《计算机应用》唯一官方网站 ›› 2025, Vol. 45 ›› Issue (10): 3241-3251.DOI: 10.11772/j.issn.1001-9081.2024091331

• 网络空间安全 • 上一篇    

基于自适应动态区间策略的工业控制协议模糊测试方法TDRFuzzer

宗学军1,2(), 韩冰1,2, 王国刚1,2, 宁博伟2,3, 何戡1,2, 连莲1,2   

  1. 1.沈阳化工大学 信息工程学院,沈阳 110142
    2.辽宁省石油化工行业信息安全重点实验室(沈阳化工大学),沈阳 110142
    3.沈阳工业大学 人工智能学院,沈阳 110870
  • 收稿日期:2024-09-20 修回日期:2024-12-19 接受日期:2024-12-26 发布日期:2025-01-14 出版日期:2025-10-10
  • 通讯作者: 宗学军
  • 作者简介:宗学军(1970—),男,辽宁沈阳人,教授,硕士,主要研究方向:工业信息安全 Email:xuejun_zong@syuct.edu.cn
    韩冰(1998—),男,辽宁朝阳人,硕士研究生,主要研究方向:工业信息安全、漏洞挖掘
    王国刚(1977—),男,山东青岛人,教授,博士,主要研究方向:工业自动化优化控制、工业信息安全
    宁博伟(1998—),男,辽宁大连人,博士研究生,主要研究方向:工业信息安全
    何戡(1978—),男,辽宁沈阳人,副教授,硕士,主要研究方向:工业信息安全
    连莲(1981—),女,辽宁丹东人,教授,博士,主要研究方向:控制理论与控制工程、工业信息安全。
  • 基金资助:
    辽宁省科学技术计划项目(2023JH1/10400082);辽宁省人工智能创新发展计划项目(2023JH26/1030008);辽宁省科技创新平台建设计划项目([2022]36号);辽宁省自然科学基金资助项目(2023-MSLH-273)

TDRFuzzer: fuzzing method for industrial control protocols based on adaptive dynamic interval strategy

Xuejun ZONG1,2(), Bing HAN1,2, Guogang WANG1,2, Bowei NING2,3, Kan HE1,2, Lian LIAN1,2   

  1. 1.College of Information Engineering,Shenyang University of Chemical Technology,Shenyang Liaoning 110142,China
    2.Liaoning Key Laboratory of Information Security for Petrochemical Industry (Shenyang University of Chemical Technology),Shenyang Liaoning 110142,China
    3.School of Artificial Intelligence,Shenyang University of Technology,Shenyang Liaoning 110870,China
  • Received:2024-09-20 Revised:2024-12-19 Accepted:2024-12-26 Online:2025-01-14 Published:2025-10-10
  • Contact: Xuejun ZONG
  • About author:ZONG Xuejun, born in 1970, M. S., professor. His research interests include industrial information security.
    HAN Bing, born in 1998, M. S. candidate. His research interests include industrial information security, vulnerability mining.
    WANG Guogang, born in 1977, Ph. D., professor. His research interests include optimization control of industrial automation, industrial information security.
    NING Bowei, born in 1998, Ph. D. candidate. His research interests include industrial information security.
    HE Kan, born in 1978, M. S., associate professor. His research interests include industrial information security.
    LIAN Lian, born in 1981, Ph. D., professor. Her research interests include control theory and control engineering, industrial information security.
  • Supported by:
    Liaoning Provincial Science and Technology Program(2023JH1/10400082);Liaoning Provincial Artificial Intelligence Innovation Development Program(2023JH26/1030008);Liaoning Provincial Natural Science Foundation(2023-MSLH-273);Liaoning Provincial Science and Technology Innovation Platform Construction Program([2022]36)

摘要:

针对模糊测试在工业控制协议(ICP)应用中存在测试用例接受率(TCAR)低和多样性不足等问题,提出一种基于自适应动态区间策略的ICP模糊测试方法。将循环神经网络(RNN)加入Transformer的自注意力机制,以构建协议特征提取模型;使用RNN通过滑动窗口提取数据的局部特征,并引入自注意力机制进行全局特征提取,以保证TCAR;在自注意力块间添加残差连接,以传递权重分数并提高计算效率;生成过程定义动态区间策略,调节模型在任意时间步的采样范围,从而增加测试用例的多样性;在测试过程中构建字段自适应重要性函数,以定位变异关键字段。基于上述方法,设计模糊测试框架TDRFuzzer,并采用Modbus TCP、S7 comm和Ethernet/IP等3种工业协议进行实验评估。结果表明,相较于GANFuzzer、WGANFuzzer和PeachFuzzer这3种模型,TDRFuzzer的TCAR指标显著提高,且漏洞检测率(VDR)分别提高了0.073、0.035和0.150个百分点,表明TDRFuzzer具备更强的ICP漏洞挖掘能力。

关键词: 模糊测试, 工业控制协议, 漏洞挖掘, Transformer, 循环神经网络

Abstract:

Aiming at the problems of low Test Case Acceptance Rate (TCAR) and lack of diversity in application of fuzzing in Industrial Control Protocols (ICPs), a fuzzing method for ICPs based on adaptive dynamic interval strategy was proposed. Recurrent Neural Network (RNN) was added to self-attention mechanism in Transformer to construct a protocol feature extraction model; RNN was used to extract local features of the data through a sliding window, and the self-attention mechanism was introduced to carry out global feature extraction, so as to ensure the TCAR; the residual connection was added between the attention blocks to transfer the weight scores and improve the computational efficiency; a dynamic interval strategy was generated to adjust sampling range of the model at any time step, so as to increase diversity of the test cases; in the testing process, the field adaptive importance function was constructed to locate the key variant fields. Based on the above method, a fuzzing framework TDRFuzzer was designed and experimentally evaluated using three industrial protocols: Modbus TCP, S7 comm, and Ethernet/IP. The results show that compared to three models: GANFuzzer, WGANFuzzer, and PeachFuzzer, TDRFuzzer has the TCAR increased significantly, and the Vulnerability Detection Rate (VDR) increased by 0.073, 0.035, and 0.150 percentage points, respectively. This indicates that TDRFuzzer has stronger vulnerability mining capability for ICPs.

Key words: fuzzing, Industrial Control Protocol (ICP), vulnerability mining, Transformer, Recurrent Neural Network (RNN)

中图分类号: