Xiang Zhu

Dr. Xiang Zhu

National Space Science Center, Beijing, China
orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
7 November 2022
Verified
Conference Paper
Open access
SSSijie Shen
Sijie Shen
XZXiang Zhu
Xiang Zhu
YDYihong Dong
Yihong Dong
QGQizhi Guo
Qizhi Guo
YZYankun Zhen
Yankun Zhen
GLGe Li
Ge Li
Code generation aims to generate a code snippet automatically from natural language descriptions. Generally, the mainstream code generation methods rely on a large amount of paired training data, including both the natural language description and the code. However, in some domain-specific scenarios, building such a large paired corpus for code generation is difficult because there is no directly available pairing data, and a lot of effort is required to manually write the code descriptions to construct a high-quality training dataset. Due to the limited training data, the generation model cannot be well trained and is likely to be overfitting, making the model's performance unsatisfactory for real-world use. To this end, in this paper, we propose a task augmentation method that incorporates domain knowledge into code generation models through auxiliary tasks and a Subtoken-TranX model by extending the original TranX model to support subtoken-level code generation. To verify our proposed approach, we collect a real-world code generation dataset and conduct experiments on it. Our experimental results demonstrate that the subtoken-level TranX model outperforms the original TranX model and the Transformer model on our dataset, and the exact match accuracy of Subtoken-TranX improves significantly by 12.75% with the help of our task augmentation method. The model performance on several code categories has satisfied the requirements for application in industrial systems. Our proposed approach has been adopted by Alibaba's BizCook platform. To the best of our knowledge, this is the first domain code generation system adopted in industrial development environments.

20 January 2023 in Electronics
Verified
Journal Article
Open access
PDF file
JXJindou Xin
Jindou Xin
XZXiang Zhu
Xiang Zhu
YMYingqi Ma
Yingqi Ma
JHJianwei Han
Jianwei Han
Bulk silicon Complementary Metal Oxide Semiconductor (CMOS) devices have distinct single event latch-up (SEL) problems in aerospace. Therefore, it is essential that CMOS devices are designed with appropriate circuit-level methods. Traditional resistor hardness satisfies the current aerospace trend of low cost, high performance, and miniaturization. Therefore conventional resistor hardness is often applied in circuit-level designs due to the reduction of latch-up current. In circuits containing a DC-DC buck converter, the resistor is connected to the back of the converter in the traditional method. However, the traditional method is unable to take devices out of the latch-up owing to the small resistance range. To solve this problem, the paper proposes an improved design for the resistor in front of the DC-DC buck converter. The proposed method enables the devices to exit the latch-up by increasing the resistance range according to the input characteristic of the DC-DC buck converter. The paper quantifies the range of the resistor through the parametric model containing the resistor and the DC-DC buck converter. Two CMOS devices are chosen for pulsed laser experiments, verifying that the proposed method increases the resistance ranges by 300% to 400% compared to the conventional method. It is also demonstrated that the proposed method exits the devices from latch-up within the resistor ranges. That is, the resistance ranges of 34 Ω~41 Ω and 51 Ω~56 Ω reduce the latch-up currents of the devices to below holding currents of 72.1 mA and 24.2 mA, respectively.

22 June 2023 in Electronics
Verified
Journal Article
Open access
PDF file
ZMZiqi Mai
Ziqi Mai
XZXiang Zhu
Xiang Zhu
HLHongwei Li
Hongwei Li
JHJianwei Han
Jianwei Han
THTao He
Tao He
Single Event Functional Interrupt (SEFI) poses a severe threat to the normal operation of spacecraft. This paper investigates SEFI in Analog-to-Digital Converters (ADCs) with storage units using precision positioning of pulsed lasers. Based on the experiment, it was discovered that a bit flip in the configuration registers in ADCs results in changes in parameters such as digital filter frequency, operating mode, and gain, leading to an upward or downward offset of ADC output codes. Similarly, a bit flip in the calibration registers also causes ADC output codes to shift upwards or downwards, or even output a value of zero. Furthermore, it was observed that SEFI phenomena can occur due to current latch-up in ADC input pins, causing the inability to read or write data in ADC storage units. This current latch-up can be resolved through power cycling or configuring the pins into a high-impedance state. This work highlights the significance of SEFI phenomena in ADCs, emphasizing the serious threat posed by storage unit flipping-induced SEFI to the proper functioning of ADCs. Moreover, the SEFI phenomenon caused by current latch-up in input pins is difficult to detect in practice, making it highly elusive. Once it occurs, it severely impacts the functionality of ADCs.

26 December 2024 in IEEE Transactions on Nuclear Science
Verified
Journal Article
Closed access
JWJieyi Wang
Jieyi Wang
HZHuanxin Zhang
Huanxin Zhang
XZXinyu Zhu
Xinyu Zhu
GSGuohong Shen
Guohong Shen
ZCZheng Chang
Zheng Chang
YMYingqi Ma
Yingqi Ma
This study investigates the dynamics of Single Event Upsets (SEUs) in space environments as detected by the NSSC’s the Space Particle Radiation Effect Comprehensive Measuring Instrument (SPRECMI), which comprises Linear Energy Transfer (LET) and Proton detectors along with an SEU Monitor. Positioned at an altitude of 1,025 km and an inclination of 63.5°, the in-flight data reveal that inducing factors from the South Atlantic Anomaly (SAA) are predominant contributors to SEUs with 97.34% of SEUs occurring within this region despite its limited exposure time of 12.75% of the daily orbit. Our results not only underscore the dominant role of protons in influencing SEU rates but also validate the accuracy of SEU rate calculations derived from particle detection over long-term periods achieving a close agreement with a relative error of 7.01%. Nevertheless, the study finds limited daily-scale correlational dynamics between SEU rates and individual inducing factors, indicating the need for further research into optimal temporal scales for analyzing these relationships. Moreover, this study presents a systematic data processing and analysis case using comprehensive in-flight detection data and provides experimental evidence for the inducing factors of SEU.

23 December 2022 in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Verified
Journal Article
Closed access
HLHaoyuan Li
Haoyuan Li
HCHua Chen
Hua Chen
XHXucang Han
Xucang Han
WCWeiqiong Cao
Weiqiong Cao
XZXiang Zhu
Xiang Zhu
JWJian Wang
Jian Wang
ECC is a widely used public key cryptosystem. Though it has good theoretical security, it is still vulnerable to some physical attacks due to the implementation weakness. To resist the attacks, a number of physical countermeasures have been proposed. However, there are still some implementation vulnerabilities that may be overlooked, leading to more practical and effective attacks. In this paper, we construct a new fault attack on round counter which is a component of scalar multiplications in ECDSA. The attack is divided into two parts. In the first part, the partial bits of nonce in each signature can be recovered by the fault injection on round counter. In the second part, an efficient lattice attack can be constructed to recover the private key by combining the recovered bits. Compared with other lattice-based fault attacks, our attack has the advantage of practicability and effectiveness. Especially it has less requirement of moment precision and wide applicability of scalar multiplications, which is the critical factors for practicability and effectiveness. To verify the strength of our attack, we carry on the laser injection experiments respectively on an AVR MCU(ATmega163L) and a Kintex-7 FPGA(XC7K325T). The experiment results verify the practicability and effectiveness of the attack in both software and hardware platform. Finally, we also propose two directions for efficient countermeasures against our attack.

19 October 2022 in Symmetry
Verified
Journal Article
Open access
PDF file
HJHuilong Jiang
Huilong Jiang
XZXiang Zhu
Xiang Zhu
JHJianwei Han
Jianwei Han
A Fault Attack (FA) is performed mainly under the data corruption model and poses a threat to security chips. Instruction corruption can enact the same purpose at the behavioral level, which is produced by interfering with the instruction system. Laser Fault Injection (LFI) on program memory during the instruction-fetching process, which we refer to as an instruction-fetching attack, is studied in this paper. This process bears the ability to produce a controllable instruction-fetching fault. Our work shows the implementation of the attack and its specific application case on an 8-bit microcontroller. The main contributions of this paper include: (1) We have mapped the sensitive areas precisely to the faulted instructions via laser injection and implemented controllable instruction tampering. (2) A Collision Fault Attack (CFA) scheme based on instruction-fetching fault is proposed. (3) The impacts of the faulted instructions are fully explored, including the influence on subsequent operations and key recovery. (4) The fault mechanism of the on-chip Flash is further investigated. Instruction-fetching fault means that the controller fetches a tampered instruction from the program memory under external interference, which likely gives rise to an invalid or incorrect operation. The experiment confirms that this specific fault can induce particular types of faults that are different to realize, e.g., the byte-fault model in CFA. The realization, application and mechanism of instruction-fetching fault are discussed in detail.