SSSuhyeon Song

Suhyeon Song

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Suhyeon Song has a total of 8 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Mind the Gap: In-Process Isolation for Safe Rust in WebAssembly

    by Suhyeon Song, Chaewon Shin, Donghyun Kwon

  • Open Access

    TzPy: Python Runtime for Edge Code Deployment on TrustZone

    by Suhyeon Song, Jaeyeol Park, Chaewon Shin, Donghyun Kwon

    As edge devices continue to improve in performance, their role is expanding in real-world environments such as smart factories and smart homes, where executing server-provided code on edge devices is essential for real-time processing and operational flexibility. However, edge code deployment introduces several security challenges, including the exposure of sensitive data and operations to runtime attacks, the absence of remote attestation mechanisms to validate code integrity and execution correctness, and insufficient access control for sensitive peripherals. Existing solutions address only a subset of these challenges and therefore fail to provide comprehensive security guarantees for edge devices. In this article, we present TzPy , a Python runtime designed for secure edge code deployment on ARM TrustZone. TzPy enables the execution of Python code entirely within the secure world. In addition, we design a remote attestation mechanism that verifies both the static integrity of deployed Python code and its runtime control flow through control-flow tracing. Furthermore, TzPy enforces a manifest-driven, fine-grained peripheral access control mechanism to prevent unauthorised or unintended access to sensitive devices. We evaluate TzPy on a Raspberry Pi 3B+ using a smart farm workload and a blinky application, and the results show an end-to-end overhead of 13.4%.


  • Open Access

    WARD: Efficient Memory Protection for WebAssembly on Tiny Embedded Systems

    by Chaewon Shin, Gowon Han, Suhyeon Song, Jinsun Park, Yoon-Ho Choi, Donghyun Kwon

    WebAssembly (Wasm) has recently gained attention in embedded systems due to its lightweight runtime support and broad compatibility with various languages and platforms. However, Wasm’s linear memory model lacks internal protection mechanisms, making it vulnerable to a wide range of memory safety violations. To address this, prior work has proposed techniques for protecting memory within the linear memory space. While effective to some extent, these approaches either suffer from limited detection coverage or are unsuitable for resource-constrained embedded systems. In this paper, we present WARD, a redzone-based memory protection technique tailored for Wasm in tiny embedded systems. At its core, WARD employs a randomized linear-to-physical address translation mechanism, which enables effective detection of memory violations without requiring specific hardware extensions. Our evaluation shows that WARD provides improved memory detection coverage over prior techniques, with reasonable performance of 24.2% on benchmark and 13.03% on real-world applications, and memory overhead suitable for embedded systems.

  • Open Access

    WasDom: An Efficient Write Protection for Wasm JITed Code With ARM Domain

    by Suhyeon Song, Chaewon Shin, Donghyun Kwon

    WebAssembly (Wasm) is a binary instruction format designed to run web applications efficiently and securely across different browsers, including Chrome’s V8, Firefox’s SpiderMonkey, and Safari’s JavaScript-Core. While Wasm’s Just-In-Time (JIT) compilation offers significant performance benefits by converting wasm code into machine code (JITed code), it introduces security vulnerabilities by violating the WˆX (Write XOR Execute) policy. Conventional methodologies, such as Intel MPK for safeguarding JITed code, are constrained to specific hardware and are ineffectual in mobile environments. Consequently, there is a necessity for the development of a ARM-compatible solution. This paper proposes WasDom, an efficacious write protection mechanism for wasm JITed code on ARM architecture. Leveraging ARM’s domain-based memory management, WasDom employs a randomized domain allocation strategy to permit multiple cores to access JITed code securely. The prototype of WasDom was implemented in the V8 runtime and demonstrated a minimal performance overhead of less than 11% while providing strong write protection. The system manages memory permissions dynamically through the Domain Access Control Register (DACR), ensuring that memory regions are writable during JIT compilation and executable during runtime, thus enforcing a strong WˆX policy.


  • Open Access

    metaSafer: A Technique to Detect Heap Metadata Corruption in WebAssembly

    by Suhyeon Song, Seonghwan Park, Donghyun Kwon

    WebAssembly (Wasm), a technology enabling efficient native code execution in web browsers, has seen a significant rise in adoption as a popular compilation target. This has led to the emergence of lightweight web services powered by Wasm, characterized by their small binary size and reduced data transfer overhead, thanks to the inherent efficiency of Wasm. Despite their lightweight nature, these services can deliver powerful features like image/video processing, AI and graphical application that surpass the capabilities of JavaScript. To ensure lightweight web services and enhance the overall web experience,Wasm has been extensively optimized. However, these optimizations have raised concerns about memory safety, leading to memory-related vulnerabilities. Wasm’s characteristic memory structure, linear memory, has vulnerabilities that provide various attack vectors to attackers. In particular, it presents various attack possibilities through metadata modification containing memory structure information. Attackers can exploit heap memory overflow in Wasm applications, allowing them to target arbitrary memory addresses, modify data, or execute arbitrary code. Such overflows can corrupt memory metadata, resulting in incorrect memory behavior. While research has mitigate memory-related weaknesses in languages such as C and C++ and architectures like X86 in recent decades, the direct application of security solutions designed for different domains to Wasm is not a practical approach. Consequently, allocators in Wasm remain vulnerable to issues like heap overflow and metadata corruption. Thus, there is a pressing need for tailored memory safety techniques and solutions that accommodateWasm’s architecture-agnostic and linear memory structures. In this paper, we propose metaSafer as a solution. By shadowing metadata from Wasm linear memory to JavaScript virtual machine memory and conducting metadata verification, metaSafer effectively blocks attack attempts and vectors. Notably, our solution achieves fast memory shadowing and validation while maintaining a small code size. Through various verification processes, we measured the performance and code size of metaSafer, revealing that it is a software-only security solution with no additional hardware requirements. metaSafer demonstrates robust metadata protection forWasm applications with an acceptable performance overhead of up to 8% in SQLite [1] speed tests and Polybench [2] benchmarks.