Search

In recent years, the rapid development of computer software has led to numerous security problems, particularly software vulnerabilities. These flaws can cause significant harm to users' privacy and property. Current security defect detection technology relies on manual or professional reasoning, leading to missed detection and high false detection rates. Artificial intelligence technology has led to the development of neural network models based on machine learning or deep learning to intelligently mine holes, reducing missed alarms and false alarms. So, this project aims to study Java source code defect detection methods for defects like null pointer reference exception, XSS (Transform), and Structured Query Language (SQL) injection. Also, the project uses open-source Javalang to translate the Java source code, conducts a deep search on the AST to obtain the empty syntax feature library, and converts the Java source code into a dependency graph. The feature vector is then used as the learning target for the neural network. Four types of Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), Bi-directional Long Short-Term Memory (BiLSTM), and Attention Mechanism + Bidirectional LSTM, are used to investigate various code defects, including blank pointer reference exception, XSS, and SQL injection defects. Experimental results show that the attention mechanism in two-dimensional BLSTM is the most effective for object recognition, verifying the correctness of the method. [ABSTRACT FROM AUTHOR]

As the number of Internet of Thing (IoT) devices increases, attacks against their vulnerabilities have become a serious threat. The web servers (WSs) in IoT devices provide management services for end-users, which are currently the major attack surface. Several fuzzing solutions for identifying vulnerabilities in IoT devices have been proposed, but there is currently no grey-box fuzzer specifically designed for the unique features of WSs in IoT to effectively detect memory corruption vulnerabilities. We design and implement KVFL, an efficient grey-box fuzzer, to address the issues of low throughput and slow exploration of deep code when fuzzing for IoT WSs. Firstly, KVFL employs a delicate hooking technology that heuristically hijacks and emulates hardware-dependent functions, ensuring WSs can be accurately and efficiently emulated in user-mode. On this basis, KVFL fully utilizes the loop parsing HTTP requests feature of WSs through a redesigned fork-server, to minimize nonessential rebooting losses of the target, thereby significantly improving fuzzing throughput. Secondly, KVFL leverages code coverage feedback to automatically infer a set of valid Keys and derive a Key-Value mutation. This enables the generation of high-quality test cases that can facilitate deeper code exploration of WSs. The evaluation results show that compared to the state-of-the-art IoT grey-box fuzzer FIRM-AFL, KVFL improves the throughput by over 2× and explores 4.5× more edges. Additionally, it identifies all 1-day vulnerabilities with over 7× faster speed than the baseline and detects three previously unknown 0-day vulnerabilities. These all indicate that KVFL is effective and efficient at fuzzing IoT WSs. [ABSTRACT FROM AUTHOR]