基于HMM-Viterbi算法的认知无线电频谱感知研究
Research on Cognitive Radio Spectrum Sensing Based on the HMM-Viterbi Algorithm
杨娜 1朱敏 1李文轩 1沈晏如 1王诗1
作者信息
- 1. 辽宁工程技术大学电子与信息工程学院，辽宁葫芦岛，125105
- 折叠
摘要
针对认知无线电频谱感知在低信噪比与有限观测条件下性能不足的问题，本文提出了基于隐马尔可夫模型与Viterbi算法的频谱感知方法。为研究该算法的频谱感知性能，本文构建了描述授权用户频谱占用时序特征的隐马尔可夫模型，将感知问题转化为隐藏状态系列的最优解码，采用Baum-Welch算法从含噪声观测中学习模型参数，利用Viterbi算法通过动态规划求解全局最优序列，并针对单时隙部分观测约束提出观测概率适配优化策略。选取感知准确率、检测概率和虚警概率为评价指标，在不同信噪比条件下进行蒙特卡洛仿真实验，与奈曼-皮尔逊检测及相关性聚类算法进行对比分析。仿真结果表明，所提算法在检测概率和感知准确率等指标上对比于传统频谱感知算法具有更好的表现，尤其在-10dB至5dB中低信噪比区间性能优势显著，在0dB时感知准确率达80.3%，相较于对比算法提升15.9个百分点。本文明确了算法的性能边界和适用条件，验证了HMM-Viterbi框架在频谱感知中的可行性，可为认知无线电动态频谱接入提供技术支撑。
Abstract
In response to the issue of insufficient detection performance of cognitive radio spectrum sensing under low signal-to-noise ratio and limited observation conditions, this paper proposes a spectrum sensing method based on the Hidden Markov Model and the Viterbi algorithm. To investigate the spectrum sensing performance of this algorithm, the paper constructs a Hidden Markov Model characterizing the temporal occupancy patterns of the primary user, transforming the sensing problem into an optimal decoding problem of hidden state sequences. The Baum-Welch algorithm is employed to learn model parameters from noisy observations, and the Viterbi algorithm is utilized to obtain the globally optimal state sequence through dynamic programming. Furthermore, an observation probability adaptation strategy is proposed to address the constraint of partial observation within a single time slot. The sensing accuracy, detection probability, and false alarm probability are selected as evaluation metrics, and Monte Carlo simulation experiments are conducted under various signal-to-noise ratio conditions, with comparative analyses against the Neyman-Pearson detection and correlation-based clustering algorithms. The spectrum sensing performance of the proposed algorithm is quantifiable under metrics such as detection probability, false alarm probability, and sensing accuracy. Moreover, in comparison to traditional spectrum sensing algorithms, the proposed algorithm demonstrates superior performance across various indicators, particularly achieving significant advantages in the low-to-moderate signal-to-noise ratio region from -10dB to 5dB, with a sensing accuracy of 80.3% at 0dB, representing a 15.9 percentage point improvement over the benchmark algorithms. This paper clarifies the performance boundaries and applicable conditions of the proposed algorithm, validates the feasibility of the HMM-Viterbi framework in spectrum sensing, and provides technical support for dynamic spectrum access in cognitive radio networks.关键词
认知无线电/频谱感知/隐马尔可夫模型/Viterbi算法/动态规划Key words
Cognitive radio/Spectrum sensing/Hidden Markov Model/Viterbi algorithm/Dynamic programming引用本文复制引用
杨娜,朱敏,李文轩,沈晏如,王诗.基于HMM-Viterbi算法的认知无线电频谱感知研究[EB/OL].(2026-05-09)[2026-05-10].http://www.paper.edu.cn/releasepaper/content/202605-24.学科分类
信息科学、信息技术/控制理论、控制技术/无线通信/电子技术应用
评论