A hybrid PSO and GA algorithm with rescheduling for task offloading in device–edge–cloud collaborative computing

Cluster Computing, Journal Year: 2024, Volume and Issue: 28(2)

Published: Nov. 26, 2024

Language: Английский

---

Federated Learning on Raspberry Pi 4: A Comprehensive Power Consumption Analysis

Published: Dec. 4, 2023

Edge Computing, a rapidly evolving sector within information technology, redefines data processing and analysis by shifting it closer to the source, away from centralized cloud servers. This paradigm promises substantial benefits for diverse applications. In realm of Artificial Intelligence Machine Learning, Federated Learning emerges as pioneering technique that harnesses Computing statistical model training. presents numerous advantages over traditional including reduced latency, heightened privacy, real-time processing. Nonetheless, introduces concerns regarding energy consumption, particularly battery-powered devices designed remote or harsh environments. study provides comprehensive assessment power consumption context operations. To achieve this, Raspberry Pi 4 an INA 219 current sensor are employed. Results show that, during communication operations, target device increases minimum 8% maximum 32% with respect its idle state. During local training operations respectively up CNN 40% RNN model.

Language: Английский

---

Mitigating content poisoning attacks in named data networking: a survey of recent solutions, limitations, challenges and future research directions

Artificial Intelligence Review, Journal Year: 2024, Volume and Issue: 58(2)

Published: Dec. 20, 2024

Abstract Named Data Networking (NDN) is one of the capable applicants for future Internet architecture, where communications focus on content rather than providing content. NDN implements Information-Centric (ICN) with its unique node structure and significant characteristics such as built-in mobility support, multicast efficient distribution to end-users. It has several key features, including inherent security, that protect communication channel. Despite good features provides, it nonetheless vulnerable a variety attacks, most critical them Content Poisoning Attack (CPA). In this survey, existing solutions presented prevention CPA in paradigm have been critically analyzed. Furthermore, we also compared suggested schemes based latency, overhead, security. addition, shown possibility other possible attacks schemes. Finally, adds some open research challanges.

Language: Английский

---

Energy and Latency-aware Computation Load Distribution of Hybrid Split and Federated Learning on IoT Devices

Published: Dec. 13, 2023

Split learning (SL) and Federated Learning (FL) are popular distributed frameworks used to increase data privacy reduce computation loads of Internet Things (IoT) devices. However, one the major challenges on IoT devices is determining portion load be assigned for compared server side. The contributions existing works in literature either limited by consideration homogeneous resources available at all or not distributing among an efficient way. In this paper, we propose adaptive clustering-based distribution method devices, with heterogeneous resource capacities, participating model training. clustering makes optimal determination split point model, which scalable even a large number numerical evaluation proposed implemented using Python 3.0 comparative performance results show that policy models reduces time 160 times average usual brute force method.

Language: Английский

---

Energy Consumption Aware Delay Minimization for UAV Enabled Internet of Vehicle

Published: Dec. 9, 2023

Language: Английский

---