Network engineers are essential to the management of computersystems, as they configure services and devices on the network to guarantee effective data routing and improve computernetworks which make easier to perfo...
详细信息
ISBN:
(纸本)9798350368130
Network engineers are essential to the management of computersystems, as they configure services and devices on the network to guarantee effective data routing and improve computernetworks which make easier to perform everyday tasks like sending and receiving emails as well as more sophisticated ones like cloud computing and online gaming. In addition, modern networks have to serve a rising number of IoT devices without compromising performance and handle more sophisticated cybersecurity threats. Network engineers must always be learning and adapting to new protocols and technologies if they are to successfully tackle these difficulties. Moreover, the introduction of Large Language Models (LLMs) that are available as open-source software has revolutionized technical innovation by enabling the automation of network setups and augmenting the capabilities of network administration. These developments represent an important step forward in the field of network engineering, with the goal of maximizing efficiency and guaranteeing strong network security and functioning.
Recent breakthroughs in wireless networks have promised scalable coordinated transmissions for future ubiquitous communications, such as non-terrestrial low-earth orbit (LEO) satellite networks, for emerging commercia...
详细信息
ISBN:
(纸本)9798350317657;9798350317640
Recent breakthroughs in wireless networks have promised scalable coordinated transmissions for future ubiquitous communications, such as non-terrestrial low-earth orbit (LEO) satellite networks, for emerging commercial and military applications. However, overcoming the data throughput issue with such distributed radio systems in practical field trials and electromagnetic contested environments is challenging. This paper develops novel hardware-in-the-loop (HIL) prototyping and evaluation of a distributed software-defined radio (SDR) system and the corresponding optimal signal reception combining technique for maximizing multi-radio throughput. A programmable in-lab testbed with multiple SDR-mounted aerial devices and a transmission control protocol (TCP) backhaul is built to emulate LEO satellite channels and satellite coordination. A distributed maximum ratio combining (d-MRC), is provided and implemented to process multi-radio receptions for optimal spectral efficiency. Comprehensive over-the-air tests with both bit-streams and image transmissions validate our designed approach's real-world feasibility and its superior performance compared to an individual SDR throughput. Thus, this work introduces a HIL system integrations with coordinated commercial-off-the-shelf SDRs, providing a reference implementation for extensive wireless aerial use cases.
Power electronic-based systems exhibit non-linear dynamics requiring simultaneous control of multiple control objectives. It is therefore expected that controllers that can cope with those nonlinearities will have a b...
详细信息
ISBN:
(纸本)9798350361018;9798350361001
Power electronic-based systems exhibit non-linear dynamics requiring simultaneous control of multiple control objectives. It is therefore expected that controllers that can cope with those nonlinearities will have a better performance than controllers requiring system linearization or nesting of the control objectives in a cascaded structure. However, the problem remains how to quantify their robustness and make a fair comparison between different non-linear controllers. The conventional tools used for the robustness validation of linear controllers cannot directly be applied to different non-linear controllers. Therefore, this paper demonstrates an approach based on statistical model checking for performing controller comparisons. The performance and robustness of different controllers (linear, model predictive, and neural networks-based) were compared in the same stochastic environment. Using this approach, a statistical estimate can be obtained for how the controller performance will be affected under different scenarios.
The current paper presents the implementation of an orchestrator for distributed edge-to-cloud systems based on heterogeneous nodes in urban environments. The orchestrator considers available resources, specifically v...
详细信息
ISBN:
(纸本)9798350364279;9798350364262
The current paper presents the implementation of an orchestrator for distributed edge-to-cloud systems based on heterogeneous nodes in urban environments. The orchestrator considers available resources, specifically video feeds from surveillance cameras, distributed across levels (i.e., from cloud to edge and viceversa) as a continuum and organizes them, taking into account their properties such as distance from the data source, processing latency, involved resources, and functionalities. The continuum from edge to cloud enables distributed processing of the video signal and network infrastructure management to optimize performance and energy efficiency, supporting mobility and workload balancing.
The complexity of Cyber-Physical systems (CPS) is increasing due to the widespread availability of inexpensive hardware, sensors, actuators, and communication links. The complexity is further heightened in a network o...
详细信息
ISBN:
(纸本)9798350371291;9798350371284
The complexity of Cyber-Physical systems (CPS) is increasing due to the widespread availability of inexpensive hardware, sensors, actuators, and communication links. The complexity is further heightened in a network of cooperating CPSs (Cyber-Physical Network (CPN)), presenting both challenges and opportunities. Designing, operating, optimizing, and maintaining such CPNs becomes more difficult with rising complexity. However, judicious utilization of the expanding computational nodes, sensors, and actuators can significantly enhance system performance, reliability, and flexibility. Therefore, integrating self-X features such as self-organization, self-adaptation, and self-healing becomes crucial for these systems. In Addition, CPNs are typically mixed-critical systems, commonly employed in areas like avionics, automotive, and healthcare. The mixed-criticality nature introduces competition among applications with hard real-time constraints, those with soft real-time constraints, and best-effort applications for the available resources. This paper shortly introduces a comprehensive adaptive middleware (Chameleon) designed for CPNs and evaluates its effectiveness. The results demonstrate the capability of Chameleon in autonomously managing system resources to meet the required constraints of applications based on their criticality.
Renewable energy hybrid mini-grids have experienced significant growth in power systems, driven primarily by the need for more resilient and reliable distributed resources in modern power systems. However, despite the...
详细信息
ISBN:
(纸本)9798331540715;9798331540708
Renewable energy hybrid mini-grids have experienced significant growth in power systems, driven primarily by the need for more resilient and reliable distributed resources in modern power systems. However, despite their many benefits, the combined operation of PV and storage systems can pose challenges in system management and coordination, especially in radial distribution networks. This paper examines the operation of microgrids in a real medium-voltage radial feeder in Brazil, incorporating distributed generation and capacitor banks. Additionally, it assesses network losses with and without implementing distributed energy resources. The study considers the role of battery energy storage systems in energy and demand response, in addition to various load and an island mode operation scenario. The findings emphasize the advantages of effectively managing hybrid microgrids, highlighting enhanced reliability and service continuity in the face of diverse disturbances.
In this paper, we analyze the performance of an algorithm for adaptive diffusion networks that controls the number of nodes sampled per iteration based on the estimation error. The goal of this solution is to keep the...
详细信息
ISBN:
(纸本)9798350362527;9798350362510
In this paper, we analyze the performance of an algorithm for adaptive diffusion networks that controls the number of nodes sampled per iteration based on the estimation error. The goal of this solution is to keep the nodes sampled while the estimation error is high in magnitude, and to cease their sampling when it is sufficiently low. Our model shows that this approach can preserve the convergence rate in comparison with the case in which every node is sampled permanently, while slightly improving the steady-state performance.
作者:
Cabello, Julia GarciaCarbo-Garcia, S.Univ Granada
Andalusian Res Inst Data Sci & Computat Intellige Dept Appl Math Granada Spain Univ Granada
Andalusian Res Inst Data Sci & Computat Intellige Dept Comp Sci & Artificial Intelligence Granada Spain
In recent years, the architecture and structure of Deep Neural networks (DNNs) have become progressively more complex in order to respond to the increasing complexity of real problems. A strategy to deal with this com...
详细信息
ISBN:
(纸本)9783031820724;9783031820731
In recent years, the architecture and structure of Deep Neural networks (DNNs) have become progressively more complex in order to respond to the increasing complexity of real problems. A strategy to deal with this complexity when it affects training would be to partition DNN training in some way: for example, by distributing it among different components of a computer network. For this, training (which is in essence the minimization of the loss function) should be performed through separated "smaller pieces". This paper offers an alternative to the gradient-based DNN training from a Dynamic Programming (DP) point of view (DP is an optimisation methodology supported by the division of a complex problem into many problems of lower complexity). To do so, conditions which enable the DNN minimization algorithm to be solved under a DP perspective are studied here. In this line, in this work is proved that any artificial neural network ANN (and thus also DNNs) with monotonic activation is separable. Furthermore, whenever ANNs are considered as a dynamical system in the form of a network (known as coupled cell networks CCNs), we show that the transmission function is a separable function assuming that the activation is non-decreasing.
This literature survey provides an overview on the current state of the practice for real-time embedded healthcare monitoring systems. These technologies enable healthcare systems to keep a constant eye on patients, c...
详细信息
ISBN:
(纸本)9798350371291;9798350371284
This literature survey provides an overview on the current state of the practice for real-time embedded healthcare monitoring systems. These technologies enable healthcare systems to keep a constant eye on patients, collecting vital health data. They can send alerts and insights to healthcare providers, allowing for faster and better care. Our survey includes an investigation on the rising use of artificial intelligence and deep learning along with the use of intelligent sensors for capturing and analyzing data in real-time health monitoring systems. We also examine the current state of connected embedded health monitoring systems via, e.g. the Internet of Things (IoT). Wireless sensor networks and Tele-Health technologies are also addressed as they are instrumental in enhancing the precision, accessibility, and real-time nature of health data, ensuring that both patients and providers are always connected. Our survey discusses the applications and advantages associated with these technologies, providing valuable insights into their role in enhancing patient care, decision support, and tracking. Additionally, we navigate through the challenges, seeking pathways to mitigate them, and ensuring that the technology is inclusive, secure, and efficient. By examining the latest research and innovations, this survey aims to illuminate the current state of real-time healthcare monitoring and guide future advancements in this critical field.
distributedsystems can be found in various applications, e.g., in robotics or autonomous driving, to achieve higher flexibility and robustness. Thereby, data flow centric applications such as Deep Neural Network (DNN...
详细信息
ISBN:
(纸本)9798350354119
distributedsystems can be found in various applications, e.g., in robotics or autonomous driving, to achieve higher flexibility and robustness. Thereby, data flow centric applications such as Deep Neural Network (DNN) inference benefit from partitioning the workload over multiple compute nodes in terms of performance and energy-efficiency. However, mapping large models on distributed embedded systems is a complex task, due to low latency and high throughput requirements combined with strict energy and memory constraints. In this paper, we present a novel approach for hardware-aware layer scheduling of DNN inference in distributed embedded systems. Therefore, our proposed framework uses a graph-based algorithm to automatically find beneficial partitioning points in a given DNN. Each of these is evaluated based on several essential system metrics such as accuracy and memory utilization, while considering the respective system constraints. We demonstrate our approach in terms of the impact of inference partitioning on various performance metrics of six different DNNs. As an example, we can achieve a 47.5% throughput increase for EfficientNet-B0 inference partitioned onto two platforms while observing high energy-efficiency.
暂无评论