Detektionsmetoder för skadlig kod i IoT-baserat smart hem : En systematisk litteraturstudie

Sammanfattning: IoT devices are being widely deployed within smart homes. Most of these devices are mass-produced at a low cost. As a result, due to the lack of security mechanisms, IoT devices become vulnerable to malware. As more IoT devices are connected to the internet, and given their inability to maintain robust security, these devices are at an increased risk of being infected with malware. Compromised IoT devices enhance the capabilities of cybercriminals and threat actors to perform attacks and distribute malware. To prevent this, proper detection mechanisms are needed. However, traditional malware detection approaches are often not feasible in an IoT environment. This study compiles current detection methods used to detect IoT-malware in smart homes. Existing malware detection solutions will be included to demonstrate the methods, usage, and effectiveness in a specific context. This was achieved by performing a qualitative systematic literature review of articles from two databases with high technological relevance. In total, 12 articles were utilized for the study. The data from these articles were subject to a thematic analysis, yielding two main themes: method and placement. The “method” theme consists of four categories: anomaly detection, signature detection, statistical analysis, and combination of methods. The “placement” theme consists of two categories: device-based and network-based. The study results indicate that both standalone methods and a combination of multiple methods are being employed for the detection of IoT-malware in smart home environments. Based on the results, anomaly-based detection emerges as the most used method for detecting IoT-malware, both on the device and within the network.