Exploring Biological Inspired Schemes for Connectivity in FANETs

Abstract

Nature has its own way of solving problems. It provides basic in stincts to the species which learn through time and experience. It is never in a hurry but accomplishes everything. All animals find their way to food without any external help. Similarly, they can pro tect themselves from any threats by using their own intelligence. It is indeed an amazing phenomenon and inspires humans to reflect and grow. It can assist in solving several problems in man made devices effortlessly. Making the man made systems intelligent is an area which is gaining popularity day by day. The novelty of nature allows effective and efficient solutions to several problems. One such system is Flying Ad hoc Networks (FANETs). These sys tems can be extremely useful for providing relief to humans in an inaccessible area. Moreover they can provide an arial overview of landscapes. If formed in a network, communication among these systems can be employed in several military and civilian applica tions. Due to continuous random mobility, flying systems are tricky to manage. FANETs present various challenges during communi cation due to the dynamic nature of network and ever-changing topology. Owing to high mobility, it is difficult to ensure a well connected network and link stability. In order to overcome these discrepancies, this thesis explores various ways to facilitate com munication among nodes, reducing the number of isolated nodes in FANETs utilizing the depth of nature and machine learning. In this thesis, we take inspiration from bats, grey wolves and dragon flies to resolve connectivity challenges. The nodes in FANETs learn from these species and use this learning to search for a neighbor, DRSML QAU ensuring connectivity. The proposed schemes result in larger cov erage area with reduced number of isolated drones using coopera tion. This improves the connectivity in FANETs adding to the net work intelligence, allowing communication despite the complexity of mobility and dynamic network topology