Key Takeaways & Executive Findings
- •• Introduces a sum-based dynamic discrete event-triggered mechanism (SDDETM) that leverages past sampled measurements and internal dynamic variables to reduce network congestion and trigger frequency. • Models deception attacks via a Bernoulli process, representing a general Markov process, to robustly handle various attack scenarios. • Co-designs a dynamic output feedback controller (DOFC) with the SDDETM parameters using the cone complement linearization (CCL) algorithm, ensuring system stability. • Validates the algorithm through two simulation examples, demonstrating effectiveness in synchronization of delayed T–S fuzzy neural networks under deception attacks.
Abstract
This paper focuses on the design of event-triggered controllers for the synchronization of delayed Takagi–Sugeno (T–S) fuzzy neural networks (NNs) under deception attacks. The traditional event-triggered mechanism (ETM) determines the next trigger based on the current sample, resulting in network congestion. Furthermore, such methods suffer from the issues of deception attacks and unmeasurable system states. To enhance the system stability, we adaptively detect the occurrence of events over a period of time. In addition, deception attacks are recharacterized to describe general scenarios. Specifically, the following enhancements are implemented: First, we use a Bernoulli process to model the occurrence of deception attacks, which can describe a variety of attack scenarios as a type of general Markov process. Second, we introduce a sum-based dynamic discrete event-triggered mechanism (SDDETM), which uses a combination of past sampled measurements and internal dynamic variables to determine subsequent triggering events. Finally, we incorporate a dynamic output feedback controller (DOFC) to ensure the system stability. The concurrent design of the DOFC and SDDETM parameters is achieved through the application of the cone complement linearization (CCL) algorithm. We further perform two simulation examples to validate the effectiveness of the algorithm.
1. Introduction
Synchronization of neural networks (NNs), as an important method, has received extensive research in recent years for NN-based control. In the network environment, the transmission of a large amount of information for controlling NNs leads to network congestion (Wang et al., 2021), making it more difficult for NNs to synchronize. Worse, network attacks (Ma et al., 2024) as well as unmeasurable system state (Liang and Huang, 2021) problems further come up, resulting in deterioration of the system performance.
For the problem of network congestion, the event-triggered mechanism (ETM) is an effective solution to alleviate the issue. However, designing event-triggering conditions in a continuous space is a significant challenge (Liu ZQ et al., 2022; Zhang D et al., 2023). Shen et al. (2023) designed a dynamic ETM with a dynamic threshold parameter (DTP) that can adaptively adjust the triggering condition based on the evolution of system states. This allows the triggering condition to be more responsive to the current system state. Bao et al. (2024) proposed a Lyapunov function-based ETM and a sampled-data-based ETM to reduce network data transmission. The former introduces a waiting time to avoid unnecessary frequent transmission, and incorporates an acknowledgment mechanism to promptly transmit data. The latter is to check the triggering condition only at sampling instants. Lei et al. (2024) designed an ETM to ensure the existence of a strictly positive minimum interevent time, thereby preventing Zeno behavior (a phenomenon where an infinite number of triggers occur in a finite time interval in event-triggered control systems) and reducing unnecessary sampling and control updates. Liu YJ et al. (2023) compared the difference between the current error and the error at the last transmission. Data are only sent when the difference exceeds the threshold. However, these mechanisms often rely on the current state of the system to determine the next trigger, which can lead to high trigger frequency and potential Zeno behavior. This can significantly degrade the performance of the system and increase the computational and communication overhead. Consequently, this results in ineffective event filtering, and fails to reach the trade-off between the network transmission efficiency and information integrity.
Moreover, in many practical applications, the system states are often unmeasurable. The communication network is vulnerable to external malicious attacks due to its openness, sharing, interconnectivity, and versatility. Control becomes challenging, especially in the case of the system being subject to uncertainties and nonlinearities, along with corrupted or incomplete state information.
Loading authentic research manuscript (Pages 1–5)...
Zhongjing YU, Duo ZHANG, Shihan KONG, Deqiang OUYANG, Hongfei LI, Junzhi YU (2025). Sum-based dynamic discrete event-triggered mechanism for synchronization of delayed neural networks under deception attacks. Frontiers of Information Technology & Electronic Engineering. https://doi.org/10.1631/FITEE_2401000
Research & Educational Purpose Only:The translations, structured abstracts, analytical annotations, and data reports provided by SinoTechIntel are intended exclusively for academic research, internal corporate R&D, and educational benchmarking. They do not constitute formal engineering, chemical safety, legal, or professional advice.
Copyright & Intellectual Property Notice: Original copyright of the underlying source articles and experimental data remains with the respective authors, institutions, and original publishing journals. SinoTechIntel claims intellectual property only over its proprietary translations, analytical syntheses, and AEO structured enhancements in accordance with international fair use and academic citation principles.
Frequently Asked Questions
What is the sum-based dynamic discrete event-triggered mechanism (SDDETM)?
The SDDETM is a novel event-triggered mechanism that determines triggering events by combining past sampled measurements and internal dynamic variables. Unlike traditional ETMs that rely solely on current samples, SDDETM uses a sum-based approach over a period of time, effectively reducing trigger frequency and network congestion.
How does the proposed method handle deception attacks?
The proposed method models deception attacks using a Bernoulli process, which can describe a variety of attack scenarios as a type of general Markov process. This allows the controller design to robustly handle unexpected data corruption or injection, enhancing system stability under malicious attacks.
What is the main advantage of SDDETM over traditional event-triggered mechanisms?
Traditional ETMs often rely on the current system state to determine the next trigger, which can lead to high trigger frequency and potential Zeno behavior. SDDETM overcomes this by incorporating past sampled measurements and internal dynamic variables, leading to more efficient event filtering and a better trade-off between network transmission efficiency and information integrity.
What algorithm is used to co-design the controller and event-triggering parameters?
The cone complement linearization (CCL) algorithm is employed to concurrently design the dynamic output feedback controller (DOFC) and the SDDETM parameters. This ensures the system's stability while optimizing the event-triggering conditions.
What are the validation results of the proposed approach?
The effectiveness of the algorithm is validated through two simulation examples, demonstrating successful synchronization of delayed Takagi–Sugeno fuzzy neural networks under deception attacks, with reduced network congestion and trigger frequency compared to conventional methods.
Related Technical Papers & Translations
Design and optimization of a high-efficiency distillation process for cellulosic fuel ethanol integrated with thermal coupling and molecular sieve adsorption
To address the challenges of high energy consumption and prominent costs in the traditional three-columns distillation process for cellulosic fuel ethanol, a distillation—molecular sieve coupling separation process is proposed. This process integrates a three-column (crude distillation column, first distillation column, second distillation column) system with a 3A molecular sieve adsorption deep dehydration unit. A thermal coupling network is constructed via differential pressure design (steam from medium/high-pressure columns as mutual heat sources, reboiler liquid waste heat for feed preheating), and molecular sieve adsorption conditions are optimized. The study first performs a thermodynamic consistency test on the ethanol—water system, determines optimal non-random two-liquid (NRTL) model binary interaction parameters via experimental data regression for Aspen Plus simulation. Aiming at minimum total annual cost (TAC), Aspen Plus is used to optimize process parameters (theoretical tray number, feed location, reflux ratio, side-draw position, etc.). Economic analysis shows this process reduces CO2 emission costs by 27.56%, TAC by 15.58% (to 5.123 × 106 USD·a-1), and increases ethanol purity to ＞99.6%, providing an effective solution for green, efficient separation.
A cohesion loss model for determining residual strength of deep bedded sandstone
Rock residual strength, as an important input parameter, plays an indispensable role in proposing the reasonable and scientific scheme about stope design, underground tunnel excavation and stability evaluation of deep chambers. Therefore, previous residual strength models of rocks established were reviewed. And corresponding related problems were stated. Subsequently, starting from the effects of bedding and whole life-cycle evolution process, series of triaxial mechanical tests of deep bedded s
Federated model with contrastive learning and adaptive control variates for human activity recognition
Recent attention to privacy issues demands a communication-safe method for training human activity recognition (HAR) models on client activity data. Federated learning (FL) has become a compelling technique to facilitate model training between the server and clients while preserving data privacy. However, classical FL methods often assume independent and identically distributed (IID) data among clients. This assumption does not hold true in practical scenarios. Human activity in real-world scena