Yucheng Yu

Yucheng Yu

Auto-generated profile, based on publicly available information

Yucheng Yu

The Hong Kong Polytechnic University

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

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

Representative Publications

View All Publications
11 January 2022 in IEEE Microwave and Wireless Components Letters
Verified
Journal Article
Closed access
YYYucheng Yu
Yucheng Yu
JCJialin Cai
Jialin Cai
XZXiao-Wei Zhu
Xiao-Wei Zhu
PCPeng Chen
Peng Chen
CYChao Yu
Chao Yu
The future intelligent communication systems will dynamically adjust the transmitted signal according to the radio environment and human behavior, which will lead to the rapid change of the characteristics of power amplifier (PA) and bring new challenges for digital predistortion (DPD). In this letter, a novel self-sensing DPD (SS-DPD) technique is proposed to linearize PA driven by fast time-varied signals. By automatically sensing the features of input signal and integrating them into the neural network, the proposed model is capable of linearizing the PA operated in such time-varied scenarios without updating DPD coefficients. Furthermore, the polynomial basis functions are embedded into neural network to reduce the complexity. Experimental results on a Doherty PA driven by the fast time-varied signal show that the proposed method can achieve good performance constantly with low complexity.

10 October 2022 in IEEE Transactions on Microwave Theory and Techniques
Verified
Journal Article
Closed access
YYYucheng Yu
Yucheng Yu
PCPeng Chen
Peng Chen
XZXiao-Wei Zhu
Xiao-Wei Zhu
JZJianfeng Zhai
Jianfeng Zhai
CYChao Yu
Chao Yu
Artificial intelligence (AI) provides opportunities to enable high-efficiency wireless communication to dynamically adapt to the local environments and user demands. In this article, a continual learning digital predistortion (DPD) algorithm is proposed to linearize radio frequency (RF) power amplifier (PA) in 6G AI-empowered wireless communication. It is achieved by employing the ability of knowledge retention and knowledge transfer to merge the new PA operating state into the previous model, which makes the model more intelligent when dealing with multiple states. In addition, an augmented self-sensing model is proposed to accurately capture the common characteristics and key dynamic characteristics among different PA operating states to improve the knowledge transfer ability during the continual learning process. The experiments were carried out on a Doherty PA with dynamic configurations associated with power, bandwidth, and signal type. The experimental results have shown that the proposed technique can continuously learn when faced with unpredicted new PA operating states and so as to linearize all known states. The proposed DPD method performs with very low complexity in the long term, which is very suitable for the emerging AI-empowered dynamic scenario.

12 September 2024 in IEEE Transactions on Microwave Theory and Techniques
Verified
Journal Article
Closed access
YYYucheng Yu
Yucheng Yu
LYLuqi Yu
Luqi Yu
RLRuijia Liu
Ruijia Liu
XZXiao-Wei Zhu
Xiao-Wei Zhu
PCPeng Chen
Peng Chen
CYChao Yu
Chao Yu
Aiming at supporting diverse applications, integrated communication, sensing, and power transfer (ICSPT) is anticipated to be a key feature of 6G networks. In ICSPT scenarios, power amplifiers (PAs) suffer from dynamic effects due to rapid switching between different operating modes, posing new challenges for PA linearization. In this article, a novel digital predistortion (DPD) method is proposed to effectively compensate for PA distortion in such scenarios. To accurately characterize these dynamic effects with low complexity, we introduce a method for calculating dynamic state variables that represent the dynamic behavior of PAs. For communication mode, these variables are incorporated into a neural network (NN) model, leveraging NN’s powerful fitting capability to track changes in PA’s nonlinear characteristics due to mode switching. For sensing and power transfer modes, the variables are used to fine-tune the input signal, compensating for fluctuations in PA’s gain. Experimental validations were conducted on a millimeter-wave (mmWave) gallium nitride (GaN) PA at a center frequency of 27 GHz. The PA was excited by 40/200-MHz NR signals for communication mode, 100/500-MHz frequency-modulated continuous-wave (FMCW) signals for sensing mode, and a CW signal for power transfer mode. The test demonstrates that the proposed method significantly improves transmission performance while maintaining similar complexity.

17 January 2023 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
AFAbdelwahab Fawzy
Abdelwahab Fawzy
SSSumei Sun
Sumei Sun
Teng Joon LimTeng Joon Lim
Teng Joon Lim
YYYucheng Yu
Yucheng Yu
CYChao Yu
Chao Yu
YGYongxin Guo
Yongxin Guo
Digital pre-distortion (DPD) has recently been developed to compensate for in-phase and quadrature (IQ) imbalance and crosstalk, as well as power amplifier (PA) nonlinearity distortions in multi-input multi-output (MIMO) transmitters (TXs). Despite its limitations, most DPD models still use a simple non-iterative framework called the indirect learning architecture (ILA). This paper proposes a novel integrated DPD solution supported by iterative learning control (ILC) and a neural network (NN) model to compensate for all of these impairments simultaneously. Compared to the state-of-the-art DPD models, our proposed scheme achieves excellent in-band and out-of-band (OOB) performance. In addition, it has a significantly lower running complexity than other polynomial-based models, with 50% fewer floating-point operations (FLOPs).

8 July 2024 in IEEE Microwave and Wireless Technology Letters
Verified
Journal Article
Closed access
LYLuqi Yu
Luqi Yu
YYYucheng Yu
Yucheng Yu
GZGaojing Zhang
Gaojing Zhang
PCPeng Chen
Peng Chen
CYChao Yu
Chao Yu
In this letter, a dual-asymmetrical-Lange-coupler-based load-modulated balanced amplifier (DALC-LMBA) topology is proposed for back-off efficiency enhancement. An input asymmetrical Lange coupler (ALC) is used to sequentially turn on the balanced amplifiers (BAs) for power back-off efficiency enhancement, another output ALC is adopted to generate proper load modulation (LM) for all three transistors. To verify the proposed design approach, a 23–25 GHz DALC-LMBA was implemented in a 0.12- $\mu$ m gallium nitride (GaN) process. It achieves 34–34.5 dBm saturated output power with 18%–20% power-added-efficiency (PAE) and 14%–18% 8-dB back-off PAE across the band. When excited by a 200-MHz 7.2-dB peak-to-average power ratio 5G new radio (NR) signal with digital predistortion, it achieves 14.4%–18% average PAE and better than $-$ 41 dBc adjacent channel leakage ratio (ACLR).

9 October 2023 in IEEE Transactions on Microwave Theory and Techniques
Verified
Journal Article
Closed access
YYYucheng Yu
Yucheng Yu
LYLuqi Yu
Luqi Yu
JZJianfeng Zhai
Jianfeng Zhai
PCPeng Chen
Peng Chen
CYChao Yu
Chao Yu
In this article, a novel dynamic activation digital predistortion (DPD) technique is proposed to linearize radio frequency (RF) power amplifiers (PAs) for 6G dynamic spectrum aggregation (DSA) applications. It is achieved by developing a low-complexity single model to deal with multiple operating modes and dynamic configurations. By leveraging nonlinear prior knowledge and incorporating the dynamic configurations of each frequency band into the dual-band (DB) neural network (NN) model, a novel DPD model specifically designed for linearizing PAs operating in dynamic DB mode is proposed. To make the model compatible with single-band (SB) mode, a dynamic activation method is proposed, which can dynamically activate the model architecture based on the current operating mode. With this operation, PAs in both SB and DB modes can be efficiently linearized using a single DPD model with appropriate complexity. Experiment tests are conducted on GaN Doherty PAs in sub-6-GHz and millimeter-wave frequencies, which involve varying operating modes (SB and DB) and dynamic configurations (input power, operating frequency, and signal bandwidth). The test results demonstrate that the proposed technique achieves comparable DPD performance using a single model in contrast to conventional DPD methods that would require multiple models, which highlights the effectiveness of the proposed DPD method.