作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2021, Vol. 47 ›› Issue (1): 196-202. doi: 10.19678/j.issn.1000-3428.0056896

• 移动互联与通信技术 • 上一篇    下一篇

基于单一切面的循环平稳检测方法

勾志杭1,2, 刘剑锋3, 胡金龙2,4,5, 冯雪林2,4,5, 王宗伟1,2   

  1. 1. 重庆邮电大学 通信与信息工程学院, 重庆 400065;
    2. 中国科学院计算技术研究所 无线通信技术研究中心, 北京 100190;
    3. 国家移动卫星通信工程技术研究中心, 南京 210000;
    4. 移动计算与新型终端北京市重点实验室, 北京 100190;
    5. 中国科学院大学 计算机科学与技术学院, 北京 100049
  • 收稿日期:2019-12-12 修回日期:2020-01-14 发布日期:2021-01-13
  • 作者简介:勾志杭(1993-),男,硕士研究生,主研方向为物理层通信算法;刘剑锋,研究员级高级工程师、博士;胡金龙,高级工程师、硕士;冯雪林,高级工程师、博士研究生;王宗伟,硕士研究生。
  • 基金资助:
    北京市自然科学基金(L172049)。

Cyclostationary Detection Method Based on Single Section

GOU Zhihang1,2, LIU Jianfeng3, HU Jinlong2,4,5, FENG Xuelin2,4,5, WANG Zongwei1,2   

  1. 1. College of Communication and Information Engineering, Chongqing University of Posts and Telecommunications, Chongqing 400065, China;
    2. Wireless Communication Technology Research Center, Institute of Computing Technology of Chinese Academy of Sciences, Beijing 100190, China;
    3. National Mobile Satellite Communication Engineering Technology Research Center, Nanjing 210000, China;
    4. Beijing Key Laboratory of Mobile Computing and Pervasive Device, Beijing 100190, China;
    5. School of Computer Science and Technology, University of Chinese Academy of Sciences, Beijing 100049, China
  • Received:2019-12-12 Revised:2020-01-14 Published:2021-01-13

摘要: 在非授权频段的频谱感知方法中,循环平稳检测法具有鲁棒性好与信号类型识别性强的特点,但其计算复杂度很高。在改进传统快速傅里叶变换累加方法(FAM)的基础上,提出一种采用单一切面的循环平稳检测方法。根据循环平稳特性信号峰值变化规律推导出特定切面约束条件,并构建f-切面和α-切面两种循环谱单一切面,在无需建立完整循环谱的情况下对不同类型信号进行循环平稳特征检测。实验结果表明,与传统FAM相比,该方法可大幅降低计算复杂度,当信噪比低于-16 dB时其检测性能略低于传统FAM,在信噪比达到-16 dB后其检测性能与传统FAM一致。

关键词: 非授权频谱, 循环平稳, 频谱感知, 循环谱, 快速傅里叶变换累加方法

Abstract: Among methods for spectrum sensing of unlicensed frequency bands,the cyclostationary detection method is characterized by strong robustness and signal type recognition,but it suffers from high computational complexity.Based on the improved traditional Fast Fourier Transformation Accumulation Method (FAM),this paper proposes a cyclostationary detection method using single section.According to the peak variation law of cyclostationary characteristic signal,the constraint conditions of a specific section are deduced,and two single sections of cyclic spectrum,namely f-section and α-section,are constructed to detect the cyclostationary characteristics of different types of signals without establishing complete cyclic spectrum.Experimental results show that compared with traditional FAM,the proposed method can greatly reduce the computational complexity.When the Signal to Noise Ratio(SNR) is lower than -16 dB,its detection performance is slightly lower than that of the traditional FAM.When the SNR reaches -16 dB,its detection performance is consistent with that of the traditional FAM.

Key words: unlicensed spectrum, cyclostationary, spectrum sensing, cyclic spectrum, Fast Fourier Transformation Accumulation Method(FAM)

中图分类号: