Yu-Chien Lin

Yu-Chien Lin

Auto-generated profile, based on publicly available information

Yu-Chien Lin

Department of Electrical and Computer Engineering, University of California at Davis, Davis, CA, USA

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

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

Representative Publications

View All Publications
1 September 2019
Verified
Conference Paper
Closed access
CLChia-Hung Lin
Chia-Hung Lin
YLYu-Chien Lin
Yu-Chien Lin
YBYue Bai
Yue Bai
WCWei-Ho Chung
Wei-Ho Chung
TLTa-Sung Lee
Ta-Sung Lee
HHHeikki Huttunen
Heikki Huttunen
The well-known cell-averaging constant false alarm rate (CA-CFAR) scheme and its variants suffer from masking effect in multi-target scenarios. Although order-statistic CFAR (OS-CFAR) scheme performs well in such scenarios, it is compromised with high computational complexity. To handle masking effects with a lower computational cost, in this paper, we propose a deep-learning based CFAR (DL- CFAR) scheme. DL-CFAR is the first attempt to improve the noise estimation process in CFAR based on deep learning. Simulation results demonstrate that DL-CFAR outperforms conventional CFAR schemes in the presence of masking effects. Furthermore, it can outperform conventional CFAR schemes significantly under various signal-to-noise ratio conditions. We hope that this work will encourage other researchers to introduce advanced machine learning technique into the field of target detection.

2 July 2019 in IEEE Access
Verified
Journal Article
Open access
YLYu-Chien Lin
Yu-Chien Lin
TLTa-Sung Lee
Ta-Sung Lee
YPYun-Han Pan
Yun-Han Pan
KLKuan-Hen Lin
Kuan-Hen Lin
In this paper, we propose a parameter estimation method for multiple-input-multiple-output (MIMO) automotive radars that consists of two stages. The first stage is a low-complexity three-dimensional (3D) constant false alarm rate (CFAR) detection technique that exploits spatial filtering to extend radar coverage, and it performs low-complexity peak detection. The second stage is an ESPRIT-based directionof-arrival (DOA) estimation technique that adopts time-frequency resource division to generate high-quality snapshots, and it performs DOA estimation of targets without knowledge of the target number. Computer simulations reveal that the proposed method achieves the performance of the two-dimensional ordered statistic CFAR (2D OS-CFAR) while having much lower computational complexity, and it offers higher resolution DOA estimation compared to conventional MIMO radars.

30 November 2021 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
HHHao-Wei Hsu
Hao-Wei Hsu
MLMing-Chun Lee
Ming-Chun Lee
MGMeng-Xun Gu
Meng-Xun Gu
YLYu-Chien Lin
Yu-Chien Lin
TLTa-Sung Lee
Ta-Sung Lee
In this paper, we consider the orthogonal frequency division modulation (OFDM) based integrated radar-communication (Rad-Com) for automobile systems operating in millimeter wave band with imperfect time-frequency synchronization for communications. Since different pilot power allocation and placement strategies can result in different performance, we investigate the power allocation and placement design for the Rad-Com system. We analyze the effect of power allocation and pilot placement on the system performance. Based on the analysis, we provide, along with a channel estimation scheme, the pilot power allocation and placement designs that can optimize the radar and communication performance, respectively. Moreover, the tradeoff between radar and communication performance is discussed and a tradeoff design is presented. We conduct simulations to validate our analysis and evaluate the proposed designs. Results show that our analysis is valid and insightful and the proposed designs are effective.

23 August 2020 in Journal of Signal Processing Systems
Verified
Journal Article
Closed access
PDF file
CLChia-Hung Lin
Chia-Hung Lin
YLYu-Chien Lin
Yu-Chien Lin
YWYen-Jung Wu
Yen-Jung Wu
WCWei-Ho Chung
Wei-Ho Chung
TLTa-Sung Lee
Ta-Sung Lee

13 July 2021 in IEEE Wireless Communications Letters
Verified
Journal Article
Closed access
YLYu-Chien Lin
Yu-Chien Lin
ZLZhenyu Liu
Zhenyu Liu
TLTa-Sung Lee
Ta-Sung Lee
ZDZhi Ding
Zhi Ding
Large scale MIMO FDD systems are often hampered by bandwidth required to feedback downlink CSI. Previous works have made notable progresses in efficient CSI encoding and recovery by taking advantage of FDD uplink/downlink reciprocity between their CSI magnitudes. Such framework separately encodes CSI phase and magnitude. To further enhance feedback efficiency, we propose a new deep learning architecture for phase encoding based on limited CSI feedback and magnitude-aided information. Our contribution features a framework with a modified loss function to enable end-to-end joint optimization of CSI magnitude and phase recovery. Our test results show superior performance in indoor/outdoor scenarios.

15 July 2020 in IEEE Sensors Journal
Verified
Journal Article
Closed access
YLYu-Chien Lin
Yu-Chien Lin
TLTa-Sung Lee
Ta-Sung Lee
In this paper, we propose a low-complexity highresolution direction-of-arrival (DoA) method (max-MUSIC) by exploiting multiple subarrays of a sparse multiple-input multipleoutput (MIMO) radar. In this method, specific subarrays for different MIMO configurations are selected according to a proposed selection criterion for generating diverse MUSIC spatial spectrums. Subsequently, we design a new spatial spectrum function that can perform high DoA estimation accuracy and resolution without spatial aliasing, by leveraging the advantages of the diverse spatial spectrums. With a given number of antenna elements, enhancements in array aperture size can improve estimation accuracy while causing severe aliasing, which cannot be mitigated. To define a legitimate MIMO configuration for aliasing suppression, a criterion that offers an optimal compromise between DoA estimation performance and anti-aliasing ability is provided. Simulations are provided to verify the effectiveness of the proposed method.