Kiat Seng Yeo

Prof. Dr. Kiat Seng Yeo

Singapore University of Technology and Design, Singapore
orcid

Research Focus & Metrics

Academic Career Length

32Years (1995-Present)

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

Representative Publications

View All Publications
13 October 2009 in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Verified
Journal Article
Closed access
NZNing Zhu
Ning Zhu
WGWang Ling Goh
Wang Ling Goh
WZWeija Zhang
Weija Zhang
Kiat Seng YeoKiat Seng Yeo
Kiat Seng Yeo
ZKZhi Hui Kong
Zhi Hui Kong
In modern VLSI technology, the occurrence of all kinds of errors has become inevitable. By adopting an emerging concept in VLSI design and test, error tolerance (ET), a novel error-tolerant adder (ETA) is proposed. The ETA is able to ease the strict restriction on accuracy, and at the same time achieve tremendous improvements in both the power consumption and speed performance. When compared to its conventional counterparts, the proposed ETA is able to attain more than 65% improvement in the Power-Delay Product (PDP). One important potential application of the proposed ETA is in digital signal processing systems that can tolerate certain amount of errors.

16 January 2023 in IEEE Access
Verified
Journal Article
Open access
JJJ. Jithish
J. Jithish
BABithin Alangot
Bithin Alangot
NMNagarajan Mahalingam
Nagarajan Mahalingam
Kiat Seng YeoKiat Seng Yeo
Kiat Seng Yeo
The smart grid integrates Information and Communication Technologies (ICT) into the traditional power grid to manage the generation, distribution, and consumption of electrical energy. Despite its many advantages, it faces significant challenges, such as detecting abnormal behaviours in the grid. Identifying anomalous behaviours helps to discover unusual user power consumption, faulty infrastructure, power outages, equipment failures, energy thefts, or cyberattacks. Machine learning (ML)-based techniques on smart meter data has shown remarkable results in anomaly detection. However, traditional ML-based anomaly detection requires smart meters to share local data with a central server, which raises concerns regarding data security and user privacy. Server-based model training faces additional challenges, such as the requirement of centralised computing power, reliable network communication, large bandwidth capacity, and latency issues, all of which affect the real-time anomaly detection performance. Motivated by these concerns, we propose a Federated Learning (FL)-based smart grid anomaly detection scheme where ML models are trained locally in smart meters without sharing data with a central server, thus ensuring user privacy. In the proposed approach, a global model is downloaded from the server to smart meters for on-device training. After local training, local model parameters are sent to the server to improve the global model. We secure the model parameter updates from adversaries using the SSL/TLS protocol. Using standard datasets, we investigate the anomaly detection performance of federated learning and observe that FL models achieve anomaly detection performance comparable to centralised ML models while ensuring user privacy. Further, our study shows that the proposed FL-based models perform efficiently in terms of memory, CPU usage, bandwidth and power consumption at edge devices and are suitable for implementation in resource-constrained environments, such as smart meters, for anomaly detection.

6 March 2006 in IEEE Transactions on Microwave Theory and Techniques
Verified
Journal Article
Closed access
KMKaixue Ma
Kaixue Ma
JMJian-Guo Ma
Jian-Guo Ma
Kiat Seng YeoKiat Seng Yeo
Kiat Seng Yeo
MDManh Anh Do
Manh Anh Do
This paper presents the characteristics of a miniaturized microstrip filter, which has two separate coupling paths: electric coupling path and magnetic coupling path between two resonators. Either magnetic coupling or electric coupling in two paths can be dominant in the total coupling coefficient of the inter-stage resonators with the similar configuration, but different positions of transmission zero points (ZPs). Based on the proposed filter topology, second- and fourth-order filters have been designed and fabricated for the first time. Advantages of using this type of filter are not only its low insertion loss and much more compact size, but also its controllable transmission ZPs.

1 November 2010
Verified
Conference Paper
Closed access
NZNing Zhu
Ning Zhu
WGWang Ling Goh
Wang Ling Goh
GWGang Wang
Gang Wang
Kiat Seng YeoKiat Seng Yeo
Kiat Seng Yeo
The tradeoff between power consumption and speed performance has become a major design consideration when devices approach the sub-100 nm regime. It is especially critical when dealing with large data set, whereby the system is degraded in terms of power and speed. If the application can accept some errors, i.e. the application is Error - tolerant (ET), a large reduction in power and an increased in speed can be simultaneously achieved. In this paper, we shall present a novel low-power and high-speed Error-Tolerant Adder Type IV design called ETAIV. The proposed ETAIV is an enhancement of our earlier design, ETAII in terms of speed and accuracy.

23 February 2015 in IEEE Microwave and Wireless Components Letters
Verified
Journal Article
Closed access
SXShanshan Xu
Shanshan Xu
KMKaixue Ma
Kaixue Ma
Fanyi MengFanyi Meng
Fanyi Meng
Kiat Seng YeoKiat Seng Yeo
Kiat Seng Yeo
This letter reports a novel back-to-back E-shaped defected ground structure (DGS) and two-side loading scheme for miniaturized dual-band substrate integrated waveguide (SIW) bandpass filter (BPF) designs. The SIW loaded by novel DGS supports evanescent-mode wave propagation below the cut-off frequency of SIW, while achieves two transmission poles at the passband and two controllable transmission zero points. Furthermore, the novel loading scheme by loading different-sized DGS resonators on two sides of SIW is proposed for dual-band BPF designs. To validate the design and analysis, a 2.4/5.2 GHz filter prototype was designed. In good agreement with theoretical results, the measured results of the dual-band filter achieve insertion losses of 3.6 and 3.1 dB, and fractional bandwidths of 5.8% and 6.45% at 2.4 and 5.2 GHz, respectively. The filter size is only 408 mil × 420 mil.

14 August 2006 in IEEE Transactions on Circuits and Systems I: Regular Papers
Verified
Journal Article
Closed access
YLYang Lu
Yang Lu
Kiat Seng YeoKiat Seng Yeo
Kiat Seng Yeo
ACA. Cabuk
A. Cabuk
JMJianguo Ma
Jianguo Ma
MDManh Anh Do
Manh Anh Do
ZLZhenghao Lu
Zhenghao Lu
An ultra-wideband (UWB) 3.1- to 10.6-GHz low-noise amplifier (LNA) employing a common-gate stage for wideband input matching is presented in this paper. Designed in a commercial 0.18-mum 1.8-V standard RFCMOS technology, the proposed UWB LNA achieves fully on-chip circuit implementation, contributing to the realization of a single-chip CMOS UWB receiver. The proposed UWB LNA achieves 16.7plusmn0.8 dB power gain with a good input match (S11<-9 dB) over the 7500-MHz bandwidth (from 3.1 GHz to 10.6 GHz), and an average noise figure of 4.0 dB, while drawing 18.4-mA dc biasing current from the 1.8-V power supply. A gain control mechanism is also introduced for the first time in the proposed design by varying the biasing current of the gain stage without influencing the other figures of merit of the circuit so as to accommodate the UWB LNA in various UWB wireless transmission systems with different link budgets