一种基于向量回归的无人机通信信道选择方法
|
摘要
由于无人机通常工作于复杂的电磁环境与多变的地理环境中,因此无人机通信通常采用跳频技术对抗干扰。然而传统的无人机跳频通信仅仅是简单的固定跳频信道接入,如当前使用信道发生用户冲突或者突然衰落时,通信可能会延迟甚至中断。因此,文章结合认知无线电技术(CR),提出一种基于向量回归的无人机认知跳频通信信道选择方法,运用基于向量回归建立预测最小信噪比模型,基于CR建立跳频信道集并在当前跳频通信时同时检测下一跳通信信道状态及动态选择信道,从而保证了跳频接入的可靠性。仿真表明,此方法较传统跳频通信有更高的传输速率和较低的误码率,对无人机的通信质量有较大提升。
Since unmanned aerial vehicles(UAV)are usually employed in complex electromagnetic environments and diverse geographic environments,the frequency-hopping technology is often used in UAV communications to compete interference.However,the traditional frequency-hopping technology is just a fixed access of frequency-hopping communication channel.Once the current using channel has user conflicts or a sudden fading,the communication may delay or interrupt.So,combine with cognitive radio(CR)technology,a frequency-hopping communication channel selection method for UAV based on RVR(Relevance Vector Regression)is addressed.The method based on RVR to establish the model to predict minimum SNR,and constitute the frequency-hopping channel set based on cognitive radio.The next frequency-hopping channel will be estimated meanwhile when the frequency-hopping is going in order to ensure the reliability of the frequency-hopping access.The simulation indicated that the method proposed has better data rate and lower error rate,and an obvious improvement of the communication quality.
Since unmanned aerial vehicles(UAV)are usually employed in complex electromagnetic environments and diverse geographic environments,the frequency-hopping technology is often used in UAV communications to compete interference.However,the traditional frequency-hopping technology is just a fixed access of frequency-hopping communication channel.Once the current using channel has user conflicts or a sudden fading,the communication may delay or interrupt.So,combine with cognitive radio(CR)technology,a frequency-hopping communication channel selection method for UAV based on RVR(Relevance Vector Regression)is addressed.The method based on RVR to establish the model to predict minimum SNR,and constitute the frequency-hopping channel set based on cognitive radio.The next frequency-hopping channel will be estimated meanwhile when the frequency-hopping is going in order to ensure the reliability of the frequency-hopping access.The simulation indicated that the method proposed has better data rate and lower error rate,and an obvious improvement of the communication quality.
引文
[1]VACHTSEVANOS G J,VALAVANIS K P.Handbook of unmanned aerial vehicles[M].Berlin:Springer Netherlands,2014.
[2]张宇航,张晔.SVM和RVM对高光谱图像分类的应用潜能分析[J].哈尔滨工业大学学报,2012,44(3):34-39.
[3]YUAN J,YU T,WANG K S,et al.Adaptive spherical gaussian kernel for fast relevance vector machine regression[C].Proceedings of the 7th World Congress on Intelligent Control and Automation,2008:2071-2078.
[4]SHEN Y,LIU G H,LIU H.Classification method of power quality disturbances based on RVM[C].Proceedings of the 8th World Congress on Intelligent Control and Automation,2010:6130-6135.
[5]董争.基于相关向量机的大规模分类问题的研究[D].广州:华南理工大学,2009.
[6]吴冰.相关向量回归元建模关键技术及其应用研究[D].长沙:国防科学技术大学,2011.
[7]TIPPING M E.Sparse bayesian learning and the relevance vector machine[J].Journal of Machine Learning Research,2001(3):211-244.
[8]HATA M.Empirical formula for propogation loss in land mobile radio services[C].IEEE Transactions on Vehicular Technology,1980,29(3):317-325.
[9]张延华,段占云,沈兰荪,等.Okumura-Hata传播预测模型的可视化仿真研究[J].电波科学学报,2001,16(1):89-92.
[10]张文秋,丁文锐,刘春辉.一种无人机数据链信道选择和功率控制方法[J].北京航空航天大学学报,2017,43(3):583-591.
[2]张宇航,张晔.SVM和RVM对高光谱图像分类的应用潜能分析[J].哈尔滨工业大学学报,2012,44(3):34-39.
[3]YUAN J,YU T,WANG K S,et al.Adaptive spherical gaussian kernel for fast relevance vector machine regression[C].Proceedings of the 7th World Congress on Intelligent Control and Automation,2008:2071-2078.
[4]SHEN Y,LIU G H,LIU H.Classification method of power quality disturbances based on RVM[C].Proceedings of the 8th World Congress on Intelligent Control and Automation,2010:6130-6135.
[5]董争.基于相关向量机的大规模分类问题的研究[D].广州:华南理工大学,2009.
[6]吴冰.相关向量回归元建模关键技术及其应用研究[D].长沙:国防科学技术大学,2011.
[7]TIPPING M E.Sparse bayesian learning and the relevance vector machine[J].Journal of Machine Learning Research,2001(3):211-244.
[8]HATA M.Empirical formula for propogation loss in land mobile radio services[C].IEEE Transactions on Vehicular Technology,1980,29(3):317-325.
[9]张延华,段占云,沈兰荪,等.Okumura-Hata传播预测模型的可视化仿真研究[J].电波科学学报,2001,16(1):89-92.
[10]张文秋,丁文锐,刘春辉.一种无人机数据链信道选择和功率控制方法[J].北京航空航天大学学报,2017,43(3):583-591.