A FUSED MACHINE LEARNING INTRUSION DETECTION MODEL IN MANETS.

Abstract

Mobile Ad-Hoc Networks – MANETs are prevalent in healthcare monitoring of high blood pressure, high cholesterol levels and various heart conditions and cardiac misnomers like syncope, third murmurs and atrial fibrillation. These irregularities that cause mysterious fainting, unexplained stroke, heart palpitations and atrial fibrillation need to be monitored remotely, accurately and effortlessly. However, the growth and provision of MANETs in smart healthcare monitoring has faced various security obstacles, primarily security. The characteristic mobility of these health monitoring devices as well as their inherently dynamic network topology, causes the connectivity structure to change frequently and unpredictably. Further, these smart devices have limited resources in storage, processor capability and memory, thus these weaknesses and inherent nature makes them subject to attacks like Denial of Service (DoS) attacks. These attacks on MANETs can reduce or mask the monitoring of health deterioration which can in turn lead to death, immobility or temporary functional disability. There is need to provide resilient security methodologies that do not require enormous computing resources. While entry prevention is the most viable disposition, it is not always possible to stop unauthorized access. Thus, it is critical to investigate use of machine learning based intrusion detection to buttress and provide sufficient security against DOS and other attacks in MANETs. Various anomaly-based intrusion detection systems employ varying techniques to identify anomalies in the context of diverse and valid variables. Most of these techniques however fail to capture and take account the physiognomies of MANETs. In the intervening time, usage of internet of things in the provision of smart healthcare is expanding and the inherent risks snowballing. Attacks aimed at MANETs are increasing to an alarming extent. This study employed a fusion of machine learning techniques through both simulation and a running prototype to achieve a more resilient intrusion detection system. The study was implemented and evaluated on a MANET environment on both Linux NS 2 and further implemented on a network of Smart wearable devices and Raspberry Pi. The results sowed that it is possible to identify and reduce cases of DDOS and blackhole attacks on MANETs by using intrusion detection system improved through machine learning. This study contributed to the body of knowledge in the field intrusion detection systems through ubiquitous learning.