New Approaches of Cloud Services Access using Tonido Cloud Server for Real-Time Applications

Abstract

A Tonido cloud server provides a private cloud storage solution and synchronizes customers and employees with the required cloud services over the enterprise. Generally, access to any cloud services by users is via the Internet connection, which can face some problems, and then users may encounter in accessing these services due to a weak Internet connection or heavy load sometimes especially with live video streaming applications overcloud. In this work, flexible and inexpensive proposed accessing methods are submitted and implemented concerning real-time applications that enable users to access cloud services locally and regionally. Practically, to simulate our network connection, we proposed to use the Raspberry-pi3 model B+ as a router wireless LAN (WLAN) that enables users to have the cloud services using different access approaches such as wireless and wireline connections. As a case study for a real-time application over the cloud server, it is suggested to do a live video streaming using an IP webcam and IVIDEON cloud where the streaming video can be accessed via the cloud server at any time with different users taking into account the proposed practical connections. Practical experiments showed and proved that accessing real-time applications of cloud services via wireline and wireless connections is improved by using Tonido cloud server's facilities.

Efficient Energy Management for a Proposed Integrated Internet of Things-Electric Smart Meter (2IOT-ESM) System

“…Another research proposed Smart Energy Monitoring System (SEMS) (Govindarajan, et al, 2019). This work introduced SEMS using the cloud-computing framework, which is useful for tracking data and storing data in the cloud server (Abdulredah, et al, 2020) using the protocol of two-way messaging. This research has a drawback, it didn't consider the theft case for the energy consumption from some consumers.…”

Increasing validation accuracy of a face mask detection by new deep learning model-based classification

“…The multi-scale feature maps alleviate the difficulty of detecting small faces, while the aforementioned multi-scale aspect ratios reduce processing costs and trust rectification, which is in line with biological intuition and might further improve the detection rate. The suggested method, FDDB, outperformed the latest generation of CNN based algorithms in a public benchmark, but with fewer limitations [5], [6]. Face detection is critical in the development of facial recognition, expression, tracking, and classification.…”

“…The 1 st and the 2 nd moments refer to the mean and the uncentered variance, respectively. To achieve moments estimation, Adam employs exponentially moving averages, the computation of which being achieved on the gradient that is evaluated on a current mini-batch, as expressed as in ( 5) and (6).…”