An Internet of Things (IoT) Consumer Data Network Security Assessment that Uses Variant Mode Decomposition (VMD) and Fused CNN-Attention Bi- LSTM

More and more people are able to access the internet and participate in social media, leaving most companies open to a variety of threats and assaults. Consequently, intrusion detection systems (IDSs) are seen as a crucial part of network security for organizations. Their job is to detect any network breaches as soon as they occur, making them the first line of defense towards cyber-attacks. Intrusion detection systems primarily seek for harmful activity in the network by analyzing its traffic. Due to the ever-increasing demand for network services, networks nowadays are experiencing massive expansion. The created algorithms may find their application limited as a result of the different data kinds and increased network complexity caused by this growth. Also, the frequency and severity of viruses and other harmful assaults are evolving. Cybercriminals are more interested than ever before in the IoT due to its phenomenal development pace. The assertion is supported by the increasing frequency of cyberattacks targeting intermediary communication media and Internet of Things devices. Severe service disruption and financial loss may occur from attacks on the Internet of Things (IoT) that go unnoticed for a long time. Additionally, it presents a risk to one's privacy. For IoT-enabled services to be dependable, safe, and lucrative, real-time intrusion detection on IoT devices is crucial. New intrusion detection solution for Internet of Things (IoT) devices based on Deep Learning (DL) is presented in this study. To prevent assaults on linked Internet of Things (IoT) devices, this smart system employs a four-layer deep Fully Connected (FC) designing networks to identify malicious data. For easier implementation, the suggested system was built to not rely on any particular communication protocol.