Yaoxin Duan

Yaoxin Duan

Auto-generated profile, based on publicly available information

Yaoxin Duan

Key Laboratory of Industrial Internet of Things and Networked Control, Ministry of Education; the School of Automation/School of Industrial Internet, Chongqing University of Posts and Telecommunications, Chongqing, China

Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

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

Representative Publications

View All Publications
17 April 2019 in IEEE Sensors Journal
Verified
Journal Article
Closed access
YDYaoxin Duan
Yaoxin Duan
KLKam-Yiu Lam
Kam-Yiu Lam
VLVictor C.S. Lee
Victor C.S. Lee
WNWendi Nie
Wendi Nie
KLKai Liu
Kai Liu
CXChun Jason Xue
Chun Jason Xue
Received signal strength (RSS)/channel state information (CSI) fingerprinting techniques have been widely adopted for WLAN-based indoor localization. However, in most of the RSS/CSI fingerprinting techniques, an application/software has to be installed on the target for uploading RSS/CSI information to the localization system. As a result, RSS/CSI fingerprinting cannot achieve passive localization (application-free), which is an essential requirement in many existing localization systems, especially for commercial and military purposes. In this paper, we propose Data Rate (DR) fingerprinting to achieve passive localization. DR fingerprinting is compatible with most off-the-shelf wireless fidelity (Wi-Fi) devices and can be directly implemented in existing wireless local area network (WLAN) based localization system without any extra hardware. In DR fingerprinting, data rates are used to replace RSS/CSI to form fingerprints, since data rate information can be directly obtained by access points. However, the inherent features of data rate, including low-resolution and serious fluctuation, etc., significantly impair the performance of DR fingerprinting. For implementing DR fingerprinting, we leverage multiple transmission power levels, propose a time window mechanism with different fingerprint formulations, design an access point (AP) switching strategy, and design a new matching algorithm named dynamic nearest neighbors (DNN). We conducted extensive experiments in a real-world testbed to study the performance of DR fingerprinting. Specifically, we compared DR fingerprinting with state-of-the-art RSS/CSI fingerprinting techniques. The experimental results illustrate that DR fingerprinting can achieve the localization accuracy which is comparable to that of RSS/CSI fingerprinting with the additional benefit of achieving passive localization.

8 January 2025 in IEEE Transactions on Information Forensics and Security
Verified
Journal Article
Closed access
DZDesheng Zheng
Desheng Zheng
WKWuping Ke
Wuping Ke
Xiaoyu LiXiaoyu Li
Xiaoyu Li
YDYaoxin Duan
Yaoxin Duan
GYGuangqiang Yin
Guangqiang Yin
FMfan min
fan min
Adversarial examples have posed a serious threat to deep neural networks due to their transferability. Existing transfer-based attacks tend to improve the transferability of adversarial examples by destroying intrinsic features. However, prior work typically employed single-dimensional or additive importance estimates, which provide inaccurate representations of features. In this work, we propose the Multi-Feature Attention Attack (MFAA), which fuses multiple layers of feature representations to disrupt category-related features and thus improve the transferability of the adversarial examples. First, MFAA introduces a layer-aggregation gradient (LAG) to obtain guidance maps, which reflect the importance of features in multiple scales. Second, it generates ensemble attention (EA), preserving object-specific features and offsetting model-specific features based on the guidance maps. Third, EA is iteratively disturbed to achieve high transferability of the adversarial examples. Empirical evaluation on the standard ImageNet dataset shows that adversarial examples crafted by MFAA can effectively attack different networks. Compared to the state-of-the-art transferable attacks, our attack improves the average attack success rate of the black-box model with defense from 88.5% to 94.1% on single-model attacks and from 86.6% to 95.1% on ensemble attacks. Our code is available at Github: https://github.com/KWPCCC/MFAA.

22 March 2018 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
WNWendi Nie
Wendi Nie
VLVictor C. S. Lee
Victor C. S. Lee
DNDusit Niyato
Dusit Niyato
YDYaoxin Duan
Yaoxin Duan
KLKai Liu
Kai Liu
SNSarana Nutanong
Sarana Nutanong
Considerable research attention has been dedicated to Vehicular Sensor Networks (VSNs) because of its great potential in traffic monitoring. By taking advantage of sensors embedded in vehicles, a VSN harvests data while vehicles are traveling along the roads and then updates the collected data to the infrastructure to support Intelligent Transportation System (ITS) applications. To meet the data collection requirements of different ITS applications, a huge number of update packets are generated, which may exhaust the available wireless communication bandwidth. To improve the efficiency of utilization of wireless communication bandwidth, in this study, we propose a Quality-oriented Data Collection (QDC) scheme which aims to effectively support both the accuracy and real-time requirements stipulated by ITS applications, while reducing communication overhead. We formulate a Minimized Communication Overhead (MCO) problem and propose two algorithms, Mixed Integer Linear Programming (MILP) and Deviation-Detection (DD), to solve the MCO problem. MILP can obtain the optimal solution to the MCO problem by having all the data collected by every vehicle while DD could achieve an efficient solution without this impractical assumption. We conducted extensive experiments by using SUMO to simulate vehicle traces in freeway and downtown environments. The experimental results have demonstrated the effectiveness of the proposed solutions.

1 June 2018
Verified
Conference Paper
Closed access
UQUmair Mujtaba Qureshi
Umair Mujtaba Qureshi
ZUZuneera UMAIR
Zuneera UMAIR
YDYaoxin Duan
Yaoxin Duan
GHGerhard Petrus Hancke
Gerhard Petrus Hancke
Bluetooth Low Energy (BLE) is widely considered for wireless Indoor Localization Systems (ILS) in which BLE Received Signal Strength (RSS) is used to derive the location of the target. The efficacy of a BLE based ILS depends on the localization accuracy. Whereas, localization accuracy depends on the stability of the BLE RSS. A number of researchers have focused on studying the impact of different parameters (such as different type of devices, environments and deployments) that cause the variations in BLE RSS. Today, BLE devices are capable of operating at multiple transmission power levels. Unlike previous studies, this paper focuses on evaluating the performance of BLE based ILS at multiple transmission power levels. Furthermore, the impact of different parameters such as multiple advertising interval, scanning interval, device orientation, distance, the effect of Line of Sight (LOS), Non-Line of Sight (NLOS) and device density over BLE RSS is also studied with multiple transmission power levels. Our main findings show that multiple transmission power level can attain localization accuracy with multiple precisions (i.e. from 2m to 5m) in an indoor classroom environment.

1 January 2020 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
YDYaoxin Duan
Yaoxin Duan
KLKam-Yiu Lam
Kam-Yiu Lam
VLVictor C. S. Lee
Victor C. S. Lee
WNWendi Nie
Wendi Nie
HLHAO LI
HAO LI
JNJoseph K. Ng
Joseph K. Ng
Passive indoor localization for mobile Wi-Fi devices, e.g., smartphones, has attracted increasing attention from research communities recently. Existing passive localization techniques leverage received signal strength (RSS) of packets transmitted by target Wi-Fi devices and do not require a dedicated software installed on the devices. However, RSS-based passive localization techniques 1) are device-dependent, which results in poor localization accuracy for a wide variety of mobile devices; and 2) cannot perform real-time passive localization. In this paper, we present a novel passive localization technique, namely packet delivery ratio (PDR) fingerprinting, to address these problems. In PDR fingerprinting, a lowest-power and highest-modulation scheme (LPHMS) is proposed to generate device-invariant packet delivery ratio, which replaces RSS to construct fingerprints, to achieve device-invariant localization accuracy. Moreover, instead of passively monitoring packets rarely sent by mobile devices, in PDR fingerprinting, access points (APs) actively transmit request-to-send (RTS) frames to trigger target devices to reply clear-to-send (CTS) frames to calculate packet delivery ratio. The RTS/CTS mechanism enables PDR fingerprinting to perform real-time localization. We have conducted extensive experiments in a real-world testbed. The experimental results demonstrate that PDR fingerprinting presents a competitive localization accuracy compared to RSS-based passive fingerprinting methods but is device-invariant.

12 July 2017 in IEEE Sensors Letters
Verified
Journal Article
Closed access
YDYaoxin Duan
Yaoxin Duan
KLKam-Yiu Lam
Kam-Yiu Lam
VLVictor C. S. Lee
Victor C. S. Lee
WNWendi Nie
Wendi Nie
HLHao Li
Hao Li
JNJoseph K. Y. Ng
Joseph K. Y. Ng
Previous fingerprinting techniques proposed for sensor-based indoor localization assume that the fingerprint database is constructed by the measurement of received signal strength (RSS) at a fixed transmit power of the transmitter sensors. In order to improve the performance of the fingerprinting technique, we consider using the measurement of path loss rather than RSS to better distinguish the signal quality at different locations. In this article, we propose an innovative fingerprint formulation, called multiple power path loss (MPPL) fingerprint, based on distinguished path losses at multiple power levels. By analyzing the signal propagation in an indoor environment, the MPPL fingerprint reveals its subtle advantage of using multiple transmit power. In addition, a location estimation algorithm is deliberately designed to exploit the performance characteristics of the MPPL fingerprint. Extensive experimental results illustrate the usability and superiority of our proposed method.