Boot Loader is the crucial program that loads the operating system in memory and initializes the system. In today's world people are constantly on move and portable system are in demand specially the USB devices due to its portability and accessibility compared to CD/DVD drives. The purpose of this paper is to design a dynamic boot loader which removes the BIOS dependency and allow user to boot from USB without changing CMOS settings. The USB is devised as plug and play portable system with puppy Linux and newly developed dynamic boot loader. The device is experimented on a computer machine with 8 GB RAM, i5 processor, 64-bit Operating system and windows 7 and observed that nearly 50% reduction in booting time i.e., the time spent in changing the boot order is eliminated compared to the static boot loader. The time spent in the BIOS is dependent on the user knowledge in changing the boot priority. The portable system allows the user to work in ease in any environment with minimum requirement of Windows XP and USB 2.0 compatible system.

In today’s world, people are constantly on move and portable systems are in demand. With technology advancement, people exploit different types of memory devices for a portable system. For any external boot medium, the BIOS boot order setting change is required. The dynamic boot loader successfully eliminated this dependency and allowed the user to directly boot from any portable USB. The usage of USB has grown exponentially in recent years and securing it has become a major concern. In this paper, the USB is devised as a highly secured portable boot medium with fingerprint authentication to ensure data security. It performs feature extraction by combining both Local Directional Pattern (LDP) and Histograms of Oriented Gradients (HOG) which improves the accuracy rate. The classification is performed by random forest classifier, such that the intended users alone are granted access to the private storage area of the USB drive.

Commonly, a USB flash drive is utilized for storing, transferring and backing up data like personal files, software, media files, etc. But users might not have much knowledge about its other hidden characteristics. It could also act as a replacement of CD/hard disk/DVD media as OSs handler resources and as a plug and play portable system. However, security is a major concern for external boot system and to fix this issue, numerous solutions were proposed and implemented. Out of all the existing security provisioning schemes, biometric based security solutions are always reliable and hassle free to process. The USB drives now available with fingerprint protection and to come out of the box, this article secures the USB drives with the combination of fingerprint and finger vein. On successful authentication, the user can boot OS from USB. The performance of the work is analysed in terms of FAR, FRR, accuracy and time consumption rates and observed that it achieves greater accuracy rate when compared with other classifiers.

With technology advancement people have started using different type of memory devices for storing data and keeping it secure has become concern in today’s world. Universal Serial Bus (USB) flash drives are leading portable storage device for storage and easy transfer of data from one computer to another. The usage of USB has grown exponentially and without security the data on the disk is at risk. Nowadays USB manufacturers offer password protection and fingerprint authentication to secure the USB data. In this paper, USB is devised as a highly secured portable boot medium with fingerprint authentication to secure the data.

Rainfall hugely impacts every aspect of human life, such as transportation, agriculture, water management, and so on. It also is a grave cause of several natural calamities, like landslides, floods, and drought, which pose a serious threat to the well-being of individuals. These concerns have necessitated the need for devising an effective technique to predict rainfall, which enables the undertaking of effective preventive measures. Several works have focused on developing efficient rainfall forecasting techniques; however, the uncertain nature of rainfall and the lack of rainfall data limit their effectiveness. This paper proposes an efficient rainfall prediction strategy using an optimized Deep Learning approach. Here, prediction is carried out using a Deep Long Short Term Memory network based on the time series data of the rainfall. Further, the prediction efficiency is enhanced by the utilization of the Circle Inspired Optimization Algorithm for the weight optimization of the Deep Long Short Term Memory. Experimental results show that the devised Circle Inspired Optimization Algorithm-Deep Long Short Term Memory reveals enhanced performance by attaining a minimal value of Relative Absolute Error at 0.023 Mean Square Error of 0.151, and Root Mean Square Error of 0.389.