一种改进的基于样本排序的信道阶数估计算法

Modified Algorithm for Channel Order Estimation via Sample Sorting

信道阶数估计是多数信道盲辨识算法的前端步骤,文章对原有的基于一阶统计量的信道阶数估计算法进行了改进。改进算法仅仅通过对接收样本数据进行排序,建立连续的样本次序对二维空间,利用阶数欠估计时的特定图形结构对信道阶数进行有效的估计,提高了原算法的准确性,并将其扩展应用于SIMO模型中,与其它基于二阶统计量的同类算法相比大大降低了计算复杂度。仿真实验证明,改进算法在较低信噪比且有限样本数据的条件下,性能依然良好,简单实用。

Channel order estimation is the front step of the majority blind channel identification algo- rithms, this paper modifis the traditional channel order estimation algorithm based on first-order statistics. The improved algorithm only sorts the received sample data, constructs the two-dimensional space of successive sample rank pairs and utilizes the particular graph structure under estimation to effectively estimate the channel order. It improves the accuracy of the original algorithm and extends it into SIMO model, greatly reduces the computational complexity compared to other similar algo- rithms based on second-order statisstics. Simulation results show that under the condition of lower signalto-noise ratio or limited sample data, the improved algorithm proves to be good, simple and practical.