基于压缩信号特征值分解的超宽带码序列估计算法

针对脉冲超宽带通信中码序列盲估计问题,根据超宽带信号为时域极窄脉冲这一主要特点,提出了一种适于较低采样率下基于特征值分解的码序列盲估计算法。结合压缩传感理论,首先利用压缩测量矩阵将信号维度降低,然后按字符周期将接收到的UWB信号截取成若干等长的时间窗,分段积累得到压缩后信号协方差矩阵的估计值,进而通过对于协方差矩阵的特征值分解得到码序列的估计,同时进一步根据所得特征值得到接收信号符号帧级的时间偏移量估计。实验验证表明,该方法可以以较低采样率,在低信噪比条件下较好地完成接收信号的码序列估计。

Estimation Algorithm for Code Sequence In IR-UWB System based on Eigenvalues Decomposition of Compressed Signal

Aiming at the problem of blind estimation for code sequence in IR-UWB system, considering the UWB signal is extreme narrow pulse, blind estimation algorithm for code sequence based on eigenvalues decomposition under low sampling rate is proposed. According to the compressed sensing theory, first, the signal dimension is reduced by using the sensing matrix. Next, the received UWB signal is divided into several time windows, the duration of which is the period of symbol. Then, a covariance matrix is computed by accumulation with windowed signal. We show that the code sequence can be reconstructed from the eigenvalues decomposition and the desynchronization time with the symbol frame level can be estimated. Experimental results show that the method can provide a good estimation below the Nyquist sampling rate, even when the received signal is far below the noise level.