CRJan 23, 2015

Learning Execution Contexts from System Call Distributions for Intrusion Detection in Embedded Systems

arXiv:1501.05963v2
Originality Incremental advance
AI Analysis

This work addresses intrusion detection for embedded systems, offering a domain-specific solution that is incremental in its approach.

The paper tackles the problem of intrusion detection in embedded systems by proposing a lightweight method that uses system call frequency distributions and cluster analysis to learn legitimate execution contexts and detect anomalies at runtime, with a prototype showing effective detection without affecting critical execution paths.

Existing techniques used for intrusion detection do not fully utilize the intrinsic properties of embedded systems. In this paper, we propose a lightweight method for detecting anomalous executions using a distribution of system call frequencies. We use a cluster analysis to learn the legitimate execution contexts of embedded applications and then monitor them at run-time to capture abnormal executions. We also present an architectural framework with minor processor modifications to aid in this process. Our prototype shows that the proposed method can effectively detect anomalous executions without relying on sophisticated analyses or affecting the critical execution paths.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes