北京邮电大学学报

  • EI核心期刊

北京邮电大学学报 ›› 2015, Vol. 38 ›› Issue (4): 34-38.doi: 10.13190/j.jbupt.2015.04.008

• 论文 • 上一篇    下一篇

云计算下手机人工免疫恶意代码检测模型

武斌1, 林幸1, 李卫东2, 芦天亮3, 张冬梅1   

  1. 1. 北京邮电大学信息安全中心, 北京 100876;
    2. 国网滑县供电公司发展策划部, 河南 安阳 456400;
    3. 中国人民公安大学网络安全保卫学院, 北京 100038
  • 收稿日期:2014-10-01 出版日期:2015-08-28 发布日期:2015-07-03
  • 作者简介:武斌(1981-),男,博士,讲师,E-mail:binwu@bupt.edu.cn.
  • 基金资助:

    国家自然科学基金项目(61101108)

Smartphone Malware Detection Model Based on Artificial Immune System in Cloud Computing

WU Bin1, LIN Xing1, LI Wei-dong2, LU Tian-liang3, ZHANG Dong-mei1   

  1. 1. Information Security Centre, Beijing University of Posts and Telecommunications, Beijing 100876, China;
    2. Development Planning Department, State Grid Huaxian power supply company, Henan Anyang 456400, China;
    3. School of Network Security Defense, People's Public Security University of China, Beijing 100038, China
  • Received:2014-10-01 Online:2015-08-28 Published:2015-07-03

摘要:

提出一种适用于云计算环境的基于人工免疫的手机恶意代码检测模型.提出扩展阴性选择算法,提取恶意代码的特征编码生成抗原,增加针对高亲和度检测器的克隆和变异算子,提高成熟检测器的生成效率,在特征检测和检测器生成阶段引入MapReduce并行处理机制,提高计算效率.仿真结果表明,检测模型对未知手机恶意代码具有较高的检测率和计算效率.

关键词: 人工免疫, 阴性选择, 手机恶意代码, 检测, 云计算

Abstract:

A smartphone malware detection model based artificial immune system(AIS) on the cloud was proposed. In this model, the extended negative selection algorithm is put forward and the antigens are generated by encoding the malwarecharacteristics. With addition of cloning with higher affinity detector and hyper-mutation, the detectors are generated efficiently. The computing rate is then improved significantly by parallel computing mechanism MapReduce during the feature coding and detector generation. Experimentshows that the detection modelhas a high detection rate and computing rate for unknown smartphone malware.

Key words: artificial immune system, negative selection, smartphone malware, detection, cloud computing

中图分类号: