Prevention of border infiltration by militants in J&K region using IoT

• Main Authors: Singh, Utkarsh, Singh, Abhyudaya P., Nayak, Sankalp, Behura, Ashutosh, Rout, Jitendra Kumar
• Format: Conference Proceeding
• Language: English
• Subjects: Algorithms; Armed forces; Artillery; Automation; Bombs; Borders; Casualties; Damage; Drones; High altitude; Lethality; Projectiles; Surveillance
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/5.0132561

With the enormous growth of IoT, it’s high time for our nation to use the brilliance of this technology for the defense sector. In the current scenario, the soldiers have to manually survey the borders, leading to a massive error probability. The primary purpose of this work is to optimize the error and give 24x7 surveillance feedback, which can capacitate the Indian Armed Forces to take action as per the data collected. With their compact body frame and exceptional maneuverability, drones can easily avoid detection and infiltrate across international borders without raising any suspicion. The drones can carry surveillance over a particular target for days without being noticed at an extremely high altitude. With advanced imagery and algorithms, drones can plot suspicious patterns and movements via surveillance of the target before striking the target with lethality and accuracy. Usage of Drones has also led to a drastic reduction in collateral damage to the Civilians and their properties compared to traditional weapons of war such as mines, Bombs, artillery shells, etc. Computer software capable of accurately modeling the movement of enemy troops and their hideouts for the missile strike that leads to minimizing collateral damage. A study estimated that civilian causalities accounted for 8%-17% of all deaths from drone strikes by the US as a part of the War on Terror Campaign in different countries like Yemen, Afghanistan, Somalia, and Pakistan. In contrast, civilian casualties of World War II are 40%-67%. There is a striking difference in how drones help in saving the lives of the innocent. Compared to other forms of surveillance, our model proposes an interlinked 24*7 all-weather reconnaissance network blanket functioning in a fully automated capacity. The automation is to be done with the help of deep learning with human intervention in case of extreme emergencies. We have also proposed the majority of the system be deployed on raspberry pi, making it more cost-efficient and modular to handle.