Identification of Cyber-Attacks in IoT-based Healthcare

The IoT has captivated the attention of the scientific and commercial sectors due to its profound influence on human existence. The IoT has lately developed as a cutting-edge platform for building smart environments. The IoT-based healthcare environment is a subsection of the IoTs wherein medical devices transmit data with one another to communicate sensitive data. The cybersecurity of IoT systems has recently been a major topic, particularly in the healthcare industry, where numerous cyber-attacks uncovered devastating IoT cybersecurity risks. Conventional network safety methods are well-established. Nevertheless, because of the limited resource nature of IoT equipment and the peculiar characteristic of IoT standards, traditional security procedures cannot be used effectively for defending IoT systems and networks from cyber-attacks. To improve the IoT security level, there is a need for IoT-specific datasets, tools, and methods. As a result, protecting the IoT-based healthcare setting from cyber-attacks becomes essential. This research's main objective is to illustrate how DRNN and SML approaches (such as RF, RC, DT, and KNN) can indeed be used to create an efficient IDS in the IoT-based healthcare setting for categorizing and predicting unforeseen cyber-attacks. Data from networks are normalized and preprocessed. Then, we used a particle swarm method with a bio-inspired design to improve characteristics. A comprehensive analysis of trials in DRNN and related SML is conducted using conventional statistics for intrusion detection. By extensive analysis, it was determined that the suggested SML model beats current methods with a 99,76 percent accuracy.