车联网无线通信技术作为智能交通系统中的重要组成部分，为各类车载通信业务以及娱乐信息服务提供了可靠的技术支撑。随着车联网业务种类和通信需求的爆发式增长，导致车联网面临频谱资源的严重短缺和通信时延等问题。认知无线电技术通过感知无线电环境中的频谱空穴，允许认知用户机会式接入空闲授权频谱，为车联网通信提供更多可利用的频谱资源。可靠的频谱感知是实现授权频谱资源再利用的关键，因而对认知车联网中的频谱感知算法研究十分必要。

针对单车辆认知车联网场景，传统的频谱感知方法受到主用户频谱活动特性和车辆移动特性的影响，导致检测性能下降的问题，提出一种基于Markov模型的单车辆频谱感知方法。该方法利用二态马尔可夫链构建主用户频谱活动统计模型，根据历史统计数据计算主用户频谱状态转移概率；接着联合考虑认知车辆的移动特性，得到主用户发射信号位于认知车辆感知范围内的内部概率，最终推导出所提方法的系统漏检概率的表达式。仿真结果表明，在相同的信噪比条件下，所提方法相较于传统方法具有更低的漏检概率，提高了单车辆系统的感知性能。

针对多车辆认知车联网场景，传统的协作频谱感知算法存在复杂度较高且在低信噪比条件下检测概率较低的问题，提出一种基于CM-ResNet的认知车联网协作频谱感知改进算法。算法结合深度学习理论将频谱感知转化为图像二分类问题，首先将由多个认知车辆组成的本地观测矩阵进行协方差运算，接着将二维矩阵归一化灰度图像处理后转化为输入到网络模型的图像数据，通过训练残差神经网络提取灰度图像中的特征，最终将训练完成的网络模型作为分类器完成在线频谱感知任务。仿真结果表明，在信噪比低至−15dB时，所提算法的检测概率达到98.2%，且在不同条件下所提算法的性能均优于传统算法，验证了所提算法的可行性。

As an important part of the intelligent transportation system, the wireless communication technology of the Internet of Vehicles provides reliable technical support for all kinds of vehicle communication services and entertainment information services. With the explosive growth of vehicle networking service types and communication requirements, the vehicle networking is facing serious shortage of spectrum resources and communication delay. By sensing the spectrum holes in the radio environment, cognitive radio technology allows cognitive users to opportunistically access the idle licensed spectrum, providing more available spectrum resources for vehicular network communication. Reliable spectrum sensing is the key to realize the reuse of licensed spectrum resources, so it is necessary to study the spectrum sensing algorithm in cognitive vehicular networks.

Aiming at the problem that the traditional spectrum sensing method is affected by the spectrum activity characteristics of the primary user and the moving characteristics of the vehicle, which leads to the decrease of detection performance in the single vehicle cognitive vehicular networking scenario, a single vehicle spectrum sensing method based on Markov model is proposed. This method uses the two-state Markov chain to construct the statistical model of the primary user 's spectrum activity, and calculates the primary user 's spectrum state transition probability according to the historical statistical data. Then, considering the mobility characteristics of the cognitive vehicle, the internal probability of the transmitted signal of the primary user within the perception range of the cognitive vehicle is obtained. Finally, the expression of the system miss detection probability of the proposed method is derived. The simulation results show that under the same signal-to-noise ratio, the proposed method has a lower missed detection probability than the traditional method, and improves the sensing performance of the single vehicle system.

For the multi-vehicle cognitive vehicular network scenario, the traditional cooperative spectrum sensing algorithm has the problems of high complexity and low detection probability under low signal-to-noise ratio conditions. An improved cooperative spectrum sensing algorithm based on CM-ResNet is proposed. The algorithm combines deep learning theory to transform spectrum sensing into image binary classification problem. Firstly, the local observation matrix composed of multiple cognitive vehicles is subjected to covariance operation. Then, the normalized gray image of the two-dimensional matrix is transformed into image data input to the network model. The features in the gray image are extracted by training the residual neural network. Finally, the trained network model is used as a classifier to complete the online spectrum sensing task. The simulation results show that the detection probability of the proposed algorithm reaches 98.2 % when the signal-to-noise ratio is as low as -15dB, and the performance of the proposed algorithm is better than that of the traditional algorithm under different conditions, which verifies the feasibility of the proposed algorithm.

[1]李百炼, 伍业钢. 智慧生态城市：内涵、架构与运行机理[J]. 国家治理, 2022(24):42-47.

[2]Hou Lingchun, Wang Yuanping, Zheng Yingheng, et al. The Impact of Vehicle Ownership on Carbon Emissions in the Transportation Sector[J]. Sustainability, 2022, 14(19):12657-12657.

[3]Lidkea V, Muresan R, Al-Dweik A, et al. Improving the Security of Cloud-Based Intelligent Transportation Systems[C]// 2019 IEEE Canadian Conference of Electrical and Computer Engineering (CCECE), Edmonton, AB, Canada, 2019:1-4.

[4]Machardy Z, Khan A, Obana K, et al. V2X Access Technologies: Regulation, Research, and Remaining Challenges[J]. IEEE Communications Surveys & Tutorials, 2018, 20(3):1858-1877.

[5]Daniel A, Paul A, Ahmad A, et al. Cooperative Intelligence of Vehicles for Intelligent Transportation Systems (ITS)[J]. Wireless Personal Communications, 2016, 87(2):461-484.

[6]Zadobrischi Eduard, Dimian Mihai, Negru Mihai. The Utility of DSRC and V2X in Road Safety Applications and Intelligent Parking: Similarities, Differences, and the Future of Vehicular Communication[J]. Sensors, 2021, 21(21):7237-7237.

[7]Naik G, Choudhury B, Park J. IEEE 802.11bd & 5G NR V2X: Evolution of Radio Access Technologies for V2X Communications[J]. IEEE Access, 2019, 7(1):70169-70184.

[8]Xiaoqin Song, Juanjuan Miao, Lei Lei, et al. Maximum-Throughput Sidelink Resource Allocation for NR-V2X Networks with the Energy-efficient CSI Transmission[J]. IEEE Access, 2020, 8:1-1.

[9]Wang B, Liu K J R. Advances in Cognitive Radio Networks: A Survey[J]. IEEE Journal of Selected Topics in Signal Processing, 2011, 5(1):5-23.

[10]Singh K, Rawat P, Bonnin J M. Cognitive Radio for Vehicular Ad Hoc Networks (CR-VANETs): Approaches and Challenges[J]. EURASIP Journal on Wireless Communications and Networking, 2014, 7(1):49-61.

[11]Chembe C, Noor R M, Ahmedy I, et al. Spectrum Sensing in Cognitive Vehicular Network: State-of-Art, Challenges and Open Issues[J]. Computer Communications, 2017, 97(1):15-30.

[12]黄钟. 车联网频谱资源管理关键技术研究[D]. 电子科技大学, 2020.

[13]Holly Gilbert Stowell. Preserving the Safety Spectrum[J]. Institute of Transportation Engineers. ITE Journal, 2020, 90(4):37-42.

[14]Hideya TAKEO, Yusuke TAKATORI. Performance Analysis of a Lane Change Decision Aid Systems Based on Vehicle Infrastructure Integration by Microscopic Traffic Simulation[J]. A - Abstracts of IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences (Japanese Edition), 2013, 96(9):650-660.

[15]Hyungjun Park, Brian L. Smith. Investigating Benefits of IntelliDrive in Freeway Operations: Lane Changing Advisory Case Study[J]. Journal of Transportation Engineering, 2012, 138(9):1113-1122.

[16]Zhaoxiang He, Xiao Qin, Pan Liu, et al. Assessing Surrogate Safety Measures using a Safety Pilot Model Deployment Dataset[J]. Transportation Research Record, 2018, 2672(38):1-11.

[17]Mumford, Richard. DRIVE C2X Project Update[J]. Microwave Journal, 2012, 55(8):45.

[18]Tatsuru DAIMON, Hiroshi MAKINO, Hiroyuki MIZUTANI, et al. Study on Safety Assist Information of Advanced Cruise-Assist Highway Systems (AHS) using VICS in Blind Curve Section of Urban Expressway[J]. Journal of Mechanical Systems for Transportation and Logistics, 2008, 1(2):192-202.

[19]关积珍. 智能交通系统发展演进及其代际特征[J]. 人工智能, 2022(04):40-49.

[20]攸连庆, 卜瑞, 时丹. 试论智能网联与未来汽车的技术变革[J]. 汽车工业研究, 2022(03):28-31.

[21]《中国公路学报》编辑部. 中国交通工程学术研究综述[J]. 中国公路学报, 2016, 29(06):1-161.

[22]李凤, 房家奕, 赵丽. 3GPP LTE-V2X标准进展及技术介绍[J]. 电信网技术, 2016(06):40-45.

[23]王飞飞, 刘梦梦, 翟晓兰. 智能网联汽车测试示范区发展模式探究[J]. 汽车实用技术, 2022, 47(13):26-30.

[24]郭俊. 我国智能网联汽车政策现状与政策趋势浅析[J]. 科技与创新, 2018(03):43-45.

[25]马小博, 彭嘉豪, 薛磊等. 5G时代车联网信息物理融合系统综合安全研究[J]. 中国科学:信息科学, 2019, 49(12):1640-1658.

[26]ShahnawazShaikh Md., Gupta Kamlesh. A Review of Spectrum Sensing Techniques for Cognitive Radio[J]. International Journal of Computer Applications, 2014, 94(8):1-5.

[27]Guowei Zhang, Zhang Guowei. Research on Cognitive Radio Spectrum Sensing Security Mechanism Based on Blockchain[J]. Journal of Physics: Conference Series,2020,1578(1): 012045.

[28]Y. Arjoune, Z. E. Mrabet, H. E. Ghazi, et al. Spectrum sensing: Enhanced energy detection technique based on noise measurement[C]// 2018 IEEE 8th Annual Computing and Communication Workshop and Conference (CCWC). IEEE, 2018: 828-834.

[29]Xin Lin Huang, Yuan Xu, Jun Wu, et al. Noncooperative Spectrum Sensing with Historical Sensing Data Mining in Cognitive Radio[J]. IEEE Transactions on Vehicular Technology, 2016, 66(10):8863-8871.

[30]李涛, 赵知劲. 基于功率谱和双门限的频谱感知算法[J]. 杭州电子科技大学学报, 2022, 42(05):28-33.

[31]倪水平, 常慧刚. 基于能量和小波变换的双门限联合频谱感知[J]. 北京邮电大学学报, 2017, 40(04):117-121.

[32]Chang Liu, Jie Wang, Xuemeng Liu, et al. Deep CM-CNN for Spectrum Sensing in Cognitive Radio[J]. IEEE Journal on Selected Areas in Communications, 2019, 37(10):2306-2321.

[33]Geng Yue, Huang Jingyi, Yang Jianxin, et al. Spectrum Sensing for Cognitive Radio Based on Feature Extraction and Deep Learning[J]. Journal of Physics: Conference Series, 2022, 2261(1).

[34]薛建伟, 唐亮, 卜智勇. 基于神经网络的认知无线电合作频谱感知[J]. 中国科学院大学学报, 2020, 37(03): 379-386.

[35]陈月, 张希, 艾文宝. 基于残差神经网络的频谱感知算法[J]. 现代电子技术, 2022,45(07):1-5.

[36]Anal Paul, Priyatham Kunarapu, Avik Banerjee, et al. Spectrum sensing in cognitive vehicular networks for uniform mobility model[J]. IET Communications, 2019, 13(19):3127-3134.

[37]张月霞, 赵义飞. 车联网中基于BP神经网络的智能协同频谱感知算法[J]. 电讯技术, 2021, 61(08):919-924.

[38]尚佳庆, 郑国强, 马华红等. 基于双门限能量检测的认知车联网协作频谱感知[J]. 计算机应用研究, 2020, 37(10):3131-3134+3138.

[39]纪玉峰, 郑敏, 谭冲等. 基于神经网络的车联网频谱感知组合算法[J]. 中国科学院大学学报, 2021, 38(04):549-556.

[40]林惠娇, 胡斌杰, 齐嘉杰. 一种门限自适应的认知车联网频谱感知算法[J]. 移动通信, 2019, 43(08):77-82.

[41]Joy Eze, Sijing Zhang, Enjie Liu, Elias Eze. Cognitive radio-enabled Internet of Vehicles: a cooperative spectrum sensing and allocation for vehicular communication[J]. IET Networks, 2018, 7(4):190-199.

[42]Aygun Bengi, Wyglinski Alexander M. A Voting-Based Distributed Cooperative Spectrum Sensing Strategy for Connected Vehicles[J]. IEEE Transactions on Vehicular Technology, 2017, 66(6):5109-5121.

[43]Chen Lingling, Fu Ke, Zhao Quanjun, et al. A multi-channel and multi-user dynamic spectrum access algorithm based on deep reinforcement learning in Cognitive Vehicular Networks with sensing error[J]. Physical Communication, 2022, 55.

[44]Zhang Liang, Li Zengqi, Jia Min. Compressed spectrum sensing for grant-free NOMA based internet of vehicles[J]. EURASIP Journal on Advances in Signal Processing, 2022, 2022(1).

[45]Hossain Md Shamim, Miah Md Sipon. Machine learning-based malicious user detection for reliable cooperative radio spectrum sensing in Cognitive Radio-Internet of Things[J]. Machine Learning with Applications, 2021, 100052.

[46]Eze E C, Zhang S J, Liu E J, et al. Advances in Vehicular Ad-Hoc Networks (VANETs): Challenges and Road-Map for Future Development[J]. International Journal of Automation and Computing, 2016, 13(1):1-18.

[47]Xinyu Wang, Min Jia, Qing Guo, et al. Detection Performance Analysis of Spectrum Sensing in Cognitive Radio Networks with Mobile Secondary Users[J]. 中国通信, 2016, 13(11):214-225.

[48]Sharma G, Sharma R. Energy Efficient Collaborative Spectrum Sensing with Clustering of Secondary Users in Cognitive Radio Networks[J]. IET Communications, 2019, 13(8):1101-1109.

[49]郭彩丽, 陈九九, 宣一荻等. 动态时空数据驱动的认知车联网频谱感知与共享技术研究[J]. 物联网学报, 2020, 4(03):96-105.

[50]Yasir Saleem, Mubashir Husain Rehmani. Primary radio user activity models for cognitive radio networks: A survey[J]. Journal of Network and Computer Applications, 2014, 43:1-16.

[51]Viriyasitavat W, Boban M, Tsai H M, et al. Vehicular Communications: Survey and Challenges of Channel and Propagation Models[J]. IEEE Vehicular Technology Magazine, 2015, 10(2):55-66.

[52]Majid H. Khoshafa, Saad Al-Ahmadi. On the Capacity of Underlay Cognitive Radio Networks Over Shadowed Multipath Fading Channels[J]. Arabian Journal for Science and Engineering,2017,42(12):5191-5199.

[53]Alexandros G. Fragkiadakis, Elias Z. Tragos, Ioannis G. Askoxylakis. A Survey on Security Threats and Detection Techniques in Cognitive Radio Networks[J]. IEEE Communications Surveys and Tutorials, 2013, 15(1):428-445.

[54]Zhu S, Guo C, Feng C, et al. User Correlation and Double Threshold Based Cooperative Spectrum Sensing in Dense Cognitive Vehicular Networks[A]. International Symposium on Wireless Communication Systems[C]. Poznan: IEEE, 2016:281-285.

[55]Yasir Saleem, Mubashir Husain Rehmani, Sherali Zeadally. Integration of Cognitive Radio Technology with unmanned aerial vehicles: Issues, opportunities, and future research challenges[J]. Journal of Network and Computer Applications, 2015, 50:15-31.

[56]Kaleem A, Zhang J, Irfan H, et al. Energy Detection Based Spectrum Sensing Strategy for CRN[A]. 2020 IEEE International Conference on Artificial Intelligence and Information Systems (ICAIIS) [C]. Dalian: IEEE, 2020:107-112.

[57]Zhang X, Gao F, Chai R, et al. Matched Filter Based Spectrum Sensing When Primary User Has Multiple Power Levels[J]. China Communications, 2015, 12(2):21-31.

[58]Alqawasmeh A F, Awin F A, Salout N, et al. Low Complexity Energy-efficient multilevel Cluster-based Cooperative Spectrum Sensing in Cognitive Radio Networks[C]// 2017 IEEE International Symposium on Signal Processing and Information Technology(ISSPIT), IEEE, 2017:117-121.

[59]Joy E, Sijing Z, Enjie L, et al. Cognitive Radio-Enabled Internet of Vehicles: a Cooperative Spectrum Sensing and Allocation for Vehicular Communication[J]. IET Networks, 2018, 7(4):190-199.

[60]Tian J,Jiang Y, Chen H, et al. Sensing of OFDM Signals with Cyclic Delay Diversity over Doubly Selective Fading Channels[J]. Transactions on Emerging Telecommunications Technologies, 2012, 23(3):234-239.

[61]Alqawasmeh A F, Awin F A, Salout N, et al. Low Complexity Energy-efficient multilevel Clusterbased Cooperative Spectrum Sensing in Cognitive Radio Networks[C]// 2017 IEEE International Symposium on Signal Processing and Information Technology(ISSPIT), IEEE, 2017:117-121.

[62]Cacciapuoti A S, Akyildiz I F, Paura L. Correlation-Aware User Selection for Cooperative Spectrum Sensing in Cognitive Radio Ad hoc Networks[J]. IEEE Journal on Selected Areas in Communications, 2012, 30(2):297-306.

[63]Alhamad R, Boujemaa H. Cooperative Spectrum Sensing with Relay Selection[J]. Telecommunication Systems, 2017, 68(4):631-642.

[64]Aziz A M. A New Adaptive Decentralized Soft Decision Combining Rule for Distributed Sensor Systems with Data Fusion[M]. A new adaptive decentralized soft decision combining rule for distributed sensor systems with data fusion, 2014, 256(1):197-210.

[65]Mohammad F R, Ciuonzo D, Mohammed Z A K. Mean-Based Blind Hard Decision Fusion Rules[J]. IEEE Signal Processing Letters, 2018, 25(5):630-634.

[66]Gu J, Wang Z, Kuen J, et al. Recent Advances in Convolutional Neural Networks[J]. Pattern Recognition, 2018, 77(1):354-377.

[67]Lecun Y, Bengio Y, Hinton G. Deep learning[J]. Nature, 2015, 521(7553):436-444.

[68]周飞燕, 金林鹏, 董军. 卷积神经网络研究综述[J]. 计算机学报, 2017, 40(06):1229-1251.

[69]刘霞. 认知车联网频谱感知技术研究[D]. 北京邮电大学, 2019.

[70]祝思婷. 车辆密集场景下的认知车联网频谱感知研究[D]. 北京邮电大学, 2017.