arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Author

Lin Su

195 papers · 1983–2026

1983: 11984: 21985: 01986: 01987: 01988: 01989: 01990: 01991: 01992: 01993: 01994: 01995: 01996: 11997: 11998: 31999: 02000: 12001: 22002: 112003: 72004: 52005: 102006: 62007: 72008: 32009: 32010: 72011: 92012: 82013: 82014: 62015: 52016: 42017: 22018: 22019: 72020: 72021: 82022: 72023: 122024: 112025: 162026: 13
1983peak 16 in 20252026

Frequent co-authors

Papers

showing 1–25 of 195 RSS

ImageBERT: Cross-modal Pre-training with Large-scale Weak-supervised Image-Text Data

Di Qi, Lin Su, Jia Song +3 more

cs.CV · · cited 280× · arXiv ↗ · more like this ·

In this paper, we introduce a new vision-language pre-trained model -- ImageBERT -- for image-text joint embedding. Our model is a Transformer-based model, which takes different modalities as input and models the relationship between them. The model is pre-trained on four tasks simultaneously: Masked Language Modeling (MLM), Masked Object Classification (MOC), Masked Region Feature Regression (MRFR), and Image Text Matching (ITM). To further enhance the pre-training quality, we have collected a Large-scale weAk-supervised Image-Text (LAIT) dataset from Web. We first pre-train the model on this dataset, then conduct a second stage pre-training on Conceptual Captions and SBU Captions. Our experiments show that multi-stage pre-training strategy outperforms single-stage pre-training. We also fine-tune and evaluate our pre-trained ImageBERT model on image retrieval and text retrieval tasks…

Performance analysis of soft handoff in CDMA cellular networks

Szu‐Lin Su, Jen‐Yeu Chen, Jane-Hwa Huang

Computer Networks and Communications · IEEE Journal on Selected Areas in Communications · · cited 126× · DOI ↗ · more like this ·

The code-division multiple-access (CDMA) scheme has been considered as one possible choice of the future standards for cellular networks because of its various advantages. Since there can be only one carrier frequency being used in CDMA systems, a handoff scheme with diversity, a so-called "soft handoff", was proposed for higher communication quality and capacity. A mathematical model is developed to analyze the soft handoff process. Markov's concept is applied to describe the system's steady state statistical behavior. System performance such as blocking probability, handoff refused probability, and channel efficiency are also determined. It is concluded that the larger the area the soft handoff region is, the better users in the cellular network will feel.

A real-time portable sign language translation system

Lih‐Jen Kau, Wan‐Lin Su, Pei-Ju Yu +1 more

Human-Computer Interaction · · cited 58× · DOI ↗ · more like this ·

In this paper, a wireless hand gesture recognition glove is proposed for real-time translation of Taiwanese sign language. To discriminate between different hand gestures, we have flex and inertial sensors embedded into the glove so that the three most important parameters, i.e., the posture of fingers, orientation of the palm, and motion of the hand, defined in Taiwanese Sign Language can be recognized without ambiguity. The finger flexion postures acquired by flex sensors, the palm orientation acquired by G-sensor, and the motion trajectory acquired by gyroscope are used as the input signals of the proposed system. The input signals will be acquired and examined periodically to see if it is a legal sign language gesture or not. Once the sampled signal can last longer than a predefined clock cycles, it is regarded as a valid gesture and will be sent to cell phone via Bluetooth for…

M<sup>3</sup>P: Learning Universal Representations via Multitask Multilingual Multimodal Pre-training

Minheng Ni, Haoyang Huang, Lin Su +5 more

Computer Vision and Pattern Recognition · · cited 57× · DOI ↗ · more like this ·

We present M <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> P, a Multitask Multilingual Multimodal Pre-trained model that combines multilingual pre-training and multimodal pre-training into a unified framework via multitask pre-training. Our goal is to learn universal representations that can map objects occurred in different modalities or texts expressed in different languages into a common semantic space. In addition, to explicitly encourage fine-grained alignment between images and non-English languages, we also propose Multimodal Code-switched Training (MCT) to combine monolingual pre-training and multimodal pre-training via a code-switch strategy. Experiments are performed on the multilingual image retrieval task across two benchmark datasets, including MSCOCO and Multi30K. M <sup xmlns:mml="http://www.w3.org/1998/Math/MathML"…

BERT-Based Approaches to Identifying Malicious URLs

Ming‐Yang Su, Kuan-Lin Su

Information Systems · Sensors · · cited 53× · DOI ↗ · more like this ·

Malicious uniform resource locators (URLs) are prevalent in cyberattacks, particularly in phishing attempts aimed at stealing sensitive information or distributing malware. Therefore, it is of paramount importance to accurately detect malicious URLs. Prior research has explored the use of deep-learning models to identify malicious URLs, using the segmentation of URL strings into character-level or word-level tokens, and embedding and employing trained models to differentiate between URLs. In this study, a bidirectional encoder representation from a transformers-based (BERT) model was devised to tokenize URL strings, employing its self-attention mechanism to enhance the understanding of correlations among tokens. Subsequently, a classifier was employed to determine whether a given URL was malicious. In evaluating the proposed methods, three different types of public datasets were…

Grey-based power control for DS-CDMA cellular mobile systems

Szu‐Lin Su, Yu-Che Su, Jen-Fa Huang

Computer Networks and Communications · IEEE Transactions on Vehicular Technology · · cited 35× · DOI ↗ · more like this ·

The propagation channel of a mobile radio system exhibits severe signal shadowing and multipath fading, which results in wide variation of signal-to-interference ratio (SIR) at the receiver. To tackle this problem, power control is used to maintain the desired link quality and thus achieve higher capacity. In order to mitigate the channel variation effect precisely, a new application of grey theory to the power control strategy in the direct-sequence code-division multiple-access cellular mobile systems is introduced. This scheme aims to predict the SIR affected by the channel variation at the receiver and issue an appropriate control signal to the transmitter. The simulation results indicate that the grey-based scheme can offer less outage probability than the previous mechanisms.

DenseNetX and GRU for the sussex-huawei locomotion-transportation recognition challenge

Yida Zhu, Haiyong Luo, Runze Chen +2 more

Computer Vision and Pattern Recognition · · cited 30× · DOI ↗ · more like this ·

The Sussex-Huawei Locomotion-Transportation (SHL) recognition challenge organized at the HASCA Workshop of UbiComp 2020 presents a large and realistic dataset with different activities and transportation. The goal of this human activity recognition challenge is to recognize eight modes of locomotion and transportation from 5-second frames of sensor data of a smartphone carried in the unknown position. In this paper, our team (We can fly) summarize our submission to the competition. We proposed a one-dimensional (1D) DenseNetX model, a deep learning method for transportation mode classification. We first convert sensor readings from the phone coordinate system to the navigation coordinate system. Then, we normalized each sensor using different maximums and minimums and construct multi-channel sensor input. Finally, 1D DenseNetX with the Gated Recurrent Unit (GRU) model output the…

Relationship between problem-solving style and mathematical literacy

Tai Wen Chun, Lin Su Wei

Education · Educational Research and Reviews · · cited 28× · DOI ↗ · more like this ·

Currently, mathematics education is focused on ensuring that students can apply the knowledge and skills they learn to everyday life; students are expected to develop their problem-solving abilities to face challenges by adopting various perspectives. When faced with a problem, students may employ different methods or patterns to solve it. If this assertion is true, then how are the various types of problem-solving styles related to mathematical literacy? This survey was conducted to investigate this critical and noteworthy topic. Research data were obtained from the 2012 Programme test for International Student Assessment in Taiwan, taken by 15-year-old students. Latent class analysis (LCA), which is appropriate for identifying otherwise unobservable subgroups within a population, was conducted to determine how students respond to problem-solving scenarios and identify patterns of…

Hybrid Convolutional Optoelectronic Reservoir Computing for Image Recognition

Philip Jacobson, Mizuki Shirao, Kerry Yu +2 more

Artificial Intelligence · Journal of Lightwave Technology · · cited 26× · DOI ↗ · more like this ·

Photonic delay-based reservoir computers (RC) have emerged as an attractive high-speed, low-power alternative to traditional digital hardware for AI. We demonstrate experimentally a novel hybrid RC scheme in which input data is first preprocessed through several convolutional layers, either trained or untrained, digitally to generate novel feature maps. These random feature maps are then processed through an optoelectronic implementation of delay-based RC. Using the MNIST dataset of handwritten digits, experiments of our proposed hybrid scheme achieve classification error of 1.6% using untrained convolutions, and an error of 1.1% using trained convolutions, results comparable to that of state-of-the-art machine learning algorithms. Additionally, our experimental implementation can offer a potential 10x decrease in model training time, compared to that of common digital alternatives.

An Android system design and implementation for Telematics services

Cheng Yong-hua, Wen‐Kuang Kuo, Szu‐Lin Su

Sociology and Political Science · · cited 20× · DOI ↗ · more like this ·

The Telematics is the integration of telecommunication and informatics, also known as ICT (Information and Communications Technology). The application trend of Telematics is to establish a user friendly vehicle communication environment to merge the information for the traffic safety, road navigation, remote business, A/V entertainment, etc. The key technologies of Telematics are the heterogeneous communication and information fusion. This paper aims to design a Android-based mobile device platform that integrated with network management functions and MOST (Media Oriented Systems Transport) technologies to utilize multiple network access and Telematics services. The platform composed a heterogeneous network management algorithm which includes roaming and sharing functions to maintain the qualities of services and satisfy different kinds of scenario needs. Besides, we implemented those…

Topology-transparent node activation scheduling schemes for multihop TDMA ad hoc networks

Szu‐Lin Su

Computer Networks and Communications · · cited 19× · DOI ↗ · more like this ·

In this paper, we study topology-transparent node activation transmission scheduling protocols for multihop TDMA ad hoc networks. We focus on quality-of-service (QoS) provisioning for each node, particularly when node mobility is considered. A framework for topology-transparent node activation scheduling based on theory of block designs is proposed. The proposed framework guarantees conflict-free transmission slots for each node in each frame by mathematical properties of block designs. Based on two mapping methods, called mapping I and mapping II, the proposed framework can be categorized into two types. With the proposed framework, we study and evaluate some series of block constructions. We then propose several topology-transparent scheduling algorithms based on the results derived. The proposed schemes maximize the minimum system throughput. Numerical results show that the proposed…

Real-time dwell scheduling of component-oriented phased array radars

Tei‐Wei Kuo, Yung‐Sheng Chao, Chin‐Fu Kuo +2 more

Hardware and Architecture · NTUR (臺灣機構典藏) · · cited 18× · DOI ↗ · more like this ·

A multi-function phased array radar must search and track suspicious targets in its surveillance space in a real-time fashion. This paper targets one of the most important issues in the design of modern phased array radars: real-time dwell scheduling. We propose a rate-based approach to schedule radar dwells in a real-time fashion. We show how to reserve radar resources to guarantee the minimum radar operation without sacrificing the stability of the system. The strength of our approach is verified by a series of simulation experiments based a real phased array radar for air defense frigates.

A Low-Interference Relay Selection for Decode-and-Forward Cooperative Network in Underlay Cognitive Radio

Chih‐Wen Chang, Po-Hsun Lin, Szu‐Lin Su

Computer Networks and Communications · · cited 17× · DOI ↗ · more like this ·

In the underlay decode-and-forward (DaF) cooperative cognitive radio (CR) network, an optimal relay can be selected by the conventional max-min selection on the condition of not violating the interference temperature (IT) limit. However, the max-min selection may cause some extra amount of inter

Topology-Independent Link Activation Scheduling Schemes for Mobile CDMA Ad Hoc Networks

Yi-Sheng Su, Szu‐Lin Su, Jung-Shian Li

Computer Networks and Communications · IEEE Transactions on Mobile Computing · · cited 17× · DOI ↗ · more like this ·

In this paper, we study medium access control (MAC) protocols with quality-of-service (QoS) support, that is, topology-independent link activation transmission scheduling, for mobile code-division multiple-access (CDMA) ad hoc networks. QoS provisioning for each communication link is guaranteed without the need to adopt transmission schedules in mobile environments. An interference model, which captures the difference between transmission and interference ranges, is considered. Under this interference model, an approach to guaranteeing conflict-free transmission slots in each frame (QoS provisioning) for each communication link is proposed. Compared with the previously known method, superior performance is obtained. We then present a topology-independent link activation scheduling framework based on the theory of group-divisible (GD) designs. By the mathematical properties of GD…

A Novel QoS Admission Control for Ad Hoc Networks

Szu‐Lin Su, Yi-Wen Su, Jing-Yen Jung

Computer Networks and Communications · · cited 17× · DOI ↗ · more like this ·

For real-time traffic requirements in mobile ad hoc networks a novel and effective QoS admission control routing protocol (QACRP) is proposed in this paper. With a limit amount of extra overhead, QACRP can avoid network congestion by a simple and precise admission control to block most of overloading flow-requests in route discovery process. System simulations show that this proposed scheme can increase the system throughput and reduce both the dropping rate and the end-to-end delay when compared with previous QoS routing schemes like QoS-aware routing protocol and CACP protocol. Hence, QACRP is surely an effective routing protocol to provide QoS guarantee for real-time traffic.

Two-stage ensemble paging strategy for cellular communication systems

Szu‐Lin Su, Kuo-Tay Chen

Computer Networks and Communications · IEEE Communications Letters · · cited 17× · DOI ↗ · more like this ·

Instead of per user basis, we propose a simple paging strategy based on the consideration of ensemble paging channel resource of each cell. Using only the information of the last known locations of mobile subscribers, the proposed paging scheme under a maximum delay constraint can improve the system performance (discovery rate, blocking rate, and average paging delay) significantly in the heavy load.

A Graph Convolution Neural Network Based Method for Insider Threat Detection

Kexiong Fei, Jiang Zhou, Lin Su +3 more

Computer Networks and Communications · · cited 16× · DOI ↗ · more like this ·

In this research, we propose Log2Graph, a new insider threat detection method based on graph convolution neural network (GCN). This method first retrieves the corresponding logs and features from log files through feature extraction. Specifically, we use an auxiliary feature of anomaly index to describe relationship between entities, such as users and hosts, instead of establish complex connections between them. Second, these logs and features are augmented through a combination of oversampling and downsampling, to prepare for the next-stage supervised learning process. Third, we use three elaborated rules to construct the graph of each user by connecting the logs according to chronological and logical relationship. At last, the graph convolution neural network constructed is used to detect insider threats. Our validation and evaluation results confirm that Log2Graph can greatly improve…

Performance of digital code tracking loops for direct-sequence spread-spectrum signals in mobile radio channels

Szu‐Lin Su, Nan‐Yang Yen

Computer Networks and Communications · IEEE Transactions on Communications · · cited 16× · DOI ↗ · more like this ·

The performance of first- and second-order non-coherent digital delay lock loops (DDLL) for direct-sequence spread-spectrum (DS-SS) signals is investigated in the mobile radio environment. The mobile radio channel is first characterized by Rayleigh fading and Doppler shift. A closed-form expression for the timing error transition probability density function of the Chapman-Kolmogorov (C-K) equation is proposed. The probability density function of the steady-state timing error for the first- and second-order DDLL is obtained by solving the C-K equation numerically, and the results are confirmed by computer simulations. Furthermore, the mean time to lose lock (MTLL) of the first-order loop is evaluated, and some numerical results and simulation results are reported. Finally, the steady-state timing error and MTLL of the first-order loop for DS-SS signals in the log-normal fading…

Leveraging Process-Action Epistemic Network Analysis to Illuminate Student Self-Regulated Learning with a Socratic Chatbot

Joel Weijia Lai, Wei Qiu, Muang Thway +5 more

Computer Science Applications · · cited 15× · DOI ↗ · more like this ·

The growing utilization of generative artificial intelligence (GenAI) has sparked discussions regarding integrating these tools into educational settings to enrich the learning experience for teachers and students. Self-regulated learning (SRL) research is pivotal in addressing this inquiry. One prevalent manifestation of GenAI is the large-language model chatbot, enabling users to seek information and assistance. This paper aims to showcase how student interaction data with a chatbot can be used in learning analytics to gain insights into self-regulated learning. This is achieved by adapting existing SRL frameworks to comprehend \NumStu students' interaction with an educational Socratic chatbot for a statistics class at the introductory undergraduate level. Chatbot conversations from students are categorized into learning actions and processes using the framework's process-action…

Reverse-link power control strategies for CDMA cellular network

Szu‐Lin Su, Shinn-Shyue Shieh

Computer Networks and Communications · · cited 14× · DOI ↗ · more like this ·

Power control (PC) is an important issue in the DS-CDMA cellular mobile radio system. For the IS-95 standard, the reverse-link PC strategy is based on fixed-step power change and it needs only one command bit per update. We call this strategy DM-PC. From the point of view of digital communication, ADM will have better performance with the same number of control bits compared to DM. So it appears to be a good idea to improve the DM-PC and be compatible with the current system as IS-95. However, the pure ADM-PC is not better than DM-PC from our earlier simulation results. Hence, we propose another PC strategy called modified ADM-PC. The simulation results show that it has better performance than DM-PC.

Joint Topology-Transparent Scheduling and QoS Routing in Ad Hoc Networks

Yi-Sheng Su, Szu‐Lin Su, Jung-Shian Li

Computer Networks and Communications · IEEE Transactions on Vehicular Technology · · cited 13× · DOI ↗ · more like this ·

This paper considers the problem of joint topologytransparent scheduling (TTS) and quality-of-service (QoS) routing in ad hoc networks and presents a joint scheme for the problem. Due to its ability to guarantee single-hop QoS support, TTS is chosen as the underlying medium-access-control (MAC) protocol. By being built on top of TTS, this paper first designs methods for bandwidth estimation and allocation (BWE and BWA, respectively) to provide QoS support without knowledge of slot status information, and then, estimates and allocates nonassigned eligible bandwidth for best effort (BE) flows. With these bandwidth management methods, this paper proposes a QoS routing protocol for a mixture of QoS and BE flows. Idealized simulation results based on the standard radio model, which ignores external sources of radio interference and protocol inefficiencies, reveal that the proposed joint…

Causal inference for recurrent event data using pseudo-observations

Chien‐Lin Su, Robert W. Platt, Jean‐François Plante

Statistics and Probability · Biostatistics · · cited 11× · DOI ↗ · more like this ·

Recurrent event data are commonly encountered in observational studies where each subject may experience a particular event repeatedly over time. In this article, we aim to compare cumulative rate functions (CRFs) of two groups when treatment assignment may depend on the unbalanced distribution of confounders. Several estimators based on pseudo-observations are proposed to adjust for the confounding effects, namely inverse probability of treatment weighting estimator, regression model-based estimators, and doubly robust estimators. The proposed marginal regression estimator and doubly robust estimators based on pseudo-observations are shown to be consistent and asymptotically normal. A bootstrap approach is proposed for the variance estimation of the proposed estimators. Model diagnostic plots of residuals are presented to assess the goodness-of-fit for the proposed regression models. A…

Performance analysis of an ffh/bfsk receiver with ratio-statistic combining in a fading channel with multitone interference

Ye‐Shun Shen, Szu‐Lin Su

Computer Networks and Communications · IEEE Transactions on Communications · · cited 11× · DOI ↗ · more like this ·

The bit-error probability (BEP) is evaluated for a fast frequency-hopping/binary frequency-shift keying spread-spectrum communication system over a frequency-nonselective, slowly fading channel with worst-case band multitone jamming and additive white Gaussian noise. A diversity reception technique with ratio-statistic combining is applied at the receiver. Both square-law and envelope detectors are utilized and analyzed. Based on circularly symmetric signal theory, the paper obtains the closed-form expressions of probability density function and cumulative distribution function of the ratio-statistic output. It is shown from the analytical results, and verified by simulation, that the BEP performance of the ratio-statistic receiver is sensitive to the fading effect on the desired signal, but is insensitive to that on the jamming tones. It is also shown that the envelope detector…

Design and performance analysis for data transmission in GSM/GPRS system with voice activity detection

Jane-Hwa Huang, Szu‐Lin Su, Jiann-Horng Chen

Computer Networks and Communications · IEEE Transactions on Vehicular Technology · · cited 11× · DOI ↗ · more like this ·

To efficiently utilize the bandwidth of cellular mobile systems and offer service of high quality to both voice and data users, we propose a protocol to integrate packet-switched data traffic into current time-division multiple-access (TDMA)-type circuit-switched digital voice systems. We analyze the performance of the proposed system, which transmits data packets in the silent periods of a conversation with voice activity detection and adapts itself to the GSM/GPRS system, which uses the idle channels to provide data services. We show that the proposed protocol can increase the bandwidth utilization efficiency and improve the throughput/delay performance of the data transmission while minimizing the impact on the current GSM/GPRS service.

A multi‐step method for passive broadband source localisation using a single vector sensor

He Li, Tong Wang, Lin Su +4 more

Signal Processing · IET Radar Sonar & Navigation · · cited 8× · DOI ↗ · more like this ·

Abstract In the deep ocean, a vector sensor is deployed near the seabed to receive direct waves from the reliable acoustic path (RAP). In the space domain, the vector sensor can estimate the vertical arrival azimuth of the signal, which is sensitive to distance. In the frequency domain, the periodic interference pattern formed by the direct and surface‐reflected wave is decided by both the source depth and the vertical arrival azimuth. In this work, a multi‐step method for passive broadband source localisation using a single‐vector sensor (SVS) is proposed. At first, the depth is estimated by the interference pattern in the frequency domain. Then the ranging problem is solved based on the depth estimated result. The impact of the surface‐generated noise is considered. Compared with the scalar vertical line array (VLA) in simulation, the localisation performance of SVS has higher…