A Novel Hybrid Deep Learning Model for Botnet Attacks Detection in a Secure IoMT Environment *

The Internet of Things (IoT) is a network of linked smart gadgets that provides access to rich data but poses security risks. However, the number of IoT-based botnet attacks has increased due to IoT device limits in processing, storage, and connectivity. A group of online devices managed by cybercriminals is known as a botnet. The IoMT (Internet of Medical Things) requires increased security in contrast to other industries because a security compromise in a healthcare system could directly affect the loss of human life. For IoT botnet malware detection, proposed a hybrid Deep Learning model that combines CNN-BLSTM-GRU for effective multiclass malware family identification. Accuracy, detection rate, and ROC AUC are performance metrics used to assess the performance of our proposed hybrid deep learning model. For IoT-based botnet attack detection, the existing model achieved an accuracy of 98.34%, and the proposed hybrid CNN-BLSTM-GRU Deep Learning-based botnet attack detection system achieves 99.25% accuracy. Future research challenges are noted, as well as the possibility of detecting botnet attacks in the context of IoT/IoMT.