Peijun Ma

Peijun Ma

Auto-generated profile, based on publicly available information

Peijun Ma


Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

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

Representative Publications

View All Publications
16 March 2022 in IEEE Electron Device Letters
Verified
Journal Article
Closed access
PDPengfei Dong
Pengfei Dong
JZJincheng Zhang
Jincheng Zhang
QYQinglong Yan
Qinglong Yan
ZLZhihong Liu
Zhihong Liu
PMPeijun Ma
Peijun Ma
YHYue Hao
Yue Hao
In this work, we show that the $\beta $ -Ga 2 O 3 Schottky Barrier Diode (SBD) can perform beyond the 1-D unipolar limit of the SiC and GaN by employing a deep trench with filled thick SiO 2 layer structure to enhance the breakdown voltage (BV). By doing so, BV and specific on- resistance ( $\text{R}_{\text {on},\text {sp}}$ ) of 5–6 kV and 3.4 $\text{m}\boldsymbol {\Omega } \cdot \text {cm}^{{2}}$ are simultaneously derived on the SBDs with $\beta $ -Ga 2 O 3 epi-layer thickness of 10 $\boldsymbol {\mu } \text{m}$ and diode radius of 90 $\boldsymbol {\mu }\text{m}$ . Therefore, the Baliga’s power figure of merit (P-FOM = BV $^{{2}} /\text{R}_{\text {on,sp}}$ ) is yielded to be 7.4-10.6 GW $/$ cm 2 . To the best of all authors’ knowledge, those P-FOMs are the highest values among all types of SBDs, which is a significant step towards the Ga 2 O 3 SBD performance improvement. Combined with negligible forward hysteresis and low reverse leakage current, vertical $\beta $ -Ga 2 O 3 SBDs with state-of-the-art BV and P-FOM show its great promise for next generation high voltage and high power applications.

23 January 2025 in IEEE Transactions on Computers
Verified
Journal Article
Closed access
PMPeijun Ma
Peijun Ma
JLJie Li
Jie Li
HLHongjin Liu
Hongjin Liu
JSJiangyi Shi
Jiangyi Shi
SZShaolin Zhang
Shaolin Zhang
YHYue Hao
Yue Hao
Currently, untrusted third-party entities are increasingly involved in various stages of IC design and manufacturing, posing a significant threat to the reliability and security of SoCs due to the presence of hardware Trojans (HTs). In this paper, gate-level HT detection methods based on graph neural networks( GNNs) are established to overcome the defects of existing machine learning, which makes it difficult to characterize circuit connection relationships. We introduce harmonic centrality in the feature engineering of gate-level HT detection, which reflects the positional information of nodes and their adjacent nodes in the graph, thereby enhancing the quality of feature engineering. We use the golden section weight optimization algorithm to configure penalty weights to alleviate the problem of extreme data imbalance. In the SAED database, GraphSAGE-LSTM model obtained a TPR of 88.06% and an average F1 score of 90.95%. In the combined HT netlist of LEDA datasets, GraphSAGE-POOL model obtains a TPR of 88.50% and the best F1 score of 92.17%. In sequential HT netlist, GraphSAGE-LSTM model performs optimally, with a TPR of 98.25% and an average F1 score of 98.59%. Compared to existing detection models, the F1 score is enhanced by 8.86% and 2.48% on combined and sequential HT datasets, respectively.

14 September 2021 in IEEE Transactions on Nuclear Science
Verified
Journal Article
Closed access
TZTian Zhu
Tian Zhu
Xuefeng ZhengXuefeng Zheng
Xuefeng Zheng
JWJia Wang
Jia Wang
MWMao-Sen Wang
Mao-Sen Wang
KCKai Chen
Kai Chen
YHYue Hao
Yue Hao
The combined effects of proton irradiation and forward gate-bias stress on the interface traps of AlGaN/GaN heterostructure have been studied in this article. It is found that the effect of proton irradiation and forward gate-bias on the shift of flat band voltage $V_{\mathrm {FB}}$ is independent. By utilizing the frequency-dependent conductance technique, it is found that the trap density $D_{\mathrm {T}}$ at metal/AlGaN interface decreases after proton irradiation and the $D_{\mathrm {T}}$ at most energy levels increases after the following forward gate-bias. The $D_{\mathrm {T}}$ at AlGaN/GaN interface increases after proton irradiation and then decreases after the following forward gate-bias. The energy level range of metal/AlGaN interface traps reduces significantly under the forward gate-bias for the irradiated devices, however, that of AlGaN/GaN interface traps decreases little. In summary, the combined effect of proton irradiation and forward gate-bias stress on the interface traps is more complex than that on the bulk traps in AlGaN layer.

24 January 2025 in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Verified
Journal Article
Closed access
PMPeijun Ma
Peijun Ma
GSGe Shang
Ge Shang
HLHongjin Liu
Hongjin Liu
JSJiangyi Shi
Jiangyi Shi
WPWeitao Pan
Weitao Pan
YHYue Hao
Yue Hao
Presents corrections to the paper, (Corrections to “GNN-Based Hardware Trojan Detection at Register Transfer Level Leveraging Multiple-Category Features”).

13 December 2024 in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Verified
Journal Article
Closed access
PMPeijun Ma
Peijun Ma
GSGe Shang
Ge Shang
HLHongjin Liu
Hongjin Liu
JSJiangyi Shi
Jiangyi Shi
WPWeitao Pan
Weitao Pan
YHYue Hao
Yue Hao
The existing hardware Trojan (HT) detection technology usually relies on the golden reference model. With the continuous improvement of circuit integration, the detection accuracy of traditional methods such as side-channel analysis has declined. Methods based on testability and switch probability analysis have shown high detection accuracy. However, these techniques have a limited detection scope and are generally ineffective at identifying HT where the Sandia controllability/observability analysis program (SCOAP) values or switch probabilities are similar to those of normal signals. Against this backdrop, this article proposes a detection method based on graph neural networks (GNNs), which can achieve HT detection at the register transfer level (RTL) without the golden reference model. First, the RTL code is transformed into a data flow graph (DFG), and node feature extraction and node label marking are carried out during the transformation process. To mitigate the impact of insufficient initial features on the GNN model performance, the node feature vector used in this article comprises 37-D node types and 6-D structural features such as the minimum distance from the primary input (PI) and primary output (PO), in-degree, and out-degree. Subsequently, several GNN models are built for node classification tasks. The best model achieves an average of 99.1% recall and 96.7% F 1-score on the open-source dataset on the Trust-Hub platform. Compared to the state-of-the-art detection results at RTL, the F 1-score in this article has increased by an average of 3.8%.

25 February 2025 in IEEE Transactions on Circuits and Systems I: Regular Papers
Verified
Journal Article
Closed access
ZHZhijian Hao
Zhijian Hao
CHChenlong He
Chenlong He
JLJiaming Liu
Jiaming Liu
QZQi Zheng
Qi Zheng
JXJinchang Xu
Jinchang Xu
YHYue Hao
Yue Hao
The introduction of multiple transform types into the Versatile Video Coding (VVC) standard has yielded notable encoding gains but also resulted in substantial computational burdens, posing two critical challenges for hardware implementation: fast kernel selection and efficient transform computation design. Existing studies typically address these challenges in isolation, lacking a holistic solution for VVC transform coding. In this paper, we presents a groundbreaking transform accelerator that unifies transform kernel selection and multiple transform circuit within a single framework. In terms of algorithms, driven by mechanistic analysis, we propose a decision tree-based kernel selection algorithm that ensures both high decision accuracy and computational efficiency. Additionally, we design a transfer matrix-based approximation algorithm for Discrete Sine Transform Type-7 and a matrix decomposition-based improved computation for Discrete Cosine Transform Type-2, significantly reducing the computational complexity. On the hardware front, we implement a high-precision and area-efficient transform accelerator, which integrates highly pipelined kernel selection and transform computation architectures. With multiple reuse and parallelism strategies, the accelerator demonstrates substantial resource efficiency advantages. Experimental results reveal that the proposed accelerator achieves a circuit resource reduction of over 44% with a slight performance degradation, while maintaining processing capabilities up to 8K@57 fps. To the best of our knowledge, this is the first comprehensive hardware solution for VVC transform coding that jointly addresses the challenges of kernel selection and transform circuit design.