Split computing: DNN inference partition with load balancing in IoT-edge platform for beyond 5G

Karjee, Jyotirmoy; S, Praveen Naik; Anand, Kartik; Bhargav, Vanamala Narasimha

doi:10.1016/j.measen.2022.100409

Cited by 10 publications

(1 citation statement)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Karjee et al [20] established split computing technology to provide a superior user experience and alleviate the issues of partly offloading the DNN model inference task from an IoT device to a trusted device called an edge. When compared with in-device inference time, the results reveal that the DNN model minimizes the inference time and balances the tasks across edge devices to significantly reduce battery drainage.…”

Section: Literature Reviewmentioning

confidence: 99%

Machine Learning-Driven Ubiquitous Mobile Edge Computing as a Solution to Network Challenges in Next-Generation IoT

Moteri

Bhatia

Alojail

2023

Systems

View full text Add to dashboard Cite

Ubiquitous mobile edge computing (MEC) using the internet of things (IoT) is a promising technology for providing low-latency and high-throughput services to end-users. Resource allocation and quality of service (QoS) optimization are critical challenges in MEC systems due to the large number of devices and applications involved. This results in poor latency with minimum throughput and energy consumption as well as a high delay rate. Therefore, this paper proposes a novel approach for resource allocation and QoS optimization in MEC using IoT by combining the hybrid kernel random Forest (HKRF) and ensemble support vector machine (ESVM) algorithms with crossover-based hunter–prey optimization (CHPO). The HKRF algorithm uses decision trees and kernel functions to capture the complex relationships between input features and output labels. The ESVM algorithm combines multiple SVM classifiers to improve the classification accuracy and robustness. The CHPO algorithm is a metaheuristic optimization algorithm that mimics the hunting behavior of predators and prey in nature. The proposed approach aims to optimize the parameters of the HKRF and ESVM algorithms and allocate resources to different applications running on the MEC network to improve the QoS metrics such as latency, throughput, and energy efficiency. The experimental results show that the proposed approach outperforms other algorithms in terms of QoS metrics and resource allocation efficiency. The throughput and the energy consumption attained by our proposed approach are 595 mbit/s and 9.4 mJ, respectively.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Machine Learning-Driven Ubiquitous Mobile Edge Computing as a Solution to Network Challenges in Next-Generation IoT

Moteri

Bhatia

Alojail

2023

Systems

View full text Add to dashboard Cite

show abstract

Exploring security and privacy enhancement technologies in the Internet of Things: A comprehensive review

Ataullah,

Chauhan

2024

Security and Privacy

View full text Add to dashboard Cite

In the era heavily influenced by Internet of Things (IoT), prioritizing strong security and the protection of user privacy is of utmost importance. This comprehensive review paper embarks on a meticulous examination of the multifaceted challenges and risks facing IoT security and privacy. It encompasses the hardware, software, and data‐in‐transit domains, shedding light on potential vulnerabilities and associated threats. In response to these concerns, this paper puts forth recommendations for effective strategies to mitigate these risks. Providing a road‐map for enhancing security and privacy in IoT environments. Furthermore, this review thoroughly assesses a multitude of solutions proposed by various authors, with the primary aim of enhancing security and privacy within the IoT landscape. The analysis provides insights into the strengths and limitations of these solutions. This is aiding in the development of a holistic comprehension of the existing status of IoT security and privacy. Moreover, the paper delves into the complexities surrounding integrating emerging technologies into the IoT framework. It explores the obstacles and challenges inherent in this process and proposes potential strategies to address these hurdles. By doing so, the review provides a holistic perspective on existing security and privacy enhancement technologies and offers guidance on navigating the dynamic landscape of emerging technologies within the IoT domain. Publications included in the review consist of journal articles, conference papers, and book chapters from reputable sources indexed in SCI (Science Citation Index), Scopus, and Web of Science.

show abstract

Lightweight Multimodal Fusion Computing Model for Emotional Streaming in Edge Platform

Karjee,

Kakwani,

Anand

et al. 2024

2024 IEEE 21st Consumer Communications &Amp; Networking Conference (CCNC)

View full text Add to dashboard Cite

Split computing: DNN inference partition with load balancing in IoT-edge platform for beyond 5G

Cited by 10 publications

References 10 publications

Machine Learning-Driven Ubiquitous Mobile Edge Computing as a Solution to Network Challenges in Next-Generation IoT

Machine Learning-Driven Ubiquitous Mobile Edge Computing as a Solution to Network Challenges in Next-Generation IoT

Exploring security and privacy enhancement technologies in the Internet of Things: A comprehensive review

Lightweight Multimodal Fusion Computing Model for Emotional Streaming in Edge Platform

Contact Info

Product

Resources

About