基于LSTM的北斗卫星信号识别方法研究

Research on Signal Recognition and Anti interference of Beidou Satellite Based on LSTM

摘要:

为有效降低和抵御干扰对信号接收与定位的影响,提高导航系统的稳定性和可靠性,设计基于LSTM的北斗卫星信号识别与抗干扰方法,以无失真地利用BDS信号。对北斗卫星真实信源实施信号调制,获取卫星信号调制样式识别算法的输入信号。设计基于LSTM的卫星信号调制样式识别算法,模型为一个传统LSTM网络与CNN网络级联的新型LSTM网络,通过CNN网络能够实施I、Q数据间特征的提取,利用LSTM网络提取数据中时序特征,并对其特征进行融合,实现北斗卫星信号的识别。并设计结合空域抗干扰技术与时域抗干扰技术的空时自适应滤波约束算法,实现北斗卫星信号窄带与宽带干扰的共同抑制,以有效提高识别的效果。实验测试结果表明,设计方法的识别准确率整体较高,最高达到接近90%,抗干扰后输出信干比最高达到78.52dB。

Abstract:

In order to effectively reduce and resist the influence of interference on signal reception and positioning, and improve the stability and reliability of navigation system, a signal recognition and anti-jamming method of Beidou satellite based on LSTM is designed to utilize BDS signals without distortion. The real information source of Beidou satellite is modulated to obtain the input signal of the satellite signal modulation pattern recognition algorithm. The satellite signal modulation pattern recognition algorithm based on LSTM is designed, and the model is a new LSTM network cascelled by the traditional LSTM network and the CNN network. The features between I and Q data can be extracted through the CNN network, and the sequential features in the data can be extracted and fused through the traditional LSTM network. Recognition of Beidou satellite signals. A spatial-time adaptive filter constraint algorithm combining spatial-time anti-jamming technology and time-domain anti-jamming technology is designed to realize the joint suppression of narrowband and wideband interference of Beidou satellite signals, so as to effectively improve the recognition effect. The experimental results show that the recognition accuracy of the design method is high, the highest is close to 90%, and the highest output signal-to-stem ratio is 78.52dB after anti-interference.