Zeyu Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Zeyu Li has a total of 42 co-authors. Here are the top co-authors by number of shared publications:
- Quan Wang
Xidian University
8 shared publications - Zhao Huang
School of Computer Science and Technology, Xian University of Posts and Telecommunications, Xian, China
7 shared publications - Junjie Wang
School of Computer Science and Technology, Xidian University, Xi’an 710071, China
5 shared publications - Xiang Li
School of decision sciences, The Hang Seng University of Hong Kong, Hong Kong 999077, China
4 shared publications - Hongxu Gao
School of Computer Science and Technology, Xidian University, Xi’an 710071, China
3 shared publications - Guangxi Zhai
School of Computer Science and Technology, Xidian University, Xi’an 710071, China
2 shared publications
Latest Publications
- Open Access
- 15 November 2026
- Published by MDPI AG in Agriculture
- Vol. 16 (22) , 2514
AgriMamba-YOLO: Grouped State Space Modeling for Wheat Head Detection and Quadrat Counting in UAV Imagery
Estimating wheat head density from unmanned aerial vehicle (UAV) imagery supports field phenotyping; however, detection still faces the challenges of dense canopies and small targets. We propose AgriMamba-YOLO, a CNN–SSM detector that retains shallow convolutional features and applies four-direction selective scanning within each deep channel group. Target-specific P2–P4 fusion incorporates stride-32 context, normalized Wasserstein distance regression, and Gaussian Soft-NMS. Replacement and component controls evaluate the hybrid configuration, scale weighting, prediction resolution, and regression settings. The model achieves 64.6% AP on GWHD 2021, 4.5 points above YOLO11n. Five UAV runs give mean AP of 86.3% and AP of 61.2% on 364 native test frames; on the same GT, YOLO11n gives 53.9% AP and 86.5% AP, a 7.25-point AP advantage for the proposed detector. Across 180 independently measured 1 m field quadrats, MAE is 12.4 and RMSE is 17.1 heads m. Complete-frame Jetson Orin Nano processing averages 1.66 s, with 0.8% of observations exceeding the 2 s capture interval. The evaluation connects held-out plot, later-date testing to field measurements, and device processing. Height group MAEs range from 10.5 to 14.0 heads m, while distinct quadrats across heights and two test plots on one date bound altitude and wider generalization inferences.- 15 November 2026
- Published by MDPI AG in Agriculture
- Vol. 16 (22) , 2514
Open AccessAgriMamba-YOLO: Grouped State Space Modeling for Wheat Head Detection and Quadrat Counting in UAV Imagery
Estimating wheat head density from unmanned aerial vehicle (UAV) imagery supports field phenotyping; however, detection still faces the challenges of dense canopies and small targets. We propose AgriMamba-YOLO, a CNN–SSM detector that retains shallow convolutional features and applies four-direction selective scanning within each deep channel group. Target-specific P2–P4 fusion incorporates stride-32 context, normalized Wasserstein distance regression, and Gaussian Soft-NMS. Replacement and component controls evaluate the hybrid configuration, scale weighting, prediction resolution, and regression settings. The model achieves 64.6% AP on GWHD 2021, 4.5 points above YOLO11n. Five UAV runs give mean AP of 86.3% and AP of 61.2% on 364 native test frames; on the same GT, YOLO11n gives 53.9% AP and 86.5% AP, a 7.25-point AP advantage for the proposed detector. Across 180 independently measured 1 m field quadrats, MAE is 12.4 and RMSE is 17.1 heads m. Complete-frame Jetson Orin Nano processing averages 1.66 s, with 0.8% of observations exceeding the 2 s capture interval. The evaluation connects held-out plot, later-date testing to field measurements, and device processing. Height group MAEs range from 10.5 to 14.0 heads m, while distinct quadrats across heights and two test plots on one date bound altitude and wider generalization inferences. - 17 October 2026
- Published by Springer Nature
- p. 609-621
SSAFI: Sorting-Based Score Anonymization Against Feature Inference in Vertical Federated Learning
- 17 October 2026
- Published by Springer Nature
- p. 609-621
SSAFI: Sorting-Based Score Anonymization Against Feature Inference in Vertical Federated Learning
- Open Access
- 9 October 2026
- Published by MDPI AG in Information
- Vol. 17 (10) , 1004
Distributed Authorization and Reliability-Aware Hardware Security for Photovoltaic Monitoring
Unattended photovoltaic monitoring nodes must maintain authenticated operation under physical access, intermittent connectivity, and environmentally induced key-reconstruction errors. We present a distributed adaptive hardware security framework (DAHSF) combining trusted-gateway threshold authorization, adaptive static random-access memory (SRAM) physical unclonable function (PUF) reconstruction, and incident-driven inspection. Independent epoch secrets separate current authorization from retained historical shares. A nested Bose–Chaudhuri–Hocquenghem (BCH) error-correction scheme reconstructs a fixed 2016-bit SRAM-PUF response, from which universal hashing extracts a 256-bit seed; federated estimation guides decoding effort. The cost and defense trade-offs are assessed under the tested workload, gateway trust, prediction-calibration, and availability conditions. On a 30-node ESP32 platform, DAHSF reduces median startup and mean node power by 4.8% and 4.4% relative to a matched lightweight baseline, with 14.3% more incremental RAM. Predicted-bound decoding reduces mean decoding time from 96 to 59 microseconds. A separate confirmatory cohort records three failures in 120,000 stressed reconstructions; its board-clustered 95% upper bound is . In the moderate attack scenario, the full inspection profile reduces compromised-node exposure by 30.0% relative to a risk-threshold policy and by 18.0% relative to frozen propagation-rate estimates. The recorded control workload has no observed 1 ms MPPT deadline misses, while fault injection still bypasses authorization in 168 of 4000 attempts.- 9 October 2026
- Published by MDPI AG in Information
- Vol. 17 (10) , 1004
Open AccessDistributed Authorization and Reliability-Aware Hardware Security for Photovoltaic Monitoring
Unattended photovoltaic monitoring nodes must maintain authenticated operation under physical access, intermittent connectivity, and environmentally induced key-reconstruction errors. We present a distributed adaptive hardware security framework (DAHSF) combining trusted-gateway threshold authorization, adaptive static random-access memory (SRAM) physical unclonable function (PUF) reconstruction, and incident-driven inspection. Independent epoch secrets separate current authorization from retained historical shares. A nested Bose–Chaudhuri–Hocquenghem (BCH) error-correction scheme reconstructs a fixed 2016-bit SRAM-PUF response, from which universal hashing extracts a 256-bit seed; federated estimation guides decoding effort. The cost and defense trade-offs are assessed under the tested workload, gateway trust, prediction-calibration, and availability conditions. On a 30-node ESP32 platform, DAHSF reduces median startup and mean node power by 4.8% and 4.4% relative to a matched lightweight baseline, with 14.3% more incremental RAM. Predicted-bound decoding reduces mean decoding time from 96 to 59 microseconds. A separate confirmatory cohort records three failures in 120,000 stressed reconstructions; its board-clustered 95% upper bound is . In the moderate attack scenario, the full inspection profile reduces compromised-node exposure by 30.0% relative to a risk-threshold policy and by 18.0% relative to frozen propagation-rate estimates. The recorded control workload has no observed 1 ms MPPT deadline misses, while fault injection still bypasses authorization in 168 of 4000 attempts. - Open Access
BIM-Constrained Elastic Fusion Navigation Framework for UAV Bridge Inspection Under Intermittent GNSS Outage
UAV bridge inspection requires centimeter-level global positioning in the engineering coordinate frame to associate detected defects with BIM component IDs and mileage stakes. Intermittent GNSS outages beneath beams, inside box girders, and in pier-dense regions cause conventional navigation methods to accumulate drift or produce discontinuous pose estimates. This paper presents BCEF-Nav, a BIM-constrained elastic fusion navigation framework for UAV bridge inspection. The framework adaptively adjusts GNSS constraints according to signal availability and replaces degraded GNSS references with semantic–geometric constraints from a high-precision as-built BIM model within a sliding-window optimizer. Semantic-guided feature matching suppresses false correspondences in repetitive bridge structures. BCEF-Nav uses only low-cost commercial sensors and is compatible with mainstream inspection UAVs. Digital twin simulations and field experiments achieved an ATE RMSE of 5.7 cm after 120 s of complete GNSS occlusion and a 100% positioning success rate with positioning errors below 10 cm, outperforming seven recent state-of-the-art baselines.
Open AccessBIM-Constrained Elastic Fusion Navigation Framework for UAV Bridge Inspection Under Intermittent GNSS Outage
UAV bridge inspection requires centimeter-level global positioning in the engineering coordinate frame to associate detected defects with BIM component IDs and mileage stakes. Intermittent GNSS outages beneath beams, inside box girders, and in pier-dense regions cause conventional navigation methods to accumulate drift or produce discontinuous pose estimates. This paper presents BCEF-Nav, a BIM-constrained elastic fusion navigation framework for UAV bridge inspection. The framework adaptively adjusts GNSS constraints according to signal availability and replaces degraded GNSS references with semantic–geometric constraints from a high-precision as-built BIM model within a sliding-window optimizer. Semantic-guided feature matching suppresses false correspondences in repetitive bridge structures. BCEF-Nav uses only low-cost commercial sensors and is compatible with mainstream inspection UAVs. Digital twin simulations and field experiments achieved an ATE RMSE of 5.7 cm after 120 s of complete GNSS occlusion and a 100% positioning success rate with positioning errors below 10 cm, outperforming seven recent state-of-the-art baselines.
- Open Access
- 11 June 2026
- Published by MDPI AG in Information
- Vol. 17 (6) , 584
Lightweight Hardware Security Framework for IoT-Based Photovoltaic Monitoring Systems Using OTP and SRAM-PUF
Distributed photovoltaic (PV) power stations are core enablers for dual-carbon goals in modern power systems, with IoT-based monitoring systems serving as their nerve center for real-time data collection and grid dispatch. However, PV monitoring nodes operate in harsh, unattended outdoor environments with severe computational resource constraints, exposing them to critical hardware security risks that can trigger cross-domain cascading hazards. Existing research focuses primarily on communication and software security, lacking systematic hardware security modeling and lightweight defense designs. Generic IoT hardware security solutions are also inapplicable due to excessive overhead. To address these gaps, this paper proposes LHSF, a lightweight hardware security framework tailored for resource-constrained PV edge nodes. It integrates an on-chip OTP-based lightweight hardware root of trust (L-HROT) with an SRAM-PUF-driven non-resident key management protocol, which implements full-lifecycle key management via a “power-on generation, on-demand usage, post-use destruction, zero-residue storage” paradigm. Experiments on ESP32 and Raspberry Pi 4B show that LHSF provides robust resistance to side-channel recovery, physical extraction, malicious firmware boot and rollback attacks, reducing fault injection bypass rate to 6.8%. Compared to standard TPM 2.0, it cuts boot delay by 60.7%, power consumption by 18.6% and memory footprint by 72.7% with negligible performance overhead. This work fills the hardware security gap for PV monitoring systems and provides a reusable technical pathway for distributed energy IoT terminals.
- 11 June 2026
- Published by MDPI AG in Information
- Vol. 17 (6) , 584
Open AccessLightweight Hardware Security Framework for IoT-Based Photovoltaic Monitoring Systems Using OTP and SRAM-PUF
Distributed photovoltaic (PV) power stations are core enablers for dual-carbon goals in modern power systems, with IoT-based monitoring systems serving as their nerve center for real-time data collection and grid dispatch. However, PV monitoring nodes operate in harsh, unattended outdoor environments with severe computational resource constraints, exposing them to critical hardware security risks that can trigger cross-domain cascading hazards. Existing research focuses primarily on communication and software security, lacking systematic hardware security modeling and lightweight defense designs. Generic IoT hardware security solutions are also inapplicable due to excessive overhead. To address these gaps, this paper proposes LHSF, a lightweight hardware security framework tailored for resource-constrained PV edge nodes. It integrates an on-chip OTP-based lightweight hardware root of trust (L-HROT) with an SRAM-PUF-driven non-resident key management protocol, which implements full-lifecycle key management via a “power-on generation, on-demand usage, post-use destruction, zero-residue storage” paradigm. Experiments on ESP32 and Raspberry Pi 4B show that LHSF provides robust resistance to side-channel recovery, physical extraction, malicious firmware boot and rollback attacks, reducing fault injection bypass rate to 6.8%. Compared to standard TPM 2.0, it cuts boot delay by 60.7%, power consumption by 18.6% and memory footprint by 72.7% with negligible performance overhead. This work fills the hardware security gap for PV monitoring systems and provides a reusable technical pathway for distributed energy IoT terminals.
- Open Access
- 27 January 2026
- Published by MDPI AG in Electronics
- Vol. 15 (3) , 548
FPGA-Based Reconfigurable System: Research Progress and New Trend on High-Reliability Key Problems
FPGA-based reconfigurable systems play a vital role in many critical domains by virtue of their unique advantages. They can effectively adapt to dynamically changing application scenarios, while featuring high parallelism and low power consumption. As a result, they have been widely adopted in key sectors such as aerospace, nuclear industry, and weapon equipment, where high performance and stability are of utmost importance. However, these systems face significant challenges. The continuous and drastic reduction in chip process size has led to increasingly complex and delicate internal circuit structures and physical characteristics. Meanwhile, the operating environments are often harsh and unpredictable. Additionally, the adoption of untrusted third-party foundries to reduce development costs further compounds these issues. Collectively, these factors make such systems highly susceptible to reliability threats, including environmental radiation, aging degradation, and malicious hardware attacks. These problems severely impact the stable operation and functionality of the systems. Therefore, ensuring the highly reliable operation of reconfigurable systems has become a critical issue that urgently needs to be addressed. There is a pressing need to summarize their technical characteristics, research status, and development trends comprehensively and in depth. In response, this paper conducts relevant research. By systematically reviewing 183 domestic and international research papers published between 2012 and 2024, it first provides a detailed analysis of the root causes of reliability issues in reconfigurable systems, thoroughly exploring their underlying mechanisms. Second, it focuses on the key technologies for achieving high reliability, encompassing four types of fault-tolerant design technologies, three types of aging mitigation technologies, and two types of hardware attack defense technologies. The paper comprehensively summarizes relevant research findings and the latest advancements in this field, offering a wealth of references for related research. Finally, it conducts a detailed comparative analysis and summary of the research hotspots in the field of high-reliability reconfigurable systems. It objectively evaluates the achievements and shortcomings of current research efforts and delves into the development trends of key technologies for high-reliability reconfigurable systems, providing clear directions for future research and practical applications.
- 27 January 2026
- Published by MDPI AG in Electronics
- Vol. 15 (3) , 548
Open AccessFPGA-Based Reconfigurable System: Research Progress and New Trend on High-Reliability Key Problems
FPGA-based reconfigurable systems play a vital role in many critical domains by virtue of their unique advantages. They can effectively adapt to dynamically changing application scenarios, while featuring high parallelism and low power consumption. As a result, they have been widely adopted in key sectors such as aerospace, nuclear industry, and weapon equipment, where high performance and stability are of utmost importance. However, these systems face significant challenges. The continuous and drastic reduction in chip process size has led to increasingly complex and delicate internal circuit structures and physical characteristics. Meanwhile, the operating environments are often harsh and unpredictable. Additionally, the adoption of untrusted third-party foundries to reduce development costs further compounds these issues. Collectively, these factors make such systems highly susceptible to reliability threats, including environmental radiation, aging degradation, and malicious hardware attacks. These problems severely impact the stable operation and functionality of the systems. Therefore, ensuring the highly reliable operation of reconfigurable systems has become a critical issue that urgently needs to be addressed. There is a pressing need to summarize their technical characteristics, research status, and development trends comprehensively and in depth. In response, this paper conducts relevant research. By systematically reviewing 183 domestic and international research papers published between 2012 and 2024, it first provides a detailed analysis of the root causes of reliability issues in reconfigurable systems, thoroughly exploring their underlying mechanisms. Second, it focuses on the key technologies for achieving high reliability, encompassing four types of fault-tolerant design technologies, three types of aging mitigation technologies, and two types of hardware attack defense technologies. The paper comprehensively summarizes relevant research findings and the latest advancements in this field, offering a wealth of references for related research. Finally, it conducts a detailed comparative analysis and summary of the research hotspots in the field of high-reliability reconfigurable systems. It objectively evaluates the achievements and shortcomings of current research efforts and delves into the development trends of key technologies for high-reliability reconfigurable systems, providing clear directions for future research and practical applications.
- Open Access
- 27 December 2024
- Published by MDPI AG in Electronics
- Vol. 14 (1) , 82
Towards Intelligent Edge Computing: A Resource- and Reliability-Aware Hybrid Scheduling Method on Multi-FPGA Systems
Multi-FPGA systems can form larger and more powerful computing units through high-speed interconnects between chips, and are beginning to be widely used by various computing service providers, especially in edge computing. However, the new computing architecture brings new challenges to efficient and reliable task scheduling. In this context, we propose a resource- and reliability-aware hybrid scheduling method on Multi-FPGA systems. First, a set of models is established based on the resource/time requirements, communication overhead, and state conversion process of tasks to further analyze the constraints of system scheduling. On this basis, the large task is divided into subtasks based on the data dependency matrix, and the Maintenance Multiple Sequence (MMS) algorithm is used to generate execution sequences for each subtask to the Multi-FPGA systems to fully exploit resources and ensure reliable operation. Compared with state-of-the-art scheduling methods, the proposed method can achieve an average increase in resource utilization of 7%; in terms of reliability, it achieves good execution gains, with an average task completion rate of 98.3% and a mean time to failure of 15.7 years.
- 27 December 2024
- Published by MDPI AG in Electronics
- Vol. 14 (1) , 82
Open AccessTowards Intelligent Edge Computing: A Resource- and Reliability-Aware Hybrid Scheduling Method on Multi-FPGA Systems
Multi-FPGA systems can form larger and more powerful computing units through high-speed interconnects between chips, and are beginning to be widely used by various computing service providers, especially in edge computing. However, the new computing architecture brings new challenges to efficient and reliable task scheduling. In this context, we propose a resource- and reliability-aware hybrid scheduling method on Multi-FPGA systems. First, a set of models is established based on the resource/time requirements, communication overhead, and state conversion process of tasks to further analyze the constraints of system scheduling. On this basis, the large task is divided into subtasks based on the data dependency matrix, and the Maintenance Multiple Sequence (MMS) algorithm is used to generate execution sequences for each subtask to the Multi-FPGA systems to fully exploit resources and ensure reliable operation. Compared with state-of-the-art scheduling methods, the proposed method can achieve an average increase in resource utilization of 7%; in terms of reliability, it achieves good execution gains, with an average task completion rate of 98.3% and a mean time to failure of 15.7 years.
- Open Access
- 19 June 2024
- Published by MDPI AG in Electronics
- Vol. 13 (12) , 2400
A Hardware Trojan Diagnosis Method for Gate-Level Netlists Based on Graph Theory
With the increasing complexity of integrated circuit design, the threat of a hardware Trojan (HT) is becoming more and more prominent. At present, the research mainly focuses on the detection of HTs, but the amount of research on the diagnosis of HTs is very small. The number of existing HT diagnosis methods is generally completed by detecting the HT nodes in the netlist. The main reason is the lack of consideration of the integrity of HTs, so the diagnosis accuracy is low. Based on the above reason, this paper proposes two implanted node search algorithms named layer-by-layer difference search (LDS) and layer-by-layer grouping difference search (LGDS). The LDS algorithm can greatly reduce the search time, and the LGDS algorithm can solve the problem of input node disorder. The two methods greatly reduce the number of nodes sorting and comparing, and therefore the time complexity is lower. Moreover, the relevance between implanted nodes is taken into account to improve the diagnosis rate. We completed experiments on an HT diagnosis; the HT implantation example is from Trust-Hub. The experimental results are shown as follows: (1) The average true positive rate (TPR) of the diagnosis using KNN, RF, or SVM with the LDS or LGDS algorithm is more than 93%, and the average true negative rate (TNR) is 100%. (2) The average proportion of implanted nodes obtained by the LDS or LGDS algorithm is more than 97%. The proposed method has a lower time complexity compared with other existing diagnosis methods, and the diagnosis time is shortened by nearly 75%.
- 19 June 2024
- Published by MDPI AG in Electronics
- Vol. 13 (12) , 2400
Open AccessA Hardware Trojan Diagnosis Method for Gate-Level Netlists Based on Graph Theory
With the increasing complexity of integrated circuit design, the threat of a hardware Trojan (HT) is becoming more and more prominent. At present, the research mainly focuses on the detection of HTs, but the amount of research on the diagnosis of HTs is very small. The number of existing HT diagnosis methods is generally completed by detecting the HT nodes in the netlist. The main reason is the lack of consideration of the integrity of HTs, so the diagnosis accuracy is low. Based on the above reason, this paper proposes two implanted node search algorithms named layer-by-layer difference search (LDS) and layer-by-layer grouping difference search (LGDS). The LDS algorithm can greatly reduce the search time, and the LGDS algorithm can solve the problem of input node disorder. The two methods greatly reduce the number of nodes sorting and comparing, and therefore the time complexity is lower. Moreover, the relevance between implanted nodes is taken into account to improve the diagnosis rate. We completed experiments on an HT diagnosis; the HT implantation example is from Trust-Hub. The experimental results are shown as follows: (1) The average true positive rate (TPR) of the diagnosis using KNN, RF, or SVM with the LDS or LGDS algorithm is more than 93%, and the average true negative rate (TNR) is 100%. (2) The average proportion of implanted nodes obtained by the LDS or LGDS algorithm is more than 97%. The proposed method has a lower time complexity compared with other existing diagnosis methods, and the diagnosis time is shortened by nearly 75%.
- Open Access
- 8 May 2024
- Published by MDPI AG in Electronics
- Vol. 13 (10) , 1821
GLRM: Geometric Layout-Based Resource Management Method on Multiple Field Programmable Gate Array Systems
Multiple field programmable gate array (Multi-FPGA) systems are capable of forming larger and more powerful computing units through high-speed interconnections between chips and are beginning to be widely used by various computing service providers. However, the new computing architecture brings new challenges to the system’s task resource management. Existing resource management methods do not fully exploit resources in Multi-FPGA systems, and it is difficult to support fast resource request and release. In this regard, we propose a geometric layout-based resource management (GLRM) method for Multi-FPGA systems. First, a geometric layout-based task combination algorithm (TCA) was proposed to ensure that the final system can use the available FPGA resources more efficiently. Then, we optimised two resource management algorithms using TCA. Compared with state-of-the-art resource management methods, TCA increases resource flexibility by an average of 6% and resource utilisation by an average of 7%, and the two optimised resource management methods are effective in improving resource management performance.
- 8 May 2024
- Published by MDPI AG in Electronics
- Vol. 13 (10) , 1821
Open AccessGLRM: Geometric Layout-Based Resource Management Method on Multiple Field Programmable Gate Array Systems
Multiple field programmable gate array (Multi-FPGA) systems are capable of forming larger and more powerful computing units through high-speed interconnections between chips and are beginning to be widely used by various computing service providers. However, the new computing architecture brings new challenges to the system’s task resource management. Existing resource management methods do not fully exploit resources in Multi-FPGA systems, and it is difficult to support fast resource request and release. In this regard, we propose a geometric layout-based resource management (GLRM) method for Multi-FPGA systems. First, a geometric layout-based task combination algorithm (TCA) was proposed to ensure that the final system can use the available FPGA resources more efficiently. Then, we optimised two resource management algorithms using TCA. Compared with state-of-the-art resource management methods, TCA increases resource flexibility by an average of 6% and resource utilisation by an average of 7%, and the two optimised resource management methods are effective in improving resource management performance.
- 1 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11 (10) , 17688-17699
Ada-FA: A Comprehensive Framework for Adaptive Fault Tolerance and Aging Mitigation in FPGAs
Commercial SRAM-based field-programmable gate arrays (FPGAs) are extremely susceptible to failures caused by external ionizing radiation or prolonged internal overloading in harsh applications, such as single event effects (SEEs) and ageing failure. Existing methods utilize the triple modular redundancy (TMR) architecture to shield against the effects of radiation on FPGA systems. However, these solutions are resource-costly and practically unnecessary. Additionally, hard faults caused by the ageing effects of long-term usage of FPGA systems is not effectively alleviated. To address these issues, we present a comprehensive framework for ensuring adaptive fault tolerance and aging mitigation in FPGAs, i.e., Ada-FA. Ada-FA is a cross-layer-aware reliability framework that includes two phases: offline and online. (1) In the offline phase, a task criticality evaluation strategy supporting fine-grained fault tolerance is proposed to reduce the hardware resource overhead. Specifically, we improve the integer linear programming (ILP) formula, which considers both fault tolerance and ageing mitigation, to obtain the optimal reliability-aware layout, thus maximizing the mean time to failure (MTTF). (2) In the online phase, we propose a runtime management architecture to further ensure the reliable operation of FPGA systems. The experimental results show that the resource usage (RU) of the proposed Ada-FA framework is reduced by 15.8% on average compared to that of existing fault-tolerant layout/scheduling methods. Moreover, our method provides a higher reliability and task accomplishment rate (TAR) than the state-of-the-art offline ageing mitigation methods.
- 1 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11 (10) , 17688-17699
Ada-FA: A Comprehensive Framework for Adaptive Fault Tolerance and Aging Mitigation in FPGAs
Commercial SRAM-based field-programmable gate arrays (FPGAs) are extremely susceptible to failures caused by external ionizing radiation or prolonged internal overloading in harsh applications, such as single event effects (SEEs) and ageing failure. Existing methods utilize the triple modular redundancy (TMR) architecture to shield against the effects of radiation on FPGA systems. However, these solutions are resource-costly and practically unnecessary. Additionally, hard faults caused by the ageing effects of long-term usage of FPGA systems is not effectively alleviated. To address these issues, we present a comprehensive framework for ensuring adaptive fault tolerance and aging mitigation in FPGAs, i.e., Ada-FA. Ada-FA is a cross-layer-aware reliability framework that includes two phases: offline and online. (1) In the offline phase, a task criticality evaluation strategy supporting fine-grained fault tolerance is proposed to reduce the hardware resource overhead. Specifically, we improve the integer linear programming (ILP) formula, which considers both fault tolerance and ageing mitigation, to obtain the optimal reliability-aware layout, thus maximizing the mean time to failure (MTTF). (2) In the online phase, we propose a runtime management architecture to further ensure the reliable operation of FPGA systems. The experimental results show that the resource usage (RU) of the proposed Ada-FA framework is reduced by 15.8% on average compared to that of existing fault-tolerant layout/scheduling methods. Moreover, our method provides a higher reliability and task accomplishment rate (TAR) than the state-of-the-art offline ageing mitigation methods.