实际上一般从数据本身估计的干扰统计量来设置所需的阈值，这个过程称为恒虚警率(CFAR)检测。
Instead, the required threshold is setusing interference statistics estimated from the data itself, a process calledconstant-false-alarm rate (CFAR) detection.
关于检测处理的内容在第六章中详细讲述。
Detection processing is described in detailin Chap. 6.
1.5.6. 测量与跟踪滤波
1.5.6. Measurements and Track Filtering
雷达系统在目标检测之后还会采用各种各样的处理操作。
Radar systems employ a wide variety ofprocessing operations after the point of target detection.
最常见的后检测处理步骤之一，也是本书关注的三个主要功能之一，即，目标跟踪，这是许多雷达系统的基本组成部分。
One of the most common postdetectionprocessing steps, and one of the three major functions of interest in thistext, is tracking of targets, an essential component of many radar systems.
目标跟踪包括对（通常多个）检测目标的位置测量，然后进行跟踪滤波。
Tracking is comprised of (usually multiple)measurements of the position of detected targets followed by track filtering.
雷达信号处理器使用阈值方法检测目标是否存在。
The radar signal processor detects thepresence of targets using threshold detection methods.
被检测目标的距离、角度和多普勒分辨率单元提供了该目标在这些坐标中的粗略估计。
The range, angle, and Doppler resolutioncell in which a target is detected provide a coarse estimate of its location inthose coordinates.
一旦检测到目标，雷达将寻求通过使用信号处理方法来细化估计参数的范围，以便更精确地估计发生阈值跨越的脉冲传输之后的时间延迟、目标相对于天线主波束方向的角度或其径向速度。
Once detected, the radar will seek torefine the estimated range by using signal processing methods to more preciselyestimate the time delay after pulse transmission at which the thresholdcrossing occurred, the angle of the target relative to the antenna mainbeamdirection, or its radial velocity.
由于干扰的存在，单次测量会有一些误差，因此可以在一次测量中提供目标位置和运动的带噪信息。
Individual measurements will have someerror due to interference, and so provide a noisy snapshot of the targetlocation and motion at one instant in time.
术语“跟踪滤波”描述了更高级、更长时间尺度的过程，该过程集成一系列这样的测量数据以估计目标随时间变化的完整轨迹。
The term track filtering describes ahigher-level, longer time scale process of integrating a series of suchmeasurements to estimate a complete trajectory of the target over time.
它通常被描述为数据处理而不是信号处理。
It is often described as data processingrather than signal processing.
由于可能存在具有交叉或紧密间隔的多个目标轨迹，所以跟踪滤波必须确定哪些测量值与哪些被跟踪目标相关，并正确地解决邻近和交叉轨迹的问题。
Because there may be multiple targets withcrossing or closely spaced trajectories, track filtering must deal with theproblems of determining which measurements to associate with which targetsbeing tracked, and with correctly resolving nearby and crossing trajectories.
为了进行跟踪滤波，已经发展了多种最优估计技术。
A variety of optimal estimation techniqueshave been developed to perform track filtering.
Bar-Shalom (1988)撰写的这本书是该领域的优秀参考资料。
An excellent reference in this area isBar-Shalom (1988).
图1.25说明了两个目标位置的一维噪声测量以及使用非常简单的alpha-beta滤波器对噪声轨迹进行滤波，这将在第9章中讨论。
Figure 1.25 illustrates a series of noisymeasurements in one dimension of the position of two targets and the filteringof that noisy trajectory using an extremely simple alpha-beta filter, to bediscussed in Chap. 9.
Figure 1.25. 使用alpha-beta滤波器跟踪两个目标的一维噪声测量。图中不同的标记表示两种不同的目标测量值。灰色线是实际位置，黑色线是滤波后的位置估计。(a)低测量噪声。(b)在较高的测量噪声中跟踪不正确的交叉切换目标。Trackfiltering of noisy measurements for two targets in one dimension using analpha-beta filter. Markers show individual measurements. Gray lines are actualposition, black lines are filtered position estimates. (a) Low measurementnoise. (b) Tracks incorrectly switch targets in higher measurement noise.
每个目标的x维位置与时间的关系用灰色线表示，因此两个目标沿x轴以不同的速度运动，一个目标在大约的时刻32处与另一个目标相交叉。
The position in the x dimension versus timefor each target is shown by the gray lines, so the two targets are moving atdifferent velocities along the x axis and one passes the other at around timestep 32.
圆圈和菱形标记表示每个位置的雷达带噪测量值。
The circle and diamond markers indicate thenoisy radar measurements of position for each.
实心黑线是由alpha-beta滤波器产生的位置平滑估计。
The solid black lines are the smoothedestimates of position produced by the alpha-beta filter.
在图(a)中，当两个目标交叉时，滤波器正确地将测量值与它们分别关联，使得每个平滑的估计值在观察时间内跟随相同的目标变化。
In part (a) of the figure, the filtercorrectly associates the measurements with each target when they cross, so thateach smoothed estimate follows the same target over the observation time.
在图1.25b中，由于噪声方差较高，导致滤波器在时刻40附近错误地交换了两个目标的轨迹。
In Fig. 1.25b, the noise variance ishigher, causing the filter to incorrectly swap the tracks around time step 40.
这表示雷达测量值到跟踪数据关联时产生了错误。
This represents an error inmeasurement-to-track data association.
各种各样的技术都可以用来解决这个问题；一些相关技术将在第9章中讨论。
A variety of techniques are available toattempt to address this problem; a few are discussed in Chap. 9.
——本文译自Mark A. Richards所著的《Fundamentals of Radar Signal Processing（Second edition）》
更多精彩文章请关注微信号：