以超声测风为应用背景,对基于相关法的时延估计性能进行分析,指出其存在的问题。利用发射信号为已知这一先验信息,提出了二次相关估计时延的解决方法。仿真结果表明,在相同的信噪比下,与相关法相比,二次相关法能够利用先验信息,提高了时延估计的正确率,并能改善系统6dB以上的噪声抑制性能。最后,对二次相关法的运算量进行了分析,并分别以TI TMS320C5000系列DSP处理器和Ahera CycloneⅡ系列FPGA作为硬件平台,估计了该方法在这两种平台上的处理时间,得出了该方法可满足实时处理的结论。
Take the ultrasonic anemograph as the application background, the performance of the estimation of the time delay based on the correlation method is analyzed, and the disadvantage is pointed out. The twice correlation method, which uses the apriori information of the sending signals, is proposed as a solution. The simulation results indicate that under the same SNR, compared with the correlation method, the accuracy can be improved when using the twice correlation method. The anti-noise performance of the system has over 6 dB improvement by using the apriori information in the twice correlation method. At the end, the computation which needed in the method has been analyzed. TI TMS320C5000 series DSP processor and Altera Cyclone Ⅱ series FPGA device are taken as the hardware platform respectively. The processing time, which needed in the new method on those two kinds of platforms, are estimated. It's concluded that the method could satisfy the real-time processing.