深空Ka频段链路的自适应传输及编码策略研究
发布时间：2018-08-06 14:56
【摘要】：具有高传输速率和大信道带宽的Ka(28~35 GHz)频段越来越多地应用在深空通信业务中。然而,深空Ka频段链路质量受太阳上合段太阳闪烁和近地段降雨的影响,链路的误码率(Bit Error Rate,BER)波动较大,限制了链路的高速率传输。基于此,本课题引入Markov理论对深空Ka频段链路的动态特性进行研究建模,并考虑发送端无法获得实时信道状态信息(Channel State Information,CSI)的限制条件,用时间序列分析法中的自回归移动平均算法(Autoregressive-moving-average,ARMA)对信道状态进行预测。由于预测误差的客观存在,使用能够抗信道估计偏差的模拟喷泉编码(Analog fountain codes,AFC)作为前向纠删技术,结合预测算法对深空Ka频段链路质量进行联合优化,确定了适应于深空Ka频段的自适应编码传输方案。具体的研究内容如下:研究深空通信太阳闪烁信道和Ka频段降雨衰落信道的动态传输,两者都符合Markov转移特性。用Markov链建模深空通信太阳闪烁信道和降雨衰落信道,并仿真对比模拟衰减时序数据与实测衰减时序数据的误差,验证了模型的可行性。利用建好的Markov链产生模拟太阳闪烁数据和模拟雨衰数据,作为后续深空通信Ka频段链路衰减的实时预测研究基础。考虑深空通信长时延情况无法获得实时CSI,课题采用预测效果好并且算法步骤简单的ARMA对Ka频段链路的太阳闪烁和降雨衰减进行实时动态预测。利用“太阳闪烁Markov链”和“降雨衰减Markov链”产生数据对ARMA算法进行模型识别和参数选择。仿真验证了该预测算法可以取得很好的预测精度,同时分析了前导数据个数和预测步长对预测精度的影响。虽然ARMA实时预测算法具有很好的预测精度,但预测误差是始终存在的。为了抵抗信道状态信息估计偏差对传输性能的影响,本课题采用在较大信噪比(Signal-to-Noise Ratio,SNR)范围内具有良好译码性能的AFC作为前向纠删分组技术。并将其与链路预测算法相结合,对深空通信Ka链路的误码率波动进行优化,确定了深空Ka频段自适应传输编码方案。理论分析了自适应编码传输方案的吞吐率,仿真分析了该方案的吞吐率,并将其与固定速率编码传输方案以及自适应LT码传输方案的吞吐率进行对比,本课题提出方案的吞吐率达到最高,从而验证了该方案可以保持深空Ka频段的数据连续性以及链路的高吞吐率。
[Abstract]:The Ka (28N 35 GHz) band with high transmission rate and large channel bandwidth is increasingly used in deep space communication services. However, the link quality in the deep space Ka band is affected by solar flicker in the upper solar segment and rainfall in the near area. The BER (bit error rate) of the link fluctuates greatly, which limits the high rate transmission of the link. Based on this, this paper introduces Markov theory to model the dynamic characteristics of deep space Ka-band links, and considers the limitation condition that the sender can not obtain real-time channel state information (Channel State). The autoregressive moving average (ARMA) algorithm in time series analysis is used to predict the channel state. Because of the objective existence of prediction error, the analog fountain code (Analog fountain codes-AFC, which can resist channel estimation deviation, is used as the forward erasure technique, and the link quality in deep space Ka band is jointly optimized by combining the prediction algorithm. The adaptive coding transmission scheme for deep space Ka band is determined. The main contents are as follows: the dynamic transmission of solar scintillation channel and Ka-band rainfall fading channel in deep space communication is studied. Both of them accord with the Markov transfer characteristics. The solar scintillation channel and rainfall fading channel in deep space communication are modeled by Markov chain, and the error between simulated attenuation time series data and measured attenuation time series data is compared to verify the feasibility of the model. Simulated solar scintillation data and simulated rain attenuation data are generated by using the constructed Markov chain, which is the basis of real-time prediction of Ka-band link attenuation in subsequent deep space communications. Considering that long time delay can not be obtained in deep space communication, ARMA, which has good prediction effect and simple algorithm steps, is used to dynamically predict solar flicker and rainfall attenuation in Ka band link. The model identification and parameter selection of the ARMA algorithm are carried out by using the data generated by "solar scintillation Markov chain" and "rainfall attenuated Markov chain". The simulation results show that the prediction algorithm can achieve good prediction accuracy. At the same time, the influence of the number of leading data and the prediction step size on the prediction accuracy is analyzed. Although the ARMA real-time prediction algorithm has a good prediction accuracy, but the prediction error is always there. In order to resist the influence of channel state information estimation error on transmission performance, the AFC with good decoding performance in the range of large signal-to-noise ratio (Signal-to-Noise) is adopted as the forward erasure correction block technique. Combining with the link prediction algorithm, the error rate fluctuation of Ka link in deep space communication is optimized, and the adaptive transmission coding scheme in deep space Ka band is determined. The throughput of the adaptive coding transmission scheme is analyzed theoretically, and the throughput of the scheme is simulated and analyzed, and the throughput of the scheme is compared with that of the fixed rate coding transmission scheme and the adaptive LT transmission scheme. In this paper, the throughput of the proposed scheme is the highest, which verifies that the scheme can maintain the data continuity in the deep space Ka band and the high throughput of the link.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN927
本文编号：2168080
[Abstract]:The Ka (28N 35 GHz) band with high transmission rate and large channel bandwidth is increasingly used in deep space communication services. However, the link quality in the deep space Ka band is affected by solar flicker in the upper solar segment and rainfall in the near area. The BER (bit error rate) of the link fluctuates greatly, which limits the high rate transmission of the link. Based on this, this paper introduces Markov theory to model the dynamic characteristics of deep space Ka-band links, and considers the limitation condition that the sender can not obtain real-time channel state information (Channel State). The autoregressive moving average (ARMA) algorithm in time series analysis is used to predict the channel state. Because of the objective existence of prediction error, the analog fountain code (Analog fountain codes-AFC, which can resist channel estimation deviation, is used as the forward erasure technique, and the link quality in deep space Ka band is jointly optimized by combining the prediction algorithm. The adaptive coding transmission scheme for deep space Ka band is determined. The main contents are as follows: the dynamic transmission of solar scintillation channel and Ka-band rainfall fading channel in deep space communication is studied. Both of them accord with the Markov transfer characteristics. The solar scintillation channel and rainfall fading channel in deep space communication are modeled by Markov chain, and the error between simulated attenuation time series data and measured attenuation time series data is compared to verify the feasibility of the model. Simulated solar scintillation data and simulated rain attenuation data are generated by using the constructed Markov chain, which is the basis of real-time prediction of Ka-band link attenuation in subsequent deep space communications. Considering that long time delay can not be obtained in deep space communication, ARMA, which has good prediction effect and simple algorithm steps, is used to dynamically predict solar flicker and rainfall attenuation in Ka band link. The model identification and parameter selection of the ARMA algorithm are carried out by using the data generated by "solar scintillation Markov chain" and "rainfall attenuated Markov chain". The simulation results show that the prediction algorithm can achieve good prediction accuracy. At the same time, the influence of the number of leading data and the prediction step size on the prediction accuracy is analyzed. Although the ARMA real-time prediction algorithm has a good prediction accuracy, but the prediction error is always there. In order to resist the influence of channel state information estimation error on transmission performance, the AFC with good decoding performance in the range of large signal-to-noise ratio (Signal-to-Noise) is adopted as the forward erasure correction block technique. Combining with the link prediction algorithm, the error rate fluctuation of Ka link in deep space communication is optimized, and the adaptive transmission coding scheme in deep space Ka band is determined. The throughput of the adaptive coding transmission scheme is analyzed theoretically, and the throughput of the scheme is simulated and analyzed, and the throughput of the scheme is compared with that of the fixed rate coding transmission scheme and the adaptive LT transmission scheme. In this paper, the throughput of the proposed scheme is the highest, which verifies that the scheme can maintain the data continuity in the deep space Ka band and the high throughput of the link.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN927
【参考文献】
相关期刊论文 前9条
1 杨峰;薛斌;刘剑;;基于非平稳时序ARIMA模型的W频段雨衰预测[J];电子与信息学报;2015年10期
2 王丽洁;朱华;孙微涛;罗文峰;;中继雨衰对载人航天的影响及解决措施[J];电讯技术;2015年04期
3 于洋;陈晓敏;;IDRP:行星际DTN路由协议[J];国防科技大学学报;2014年03期
4 王辉;魏文博;金胜;叶高峰;景建恩;张乐天;董浩;李波;谢成良;;基于同步大地电磁时间序列依赖关系的噪声处理[J];地球物理学报;2014年02期
5 周洁;焦健;杨志华;顾术实;张钦宇;;深空通信中Ka频段自适应纠删编码研究[J];宇航学报;2013年01期
6 王昱霖;张更新;张威;孔博;;深空探测与行星际互联网[J];国际太空;2012年08期
7 胡圣波;孟新;赵娜;蒋立正;;弱太阳闪烁中的深空探测信道预测[J];宇航学报;2009年01期
8 杨伟;朱巧明;李培峰;钱培德;;基于时间序列的服务器负载预测[J];计算机工程;2006年19期
9 刘嘉兴;;深空测控通信的特点和主要技术问题[J];飞行器测控学报;2005年06期
，本文编号：2168080
本文链接：https://www.wllwen.com/kejilunwen/xinxigongchenglunwen/2168080.html
