我喜欢回忆往事。三十五年前,在我攻读电子学时,据说可以用两种完全等效的方法设计滤波器,即: 确定滤波器的传递函数; 确定滤波器的脉冲响应。但第二种能导致横向滤波器实现的方法很快就被人们所遗忘;而大量的时间用在阐明如何实现对应于给定传递函数的滤波器上。为什么这样呢?原因就在于实现横向滤波器要用具有横向输出的延时线,而在三十五年前实际上没有合适的延时线可用。当时仅有的延时线只是集总常数延时线(体积很大)或铁淦氧加载同轴电缆,或磁致伸缩器件,后两者的通带都不宽。正如大家所知道的,现在的情况则完全不同。理由有二:第一,可以采用声表面波器件, I like to remember the past. Thirty-five years ago, when I was in Electronics, I was said to design filters in two completely equivalent ways, namely, to determine the transfer function of the filter and to determine the impulse response of the filter. But the second method that can lead to the realization of a transversal filter is quickly forgotten by humans; much time is spent on elucidating how to implement a filter that corresponds to a given transfer function. Why? The reason is that the horizontal filter is required to use a delay line with a horizontal output, and virtually no suitable delay line is available thirty-five years ago. The only delay line at the time was just a lumped constant delay line (bulky) or ferrite loaded coaxial cable, or a magnetostrictive device, the latter two being not wide passbands. As we all know, the situation is completely different now. There are two reasons: First, you can use SAW devices,