【作者】 张静；

【导师】 林丽媛； 张瑞涛；

【作者基本信息】 天津科技大学 ， 电子信息（专业学位）， 2024， 硕士

【摘要】 水下声学频谱资源稀缺,因此频谱感知技术对水声通信的发展至关重要。将认知无线电领域中的能量检测技术应用与水声通信中,可极大提高水声信号频谱检测性能。但是传统能量检测算法中门限固定,灵活度低。并对能量值处于门限值间的信号予以丢弃或者不做处理,导致信号错检或漏检,最终影响检测概率。另外,由于水下信道环境恶劣,大规模水下无线传感器网络在保障数据传输可靠性方面面临巨大挑战。为此,机会路由(OR)协议因其高可靠性、强鲁棒性、低延迟和高能效等特点,得到了广泛运用。然而,UWSNs中的OR协议容易受到路由攻击的影响。因此,本文主要研究频谱感知算法与入侵检测方案。针对传统感知算法存在的门限值固定,灵活度低的问题,加入权重因子,使得在信噪比不同的环境下,频谱感知算法中的门限可以根据信噪比的不同进而调节,降低误差。并对能量值处于门限值间的信号,采用集中式检测算法,并增加检测次数,每个感知用户独立检测,降低了系统整体误检率,提高了检测效率。针对路由协议易受攻击的问题,本文提出了一种入侵检测方案(IDS),该方案基于密度的噪声应用空间聚类算法(DBSCAN)的入侵检测方案(DOIDS)。在DOIDS中,采用了本地监控机制。DOIDS的网络中运行的每个节点都可以选择受信任的下一跳节点,它对潜在异常节点是否为恶意节点做出最终判断。仿真结果表明,该算法能有效提高检测准确率,并降低误报率。在低信噪比环境下,本文所提出的信噪比加权算法检测精度最低达到64.2%,最高达90%,DOIDS入侵检测算法在不同场景下的检测准确率提高幅度为3%至15%,所提出的两种算法性能均优于现有模型,并且能够显著增强水声通信系统在复杂水下环境中的适应性和稳定性,提高信号传输的可靠性和通信质量,为水下无线通信技术的发展奠定坚实基础。这些算法的不断优化和创新,为未来的水下通信和海洋探索提供更强大的技术支持。更多还原

【Abstract】 Underwater acoustic spectrum resources are scarce,making spectrum sensing technology crucial to the development of underwater acoustic communications.The application of energy detection technology from the field of cognitive radio to underwater acoustic communications can significantly improve the spectrum detection performance of underwater acoustic signals.However,traditional energy detection algorithms have fixed thresholds,resulting in low flexibility.They discard or leave untreated signals with energy values falling within the threshold range,leading to signal misdetection or missed detection and ultimately affecting the detection probability.Due to the harsh underwater channel conditions,large-scale underwater wireless sensor networks face significant challenges in ensuring reliable data transmission.Consequently,Opportunistic Routing(OR)protocols have been widely adopted due to their characteristics of high reliability,strong robustness,low end-to-end delay,and high energy efficiency.Aiming at the problems of fixed threshold and low flexibility in traditional sensing algorithms,a weighting factor is added so that the threshold in the spectrum sensing algorithm can be adjusted according to the signal-to-noise ratio in different environments,reducing errors.For signals with energy values between threshold values,a centralized detection algorithm is adopted,and the number of detections is increased.Each sensing user performs independent detection,which reduces the overall false detection rate of the system and improves detection efficiency.Aiming at the problem that routing protocols are vulnerable to attacks,this paper proposes an intrusion detection scheme(IDS),which is based on the Density-Based Spatial Clustering of Applications with Noise(DBSCAN)algorithm,named DOIDS.In DOIDS,a local monitoring mechanism is adopted.Each node running in the DOIDS network can select trusted next-hop nodes.It makes a final judgment on whether a potentially abnormal node is a malicious node.Simulation results show that the algorithm can effectively improve the detection accuracy and reduce the false alarm rate.In low signal-to-noise ratio(SNR)environments,the detection accuracy of the SNR weighting algorithm proposed in this paper ranges from a minimum of 64.2%to a maximum of 90%.The DOIDS intrusion detection algorithm improves detection accuracy by 3%to 15%in different scenarios.Both proposed algorithms demonstrate superior performance compared to existing models and significantly enhance the adaptability and stability of underwater acoustic communication systems in complex underwater environments.They improve the reliability and communication quality of signal transmission,laying a solid foundation for the development of underwater wireless communication technology.The continuous optimization and innovation of these algorithms provide more powerful technical support for future underwater communication and ocean exploration.更多还原