spectra were then identified using a least-squares algorithm and a nonlinear programing algorithm to find the optimal combination of radionuclide spectra in the library that matched the unknown spectrum with ensemble learning used to integrate the results of the two algorithms for identification. A single spectrum and mixed spectra containing 2 or 3 kinds of radionuclides were generated to evaluate the method. The results show that this method can accurately identify the radionuclides in an efficient and effective way.
Figures and Tables
|
References
|
Related Articles
|
Metrics
CIVIL ENGINEERING
Select
Pavement roughness index measurements with a 3D line laser
GUO Runhua, XU Li, PENG Huiting, CAO Zhijian, GENG Jingjie
Journal of Tsinghua University(Science and Technology). 2021,
61
(11): 1316-1324. DOI: 10.16511/j.cnki.qhdxxb.2020.26.041
Abstract
HTML
PDF
(11749KB) (
156
)
This paper describes the longitudinal layout and working principle of a 3D line laser profilometer with a moving reference algorithm based on the line laser design which can continuously measure the longitudinal contour of a track plane at 20~100 km/h. The device is placed on the left and right wheel track belts. The line laser survey path is 270 mm long and the vertical accuracy is 0.01 mm. The system was used to measure two sections of different quality asphalt pavement contours over distances of 10 m and 100 m to determine the international roughness index (IRI) for each section. The results show that the IRI at various speeds had a relative measurement error less than 3%. At low speeds, the left and right contour graph standard deviations were less than 0.3 and less than 0.6 at higher variable speeds. Measurements using a digital test vehicle gave poor results for quick starts with distorted data over the first 50 m. Thus, the initial values for the standard gold car model were taken from the first quarter of the dynamic response model results as the initial value. For the right contour graph, 91.4% of the data had coefficients of variation less than 5% and 8.6% of the coefficients of variation were less than 15%. For the left contour graph, 94.2% of the coefficients of variation were less than 5% and 5.8% of the coefficients of variation were less than 10%. The IRI results at sites 1 and 2 were in line with the actual service conditions of the road surfaces. The evaluations show that this dynamic monitoring system can accurately measure the IRI at both low speeds and variable speeds. This system can be used for pavement flatness surveys and pavement service status assessments.
Figures and Tables
|
References
|
Related Articles
|
Metrics
HYDRAULIC ENGINEERING
Select
Numerical simulations of the effect of leading edge cavitation on the vibration characteristics of an elastic hydrofoil
YAO Zhifeng, LAI Guihua, LIU Jing, ZENG Yongshun
Journal of Tsinghua University(Science and Technology). 2021,
61
(11): 1325-1333. DOI: 10.16511/j.cnki.qhdxxb.2020.22.043
Abstract
HTML
PDF
(9600KB) (
153
)
When hydraulic machinery operating conditions differ from the design conditions, leading edge cavitation can significantly disrupt the flow field and increase fluid-solid interactions (FSI) between the flow and the blade which change the system vibration characteristics. The flow around a NACA0009 elastic hydrofoil was simulated using the shear stress transfer (SST) turbulence model and the Zwart-Gerber-Belamri (ZGB) cavitation model in the three dimensional Reynolds average numerical simulation (RANS) method.The damped natural frequency and the added damping ratio were then calculated using a two-way FSI numerical method. The results were verified against existing experimental data. The computational results show that both the length and thickness of the cavitation region expand and the boundary layer on the hydrofoil leading edge grows thicker as the cavitation number decreases. The increasing cavitation length then reduces the shedding vortex frequency from the hydrofoil trailing edge. The leading edge cavity has little influence on the vibration shape of the bending mode, but greatly influences the vibration shape of the torsional mode. Enlarging the cavitation region increases the natural frequencies of the bending and torsional modes of the elastic hydrofoil in the same way, while the effects of the added damping ratio depend on the vibration mode. At an attack angle of 2° and a flow rate of 14 m/s, the natural frequency of the hydrofoil in the bending mode with leading edge cavitation increases up to 2.25% and the added damping ratio decreases by 9.00%, while the natural frequency in the torsional mode increases by 20.12% and the added damping ratio increases by 165.70%.
Figures and Tables
|
References
|
Related Articles
|
Metrics
ENVIRONMENTAL SCIENCE AND ENGINEERING
Select
Improved scatterometer sea surface wind speed predictions during rain
JIANG Zhuhui, ZHANG Wei
Journal of Tsinghua University(Science and Technology). 2021,
61
(11): 1334-1340. DOI: 10.16511/j.cnki.qhdxxb.2020.26.033
Abstract
HTML
PDF
(1522KB) (
97
)
The accuracy of sea surface wind speed measurements by spaceborne ASCAT and QUIKSCAT scatterometers during rain is improved in this study using spatial matching of the measured scatterometer sea surface wind speeds with in-situ buoy data for various spatial windows determined to be the optimal window angle. Then, the ASCAT and QUIKSCAT scatterometer sea surface wind speed errors were analyzed for typical rain rates to develop a more accurate scatterometer wind speed model during rain. Tests show that the ASCAT wind speed errors decrease from 2.5 m/s to 1.4 m/s, while the QUIKSCAT wind speed errors decrease from 4.3 m/s to 2.2 m/s.
Figures and Tables
|
References
|
Related Articles
|
Metrics
News
More
»
aaa
2024-12-26
»
2023年度优秀论文、优秀审稿人、优秀组稿人评选结果
2023-12-12
»
2022年度优秀论文、优秀审稿人、优秀组稿人评选结果
2022-12-20
»
2020年度优秀论文、优秀审稿人评选结果
2021-12-01
»
aa
2020-11-03
»
2020年度优秀论文、优秀审稿人评选结果
2020-10-28
»
第十六届“清华大学—横山亮次优秀论文奖”暨2019年度“清华之友—日立化成学术交流奖”颁奖仪式
2020-01-17
»
a
2019-01-09
»
a
2018-12-28
»
a
2018-01-19
Links
More
Copyright © Journal of Tsinghua University(Science and Technology), All Rights Reserved.
Powered by Beijing Magtech Co. Ltd