Deep Learning-based Service Distribution Model for Wireless Network Assisted Internet of Everything

Gunasekaran Manogaran, Howard University
Tu Nguyen, Kennesaw State University
Jiechao Gao, University of Virginia
Priyan Malarvizhi Kumar, Kyung Hee University


Internet of Everything (IoE) provides scalable service support for a heterogeneous class of users and applications. Concorde service delivery and quality are the focused problems retaining the user's satisfaction in the next-generation Wireless Networks (WNs). However, the flexible support for applications and users remains biased due to WN connections and unstable network architectures. This article introduces a Mutable Service Distribution Model (MSDM) for providing unified IoE application response. The proposed model handles the bias in IoE supporting WN connections and service drops between successive intervals. In this process, deep recurrent learning identifies the continuity between different intervals, preventing service overlapping. The service overlapping that degrades the Quality of Service (QoS) is distributed for different resource providers and WN architectures for delay-less responses. The bias is monitored by the learning output, ensuring re-assignment of application/ user service requests. Therefore, the service response drops are reduced with controlled time, aiding flexibility. Besides, the proposed model's performance is verified using response ratio, overhead, and assigning time.