Abstract
In 6G edge communication networks, the machine learning models play a major role in enabling intelligent decision-making in case of optimal resource allocation in case of the healthcare system. However, it causes a bottleneck, in the form of sophisticated memory calculations, between the hidden layers and the cost of communication between the edge devices/edge nodes and the cloud centres, while transmitting the data from the healthcare management system to the cloud centre via edge nodes. In order to reduce these hurdles, it is important to share workloads to further eliminate the problems related to complicated memory calculations and transmission costs. The effort aims mainly to reduce storage costs and cloud computing associated with neural networks as the complexity of the computations increases with increasing numbers of hidden layers. This study modifies federated teaching to function with distributed assignment resource settings as a distributed deep learning model. It improves the capacity to learn from the data and assigns an ideal workload depending on the limited available resources, slow network connection, and more edge devices. Current network status can be sent to the cloud centre by the edge devices and edge nodes autonomously using cybertwin, meaning that local data are often updated to calculate global data. The simulation shows how effective resource management and allocation is better than standard approaches. It is seen from the results that the proposed method achieves higher resource utilization and success rate than existing methods. Index Terms are fuzzy, healthcare, bioinformatics, 6G wireless communication, cybertwin, machine learning, neural network, and edge.
Highlights
Since the development of edge computing [1], it has emerged as a key strategic approach in a variety of application areas, especially in the fields of data aggregation, network connectivity, and other industrial tasks. e edge is regarded as an open platform for storage and computing applications because it is situated near a data source or an object on the network side
A multinode FNN wireless healthcare model is often referred as the FNN wireless healthcare model that aims to improve the precision and performance and scales according to larger data size. e increasing size of input data learning for learning reduces significantly the training errors and enables error-free complex operations [8]. is allows the distributed FNN wireless healthcare model computing to draw significant decisions and conclusions over larger data sizes or in case of complex computing. e purpose-built distributed FNN wireless healthcare model operates in distributed edge computing environment that gains advantage in terms of its performance requirement, user cases, data size, and implementation effort
E edge node is allowed to select η(t), the learning rate, and for faster computation, we have considered η(t) 1. e federated learning (FL) is described for the distributed neural networks (DNNs) in section, where it uses an individual matrix (W) in order of representing the parameters over each hidden layer. e parameters representing the full connected layers of the DNN in FL is described in the form of 2D matrix
Summary
Since the development of edge computing [1], it has emerged as a key strategic approach in a variety of application areas, especially in the fields of data aggregation, network connectivity, and other industrial tasks. e edge is regarded as an open platform for storage and computing applications because it is situated near a data source or an object on the network side. Edge computing is placed between the cloud and end devices and uses a high-speed data communication. During the onset of large-scale distributed neural network use, the limited computing resources found at the edge devices present several challenges. With the DNN serving as a secondary model, it is possible to use the following rules: edge resources allow zero or multiple edge devices, resources available, memory requirements, and user quality of service requirements. (ii) Decentralized training data distribution is a solution that optimises the reuse of valuable network resources, even in the event of an unreliable network To distribute environments such as this, FL (each iteration) enables the edge node to compute updates to the cloud centre independent of system requirements, user cases, data size, and implementation effort.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
