A Binary Sparse Uniform Modem System for Resource Allocation in 5G and Beyond Networks

In this era, resource allocation in 5th Generation (5G) and beyond networks establishes an effective, dynamic and intelligent management for resources of networks. However, the traditional Machine Learning (ML) and Deep Learning (DL) approaches exhibit challenges like latency optimization, user mobility and spectrum efficiency for improved network performance. In order to solve these issues, the proposed model utilizes Binary Sparse Uniform Modem (BSUM) for resource allocation. Primarily, the Deep Slice and Secure 5G & Long Term Evolution (LTE) Wireless dataset is used for allocating the resources. Subsequently, the preprocessing technique accompanies Normalization, Scaling and Transformation which helps for aligning feature scales, avoiding from dominant features that leads to equitable and optimized allocation of resources like spectrum and power. Next step involves extraction of features using Deep Convolutional Neural Networks (DCNN) that helps in identification of patterns and relationships and enhancing the scalability. Further, Classification model is incorporated with BSUM for adapting seamless dynamic environment of networks and preliminaries required for resource allocation. The results of the proposed model are superior to the existing models by achieving 35.92 Megabits Per Second (MBPS) and 89.23% Interactivity score.