.png){kind=icon}
.png){kind=icon}
Return
An Efficient Mobile Edge Computing based Resource Allocation using Optimal Double Weighted Support Vector Transfer Regression
.png){kind=icon}
PRE
AI
DOI：10.1007/s10723-023-09680-z.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Mobile edge computing (MEC) technology is gaining more attention in smart cities due to its powerful computation capability. However, there arise complications related to security and privacy while transmitting and processing raw data to other cloud or MEC servers. This makes the users unwilling to update their private information on the cloud servers. To tackle this issue, we proposed a novel approach for optimal task scheduling and resource allocation processes in this paper. The proposed 'double-weighted support vector transfer regression based flow direction (DSTR-FD) approach' resolves the issues of resource management of edge servers and makes optimal task offloading decisions with minimized energy consumption. Here, the model parameters such as weight functions, regularization parameters, and kernel parameters of the DSTR network are tuned using the flow direction (FD) algorithm. The proposed method thus provides better data privacy without sharing the original data with other servers along with minimizing the utilization of energy in the Internet of Things (IoT). The efficiency of the proposed DSTR-FD approach is evaluated by comparing its results with other states of art methods. The simulation experiments illustrate that the proposed DSTR-FD approach effectively minimizes the energy utilization of all IoT devices.
Keywords:
Mobile edge computing
Resource allocation
Task offloading
Double weighted support vector transfer regression
Flow direction algorithm
Smart city
Journal
IF:
2.9
Papers:
759
Citations:
1.2K