As the Internet of Things digital systems have been becoming increasingly complex, so are the cybersecurity issues, which, in turn, have been heightened by the malicious activities of Portable Executable (PE) malware and Distributed Denial of Service (DDoS) attacks. These kinds of threats constitute a significant challenge to the integrity, confidentiality, and availability of connected systems; hence, an immediate need has been identified to design effective detection mechanisms. The development of IoT-based devices has moved at a galloping speed, which was accompanied by enormous security issues of these networks with respect to potential cyber threats caused by PE malware or DDoS. The PE malware exploits vulnerabilities in executable files to breach IoT devices, which most of the time leads to unauthorized access and system compromise. However, DDoS attacks target IoT networks by causing a flood of malicious traffic against them, thereby resulting in service interruptions and great damage. The proposed system uses Multivariate Correlation Analysis (MCA) to evaluate network flow based on the extraction and analysis of spatial relationships among traffic attributes. MCA can prove to be an effective anomaly-based detection mechanism in the case of malicious behavior. The proposed system ensures scalability and real-time capability in terms of detection offered by it by using lightweight machine learning models and combining them with MCA. This strategy intends to enhance IoT environment security through a reliable framework for identifying PE malware and counteracting DDoS attacks. © 2025 Elsevier B.V., All rights reserved.