Abstract: Aimming to improve the accuracy of ultra-wideband positioning under Non-Line-Of-Sight（NLOS） environment, a non-
coherent TOA scheme is proposed by jointing double-threshold and slope estimations. By Direct Path（DP） search initially, the optimal normalized threshold fitting based on the slope of the maximum and minimum energy sample is obtained. Furthermore, double-threshold TOA estimation is adopted and improved by using the characteristics of double-exponent channel model. The Mean-Absolute Error（MAE） of the proposed algorithm is compared with other IR-UWB TOA estimations under NLOS environment. Simulations verify the outstanding performance of this scheme in different UWB channels and SNR environments.

Key words: Ultra Wide Band Impulse Radio（IR-UWB）, double-threshold algorithm, double-exponent channel model, Time of Arrival（TOA） estimation

摘要： 针对非视距脉冲超宽带（IR-UWB）定位应用中，非相干TOA（Time of Arrival）估计算法精度的不足，提出了一种基于双门限和斜率（Double-Threshold and Slope，DTS）的首达信号时间估计算法。首达路径（Direct Path，DP）经过初始估计之后，根据能量采样序列最大值与最小值的斜率，拟合出最优归一化门限表达式；再联合UWB双指数信道模型的特点对双门限检测算法加以改进，实现TOA估计。在非视距环境下仿真、比较了DTS算法与几种IR-UWB常用TOA算法的平均绝对误差（MAE）。仿真结果表明该算法在不同UWB信道和信噪比环境下均获得了优异的误差性能。

关键词: 非视距脉冲超宽带（IR-UWB）, 双门限算法, 双指数信道模型, TOA估计