Fine-tuning the bias terms of large language models (LLMs) has the potential to achieve unprecedented parameter efficiency while maintaining competitive performance, particularly in low-data regimes.However, the link between fine-tuning different bias terms (i.e., b q , b k , and b v in the query, key, or value projections) and downstream performance remains largely unclear to date.In this paper, we investigate the link between fine-tuning b q , b k , and b v with the performance of the downstream task.Our key finding is that directly fine-tuning b v generally leads to higher downstream performance in low-data regimes, in comparison to b q and b k .We extensively evaluate this unique property across a wide range of LLMs spanning encoder-only and decoderonly architectures up to 6.7B parameters (including bias-free LLMs).Our results provide strong evidence for the effectiveness of directly fine-tuning b v across various downstream tasks.

Time constants in spiking neural networks (SNNs) are crucial for determining performance. While prior work shows that learning time constants can improve accuracy, it typically assumes near-optimal initial values and rarely examines recovery from poor initializations. We systematically study how membrane and synaptic time constants affect SNN performance using multiple training strategies. Our results show that suboptimal values for time constants can reduce accuracy by nearly 10%, but networks can recover through optimization during the training process.

This work investigates lightweight personalisation of micro-electro-mechanical systems (MEMS)-based wearables using a public padel database as a case study.We compare a centralised CNN model, single-user models and two fine-tuning schemes (full and last-layer) on wrist-worn IMU data from 23 players and 13 stroke classes.Personalised models with data augmentation achieve weighted F1-scores above 90%, closing most of the gap to an optimistic single-user upper bound while reducing inter-subject variability.FLOP and memory analyses show that last-layer fine-tuning offers a favourable trade-off between accuracy and efficiency for on-device deployment in MEMS-based wearables.

Forward-only algorithms offer a promising memory-efficient alternative to Backpropagation (BP) for on-device learning. However, state-of-the-art forward-only algorithms, e.g., Forward-Forward (FF), still require a substantial amount of memory during the training process, often exceeding the limits of mobile edge and Internet of Things (IoT) devices. At the same time, existing memory-optimization techniques, e.g., binarizing parameters and activations, are mainly designed for BP, hence significantly degrading the classification performance when applied to state-of-the-art forward-only algorithms. In this paper, we propose a memory-efficient forward-only algorithm called TinyFoA, to reduce dynamic memory overhead in the training process. Our TinyFoA optimizes the memory efficiency not only by layer-wise training but also by partially updating each layer, as well as by binarizing the weights and the activations. We extensively evaluate our proposed TinyFoA against BP and other forward-only algorithms and demonstrate its effectiveness and superiority compared to state-of-the-art forward-only algorithms in terms of classification performance and training memory overhead, reducing the memory overheads by an order of magnitude.

Vision impairment increases risks such as social isolation, mobility challenges, and falls. Wearable Augmented Reality (AR) devices with Artificial Intelligence (AI) can enhance sensory perception by enabling real-time recognition of obstacles, assisting visually impaired individuals during street navigation, aiming to reduce the risk of falls. In this paper, we propose a Federated Learning (FL) framework for outdoor obstacle detection using resource-constrained edge AR devices. As such, our proposed framework is designed to optimize energy efficiency by partially fine-tuning a generic pre-trained model originally developed for visually impaired assistance. Our framework is evaluated on a testbed with NVIDIA Jetson Nano, demonstrating that FL improves accuracy over standalone models on each AR device, while achieving an accuracy comparable to centralized approaches, but without the need to transfer the local raw data on each AR device to a central server/cloud to alleviate privacy concerns.

Given two neural network classifiers with the same input and output domains, our goal is to compare the two networks in relation to each other over an entire input region (e.g., within a vicinity of an input sample). To this end, we establish the foundation of formal local implication between two networks, i.e., N2 ⇒D N1, in an entire input region D. That is, network N1 consistently makes a correct decision every time network N2 does, and it does so in an entire input region D. We further propose a sound formulation for establishing such formally-verified (provably correct) local implications. The proposed formulation is relevant in the context of several application domains, e.g., for comparing a trained network and its corresponding compact (e.g., pruned, quantized, distilled) networks. We evaluate our formulation based on the MNIST, CIFAR10, and two real-world medical datasets, to show its relevance.

Machine Learning (ML) offers many opportunities, but its reliance on personal data raises privacy concerns.One such example is the Membership Inference Attack (MIA), which aims to determine whether a specific data point was part of a model's training dataset.In this paper, we investigate this attack on Random Forests (RFs) and propose a method to quantify their vulnerability to MIA.We also demonstrate that in collaborative setups like federated learning, a client with access to the model and partial training dataset can establish MIA against other clients' training data.The effectiveness of our method is validated through experiments.

We investigate the intricate relation between robustness of a Deep Neural Network (DNN) model, a typical safety property, and membership inference, a prominent attack on privacy. To this end, we introduce the notion of Patient Membership Inference in the context of personalized health and precision medicine where personalized models are often adopted. Given a set of patients and a model trained using the data of one of them, Patient Membership Inference aims at identifying the patient whose data was used for training. For this, we leverage on the specificities of robustness of the model when considering data from different patients. In contrast to the classical membership inference, where the task is to determine whether a certain sample has been part of the training set, patient membership inference does not assume access to training data. As such, patient membership inference also demonstrates that access to training data is not necessary for membership inference and that membership inference is possible even for well-generalized models, not suffering from overfitting. We evaluate and demonstrate that robustness may be used to infer membership in the context of two healthcare application domains, i.e., epileptic seizure and cardiac-rhythm abnormality detections.

Interpretability has become a crucial component in the Machine Learning (ML) domain. This is particularly important in the context of medical and health applications, where the underlying reasons behind how an ML model makes a certain decision are as important as the decision itself for the experts. However, interpreting an ML model based on limited local data may potentially lead to inaccurate conclusions. On the other hand, centralized decision making and interpretability, by transferring the data to a centralized server, may raise privacy concerns due to the sensitivity of personal/medical data in such applications.In this paper, we propose a federated interpretability scheme based on SHAP (SHapley Additive exPlanations) value and DeepLIFT (Deep Learning Important FeaTures) to interpret ML models, without sharing sensitive data and in a privacy-preserving fashion. Our proposed federated interpretability scheme is a decentralized framework for interpreting ML models, where data remains on local devices, and only values that do not directly describe the raw data are aggregated in a privacy-preserving fashion to interpret the model.

The human brain performs tasks with an outstanding energy efficiency, i.e., with approximately 20 Watts. The state-of-the-art Artificial/Deep Neural Networks (ANN/DNN), on the other hand, have recently been shown to consume massive amounts of energy. The training of these ANNs/DNNs is done almost exclusively based on the back-propagation algorithm, which is known to be biologically implausible. This has led to a new generation of forward-only techniques, including the Forward-Forward algorithm. In this paper, we propose a lightweight inference scheme specifically designed for DNNs trained using the Forward-Forward algorithm. We have evaluated our proposed lightweight inference scheme in the case of the MNIST and CIFAR datasets, as well as two real-world applications, namely, epileptic seizure detection and cardiac arrhythmia classification using wearable technologies, where complexity overheads/energy consumption is a major constraint, and demonstrate its relevance. Our code is available at https://github.com/AminAminifar/LightFF.

Machine learning techniques often lack formal correctness guarantees, evidenced by the widespread adversarial examples that plague most deep-learning applications. This lack of formal guarantees resulted in several research efforts that aim at verifying Deep Neural Networks (DNNs), with a particular focus on safety-critical applications. However, formal verification techniques still face major scalability and precision challenges. The over-approximation introduced during the formal verification process to tackle the scalability challenge often results in inconclusive analysis. To address this challenge, we propose a novel framework to generate Verification-Friendly Neural Networks (VNNs). We present a post-training optimization framework to achieve a balance between preserving prediction performance and verification-friendliness. Our proposed framework results in VNNs that are comparable to the original DNNs in terms of prediction performance, while amenable to formal verification techniques. This essentially enables us to establish robustness for more VNNs than their DNN counterparts, in a time-efficient manner.

Machine Learning (ML) algorithms are generally designed for scenarios in which all data is stored in one data center, where the training is performed. However, in many applications, e.g., in the healthcare domain, the training data is distributed among several entities, e.g., different hospitals or patients' mobile devices/sensors. At the same time, transferring the data to a central location for learning is certainly not an option, due to privacy concerns and legal issues, and in certain cases, because of the communication and computation overheads. Federated Learning (FL) is the state-of-the-art collaborative ML approach for training an ML model across multiple parties holding local data samples, without sharing them. However, enabling learning from distributed data over such edge Internet of Things (IoT) systems (e.g., mobile-health and wearable technologies, involving sensitive personal/medical data) in a privacy-preserving fashion presents a major challenge mainly due to their stringent resource constraints, i.e., limited computing capacity, communication bandwidth, memory storage, and battery lifetime. In this paper, we propose a privacy-preserving edge FL framework for resource-constrained mobile-health and wearable technologies over the IoT infrastructure. We evaluate our proposed framework extensively and provide the implementation of our technique on Amazon's AWS cloud platform based on the seizure detection application in epilepsy monitoring using wearable technologies.

Digital contact tracing substantially improves the identification of high-risk contacts during pandemics. Despite several attempts to encourage people to use digital contact-tracing applications by developing and rolling out decentralized privacy-preserving protocols (broadcasting pseudo-random IDs over Bluetooth Low Energy---BLE), the adoption of digital contact tracing mobile applications has been limited, with privacy being one of the main concerns. In this paper, we propose a decentralized privacy-preserving contact tracing protocol, called DP-ACT, with both active and passive participants. Active participants broadcast BLE beacons with pseudo-random IDs, while passive participants model conservative users who do not broadcast BLE beacons but still listen to the broadcasted BLE beacons. We analyze the proposed protocol and discuss a set of interesting properties. The proposed protocol is evaluated using both a face-to-face individual interaction dataset and five real-world BLE datasets. Our simulation results demonstrate that the proposed DP-ACT protocol outperforms the state-of-the-art protocols in the presence of passive users.

Integrating low-power wearable systems into routine health monitoring is an ongoing challenge. Recent advances in the computation capabilities of wearables make it possible to target complex scenarios by exploiting multiple biosignals and using high-performance algorithms, such as Deep Neural Networks (DNNs). However, there is a tradeoff between the algorithms’ performance and the low-power requirements of platforms with limited resources. Besides, physically larger and multi-biosignal-based wearables bring significant discomfort to the patients. Consequently, reducing power consumption and discomfort is necessary for patients to use wearable devices continuously during everyday life. To overcome these challenges, in the context of epileptic seizure detection, we propose the Multi-to-Single Knowledge Distillation (M2SKD) approach targeting single-biosignal processing in wearable systems. The starting point is to train a highly-accurate multi-biosignal DNN, then apply M2SKD to develop a single-biosignal DNN solution for wearable systems that achieves an accuracy comparable to the original multi-biosignal DNN. To assess the practicality of our approach to real-life scenarios, we perform a comprehensive simulation experiment analysis on several edge computing platforms.

In healthcare, data privacy of patients regulations prohibits data from being moved outside the hospital, preventing international medical datasets from being centralized for AI training. Federated learning (FL) is a data privacy-focused method that trains a global model by aggregating local models from hospitals. Existing FL techniques adopt a central server-based network topology, where the server assembles the local models trained in each hospital to create a global model. However, the server could be a point of failure, and models trained in FL usually have worse performance than those trained in the centralized learning manner when the patient's data are not independent and identically distributed (Non-IID) in the hospitals. This paper presents a decentralized FL framework, including training with adaptive ensemble learning and a deployment phase using knowledge distillation. The adaptive ensemble learning step in the training phase leads to the acquisition of a specific model for each hospital that is the optimal combination of local models and models from other available hospitals. This step solves the non-IID challenges in each hospital. The deployment phase adjusts the model's complexity to meet the resource constraints of wearable systems. We evaluated the performance of our approach on edge computing platforms using EPILEPSIAE and TUSZ databases, which are public epilepsy datasets.

The state-of-the-art machine learning techniques come with limited, if at all any, formal correctness guarantees. This has been demonstrated by adversarial examples in the deep learning domain. To address this challenge, here, we propose a scalable robustness verification framework for Deep Neural Networks (DNNs). The framework relies on Linear Programming (LP) engines and builds on decades of advances in the field for analyzing convex approximations of the original network. The key insight is in the on-demand incremental refinement of these convex approximations. This refinement can be parallelized, making the framework even more scalable. We have implemented a prototype tool to verify the robustness of a large number of DNNs in epileptic seizure detection. We have compared the results with those obtained by two state-of-the-art tools for the verification of DNNs. We show that our framework is consistently more precise than the over-approximation-based tool ERAN and more scalable than the SMT-based tool Reluplex.

For patients with epilepsy, automatic epilepsy monitoring, i.e., the process of direct observation of the patient’s health status in real time, is crucial. Wearable systems provide the possibility of real-time epilepsy monitoring and alerting caregivers upon the occurrence of a seizure. In the context of the ICASSP 2023 Seizure Detection Challenge, we propose a lightweight machine-learning framework for real-time epilepsy monitoring on wearable devices. We evaluate our proposed framework on the SeizeIT2 dataset from the wearable SensorDot (SD) of Byteflies. The experimental results show that our proposed framework achieves a sensitivity of 73.6% and a specificity of 96.7% in seizure detection.

Recent years have seen growing interest in leveraging deep learning models for monitoring epilepsy patients based on electroencephalographic (EEG) signals. However, these approaches often exhibit poor generalization when applied outside of the setting in which training data was collected. Furthermore, manual labeling of EEG signals is a time-consuming process requiring expert analysis, making fine-tuning patient-specific models to new settings a costly proposition. In this work, we propose the Maximum-Mean-Discrepancy Decoder (M2D2) for automatic temporal localization and labeling of seizures in long EEG recordings to assist medical experts. We show that M2D2 achieves 76.0% and 70.4% of F1-score for temporal localization when evaluated on EEG data gathered in a different clinical setting than the training data. The results demonstrate that M2D2 yields substantially higher generalization performance than other state-of-the-art deep learning-based approaches.

Epilepsy is one of the most prevalent paroxystic neurological disorders. It is characterized by the occurrence of spontaneous seizures. About 1 out of 3 patients have drug-resistant epilepsy, thus their seizures cannot be controlled by medication. Automatic detection of epileptic seizures can substantially improve the patient's quality of life. To achieve a high-quality model, we have to collect data from various patients in a central server. However, sending the patient's raw data to this central server puts patient privacy at risk and consumes a significant amount of energy. To address these challenges, in this work, we have designed and evaluated a standard federated learning framework in the context of epileptic seizure detection using a deep learning-based approach, which operates across a cluster of machines. We evaluated the accuracy and performance of our proposed approach on the NVIDIA Jetson Nano Developer Kit based on the EPILEPSIAE database, which is one of the largest public epilepsy datasets for seizure detection. Our proposed framework achieved a sensitivity of 81.25%, a specificity of 82.00%, and a geometric mean of 81.62%. It can be implemented on embedded platforms that complete the entire training process in 1.86 hours using 344.34 mAh energy on a single battery charge. We also studied a personalized variant of the federated learning, where each machine is responsible for training a deep neural network (DNN) to learn the discriminative electrocardiography (ECG) features of the epileptic seizures of the specific person monitored based on its local data. In this context, the DNN benefitted from a well-trained model without sharing the patient's raw data with a server or a central cloud repository. We observe in our results that personalized federated learning provides an increase in all the performance metric, with a sensitivity of 90.24%, a specificity of 91.58%, and a geometric mean of 90.90%.

Today, the security of wearable and mobile-health technologies represents one of the main challenges in the Internet of Things (IoT) era. Adversarial manipulation of sensitive health-related information, e.g., if such information is used for prescribing medicine, may have irreversible consequences involving patients' lives. In this article, we demonstrate the power of such adversarial attacks based on a real-world epileptic seizure detection problem. We identify the minimum perturbation required by the adversaries to declare a seizure (ictal) sample as non-seizure (inter-ictal) in emergency situations, i.e., minimal adversarial perturbation to fool the classification algorithm.

Adversarial examples have received a lot of attention over the past decade, particularly with the rise of deep neural networks. Adversarial manipulation of sensitive health-related information, e.g., if such information is used for prescribing medicine, may have irreversible consequences, involving patients' lives. In this article, we consider adversarial perturbations in the context of medical and health applications and focus on the epileptic seizure detection problem. We formulate an optimization problem for computing universal adversarial perturbations and show that such universal perturbations may be used to declare the majority of seizure samples as non-seizure, i.e., to fool the classification algorithm, while being imperceptible to the medical expert eye.

Deep convolutional neural networks have recently emerged as a state-of-the art tool in detection of seizures. Such models offer the ability to extract complex nonlinear representations of an electroencephalogram (EEG) signal which can improve accuracy over methods relying on hand-crafted features. However, neural networks are susceptible to confounding artifacts commonly present in EEG signals and are notoriously difficult to interpret. In this work, we present a neural-network based algorithm for seizure detection which leverages recent advances in information theory to construct a signal representation containing the minimal amount of information necessary to discriminate between seizure and normal brain activity. We show our approach automatically learns representations that ignore common signal artifacts and which encode medically relevant information from the raw signal.

Today, it is common knowledge in the cyber-physical systems domain that the tight interaction between the cyber and physical elements provides the possibility of substantially improving the performance of these systems that is otherwise impossible. On the downside, however, this tight interaction with cyber elements makes it easier for an adversary to compromise the safety of the system. This becomes particularly important, since such systems typically are composed of several critical physical components, e.g., adaptive cruise control or engine control that allow deep intervention in the driving of a vehicle. As a result, it is important to ensure not only the reliability of such systems, e.g., in terms of schedulability and stability of control plants, but also resilience to adversarial attacks. In this article, we propose a security-aware methodology for routing and scheduling for control applications in Ethernet networks. The goal is to maximize the resilience of control applications within these networked control systems to malicious interference while guaranteeing the stability of all control plants, despite the stringent resource constraints in such cyber-physical systems. Our experimental evaluations demonstrate that careful optimization of available resources can significantly improve the resilience of these networked control systems to attacks.

Event detection and classification algorithms are resilient towards aggressive resource-aware optimisations. In this paper, we leverage this characteristic in the context of smart health monitoring systems. In more detail, we study the attainable benefits resulting from tailoring Support Vector Machine (SVM) inference engines devoted to the detection of epileptic seizures from ECG-derived features. We conceive and explore multiple optimisations, each effectively reducing resource budgets while minimally impacting classification performance. These strategies can be seamlessly combined, which results in 12.5X and 16X gains in energy and area, respectively, with a negligible loss, 3.2% in classification performance.

Epilepsy is one of the most common neurological disorders and affects over 65 million people worldwide. Despite the continuing advances in anti-epileptic treatments, one third of the epilepsy patients live with drug resistant seizures. Besides, the mortality rate among epileptic patients is 2 - 3 times higher than in the matching group of the general population. Wearable devices offer a promising solution for the detection of seizures in real time so as to alert family and caregivers to provide immediate assistance to the patient. However, in order for the detection system to be reliable, a considerable amount of labeled data is required to train it. Labeling epilepsy data is a costly and time-consuming process that requires manual inspection and annotation of electroencephalogram (EEG) recordings by medical experts. In this paper, we present a self-learning methodology for epileptic seizure detection without medical supervision. We propose a minimally-supervised algorithm for automatic labeling of seizures in order to generate personalized training data. We demonstrate that the median deviation of the labels from the ground truth is only 10.1 seconds or, equivalently, less than 1% of the signal length. Moreover, we show that training a real-time detection algorithm with data labeled by our algorithm produces a degradation of less than 2.5% in comparison to training it with data labeled by medical experts. We evaluated our methodology on a wearable platform and achieved a lifetime of 2.59 days on a single battery charge.

Increasing internet connectivity poses an existential threat for cyber-physical systems. Securing these safety-critical systems becomes an important challenge. Cyber-physical systems often comprise several control applications that are implemented on shared platforms where both high and low criticality tasks execute together (to reduce cost). Such resource sharing may lead to complex timing behaviors and, in turn, counter-intuitive timing anomalies that can be exploited by adversaries to destabilize a critical control system, resulting in irreversible consequences. We introduce the butterfly attack, a new attack scenario against cyber-physical systems that carefully exploits the sensitivity of control applications with respect to the implementation on the underlying execution platforms. We illustrate the possibility of such attacks using two case-studies from the automotive and avionic domains.

Real-time communication over Ethernet is becoming important in various application areas of cyber-physical systems such as industrial automation and control, avionics, and automotive networking. Since such applications are typically time critical, Ethernet technology has been enhanced to support time-driven communication through the IEEE 802.1 TSN standards. The performance and stability of control applications is strongly impacted by the timing of the network communication. Thus, in order to guarantee stability requirements, when synthesizing the communication schedule and routing, it is needed to consider the degree to which control applications can tolerate message delays and jitters. In this paper we jointly solve the message scheduling and routing problem for networked cyber-physical systems based on the time-triggered Ethernet TSN standards. Moreover, we consider this communication synthesis problem in the context of control applications and guarantee their worst-case stability, taking explicitly into consideration the impact of communication delay and jitter on control quality. Considering the inherent complexity of the network communication synthesis problem, we also propose new heuristics to improve synthesis efficiency without any major loss of quality. Experiments demonstrate the effectiveness of the proposed solutions.

Today, wearable systems are facing fundamental barriers in terms of battery lifetime and quality of their results. The main challenge in wearable systems is to increase the battery lifetime, while maintaining the machine-learning performance of the system. A recently proposed concept for overcoming this challenge is self-awareness, which increases system's knowledge of itself and the surrounding environment. This is precisely what health monitoring wearable systems require to adapt to different situations. To demonstrate the impact of introducing self-awareness in wearable technologies, we consider the epileptic seizure detection problem, as a case study. Epilepsy affects around 1% of the world's population, which can dramatically degrade the quality of life and represents a major public health issue. As a result, detection of epileptic seizures has become more important over the past decades. In this paper, we aim to introduce a new generation of self-aware wearable systems to decrease energy consumption and improve their seizures detection capabilities by introducing the notion of self-awareness in such systems. These techniques include switching to low-power mode to reduce the energy consumption and machine-learning model enhancement to improve detection quality. We incorporated our proposed techniques in the machine learning module, which detects epileptic seizures by monitoring the cardiac and respiratory systems. We evaluated the performance of our approach based on an epilepsy database of more than 141 hours, provided by the Lausanne University Hospital (CHUV). Our self-aware wearable system achieves 36% reduction in computational complexity and 10.51% improvement in detection performance.

Today, epilepsy is one of the most common chronic diseases affecting more than 65 million people worldwide and is ranked number four after migraine, Alzheimer's disease, and stroke. Despite the recent advances in anti-epileptic drugs, one-third of the epileptic patients continue to have seizures. More importantly, epilepsy-related causes of death account for 40% of mortality in high-risk patients. However, no reliable wearable device currently exists for real-time epileptic seizure detection. In this paper, we propose e-Glass, a wearable system based on four electroencephalogram (EEG) electrodes for the detection of epileptic seizures. Based on an early warning from e-Glass, it is possible to notify caregivers for rescue to avoid epilepsy-related death due to the underlying neurological disorders, sudden unexpected death in epilepsy, or accidents during seizures. We demonstrate the performance of our system using the Physionet.org CHB-MIT Scalp EEG database for epileptic children. Our experimental evaluation demonstrates that our system reaches a sensitivity of 93.80% and a specificity of 93.37%, allowing for 2.71 days of operation on a single battery charge.

Today, security can no longer be treated as a secondary issue in embedded and cyber-physical systems. Therefore, one of the main challenges in these domains is the design of secure embedded systems under stringent resource constraints and real-time requirements. However, there exists an inherent trade-off between the security protection provided and the amount of resources allocated for this purpose. That is, the more the amount of resources used for security, the higher the security, but the fewer the number of applications which can be run on the platform and meet their timing requirements. This trade-off is of high importance since embedded systems are often highly resource constrained. In this paper, we propose an efficient solution to maximize confidentiality, while also guaranteeing the timing requirements of real-time applications on shared platforms.

Today, many control applications in cyber-physical systems are implemented on shared platforms. Such resource sharing may lead to complex timing behaviors and, in turn, instability of control applications. This paper highlights a number of anomalies demonstrating complex timing behaviors caused as a result of resource sharing. Such anomalous scenarios, then, lead to a dramatic increase in design complexity, if not properly considered. Here, we demonstrate that these anomalies are, in fact, very improbable. Therefore, design methodologies for these systems should mainly be devised and tuned towards the majority of cases, as opposed to anomalies, but should also be able to handle such anomalous scenarios.

Today, a considerable portion of embedded systems, e.g., automotive and avionic, comprise several control applications. Guaranteeing the stability of these control applications in embedded systems, or cyber-physical systems, is perhaps the most fundamental requirement while implementing such applications. This is different from the classical hard real-time systems where often the acceptance criterion is meeting the deadline. In other words, in the case of control applications, guaranteeing stability is considered to be a main design goal, which is linked to the amount of delay and jitter a control application can tolerate before instability. This advocates the need for new design and analysis techniques for embedded real-time systems running control applications. In this paper, the analysis and design of such systems considering a server-based resource reservation mechanism are addressed. The benefits of employing servers are manifold: providing a compositional and scalable framework, protection against other tasks' misbehaviors, and systematic bandwidth assignment and co-design. We propose a methodology for designing bandwidth-optimal servers to stabilize control tasks. The pessimism involved in the proposed methodology is both discussed theoretically and evaluated experimentally.

Control applications are considered to be among the core applications in cyber-physical and embedded realtime systems, for which jitter is typically an important factor. This paper investigates whether it is possible to guarantee certain amount of jitter for a given set of applications on a shared platform. The effect of jitter on the stability of control applications and its relation with the latency will be discussed. The importance arises from the fact that it is considerably easier to manage the constant part of the delay (known as latency), while the process of coping with the varying part of the delay (known as jitter) is more involved. The proposed solution guarantees certain jitter limits, and at the same time does not lead to overly pessimistic latency values. The results are later used in a design optimization problem to minimize the resource utilized.

Many cyber-physical systems comprise several control applications implemented on a shared platform, for which stability is a fundamental requirement. This is as opposed to the classical hard real-time systems where often the criterion is meeting the deadline. However, the stability of control applications depends on not only the delay experienced, but also the jitter. Therefore, the notion of deadline is considered to be artificial for control applications that promotes the need for new techniques for designing cyber-physical systems. The approach in this paper is built on a server-based resource reservation mechanism, which provides compositionality, isolation, and the opportunity of systematic controller-server co-design. We address the controller-server co-design of such systems to obtain design solutions with the minimal bandwidth to guarantee stability.

Ethernet AVB is being actively considered by the automotive industry as a candidate for in-vehicle communication backbone. However, several questions pertaining to schedulability of hard real-time messages transmitted via such a switch remain unanswered. In this paper, we attempt to fill this void. We derive equations to perform worst-case response time analysis on Ethernet AVB switches by considering its credit-based shaping algorithm. Also, we propose several approaches to reduce the pessimism in the analysis to provide tighter bounds.

Many cyber-physical systems comprise several control applications sharing communication and computation resources. The design of such systems requires special attention due to the complex timing behavior that can lead to poor control quality or even instability. The two main requirements of control applications are: (1) robustness and, in particular, stability and (2) high control quality. Although it is essential to guarantee stability and provide a certain degree of robustness even in the worst-case scenario, a design procedure which merely takes the worst-case scenario into consideration can lead to a poor expected (average-case) control quality, since the design is solely tuned to a scenario that occurs very rarely. On the other hand, considering only the expected quality of control does not necessarily provide robustness and stability in the worst-case. Therefore, both the robustness and the expected control quality should be taken into account in the design process. This paper presents an efficient and integrated approach for designing high-quality cyber-physical systems with robustness guarantees.

Many embedded systems comprise several controllers sharing available resources. It is well known that such resource sharing leads to complex timing behavior that can jeopardize stability of control applications, if it is not properly taken into account in the design process, e.g., mapping and scheduling. As opposed to hard real-time systems where meeting the deadline is a critical requirement, control applications do not enforce hard deadlines. Therefore, the traditional real-time analysis approaches are not readily applicable to control applications. Rather, in the context of control applications, stability is often the main requirement to be guaranteed, and can be expressed as the amount of delay and jitter a control application can tolerate. The nominal delay and response-time jitter can be regarded as the two main factors which relate the real-time aspects of a system to control performance and stability. Therefore, it is important to analyze the impact of variations in scheduling parameters, i.e., period and priority, on the nominal delay and response-time jitter and, ultimately, on stability. Based on such an analysis, we address, in this paper, priority assignment and sensitivity analysis problems for control applications considering stability as the main requirement.

Guaranteeing stability of control applications in embedded systems, or cyber-physical systems, is perhaps the alpha and omega of implementing such applications. However, as opposed to the classical real-time systems where often the acceptance criterion is meeting the deadline, control applications do not primarily enforce hard deadlines. In the case of control applications, stability is considered to be the main design criterion and can be expressed in terms of the amount of delay and jitter a control application can tolerate before instability. Therefore, new design and analysis techniques are required for embedded control systems. In this paper, the analysis and design of such systems considering server-based resource reservation mechanism are addressed. The benefits of employing servers are manifold: (1) providing a compositional framework, (2) protection against other tasks misbehaviors, and (3) systematic bandwidth assignment. We propose a methodology for designing bandwidth-efficient servers to stabilize control tasks.

Many embedded systems comprise several controllers sharing available resources. It is well known that such resource sharing leads to complex timing behavior that degrades the quality of control, and more importantly, can jeopardize stability in the worst-case, if not properly taken into account during design. Although stability of the control applications is absolutely essential, a design flow driven by the worst-case scenario often leads to poor control quality due to the significant amount of pessimism involved and the fact that the worst-case scenario occurs very rarely. On the other hand, designing the system merely based on control quality, determined by the expected (average-case) behavior, does not guarantee the stability of control applications in the worst-case. Therefore, both control quality and worst-case stability have to be considered during the design process, i.e., period assignment, task scheduling, and control-synthesis. In this paper, we present an integrated approach for designing high-quality embedded control systems, while guaranteeing their stability.

Many embedded control systems are implemented on execution platforms with several computation nodes and communication components. Distributed embedded control systems typically comprise multiple control loops that share the available computation and communication resources of the platform. It is well known that such resource sharing leads to complex delay characteristics that degrade the control quality if not properly taken into account at design time. Scheduling in computation nodes and communication infrastructure, as well as execution periods of the controllers impact the delay characteristics and, consequently, the control quality. In addition, mapping of tasks on computation nodes affect both scheduling of tasks and messages, and the assignment of periods of the control applications. Therefore, control synthesis must be considered during mapping, scheduling, and period assignment in order to achieve high control quality. This paper presents a control-quality optimization approach for integrated mapping, scheduling, period selection, and control synthesis for distributed embedded control systems.