自然标记法相关流量测量的线性系统理论

A Linear System Theory for Correlative Flow Measurement with Natural Traces

人工标记法相关流量测量中的线性系统理论早已有之 ,但对自然标记法相关流量测量若使用相同的线性系统理论就会发现有很多矛盾之处。针对这一情况 ,作者建立了分布式注入源的物理模型。借助这一模型 ,传感器信号的自相关函数 (ACF)与传感器所处位置无关的原因、自然标记法相关流量测量中互相关函数 (CCF)的确切含义、以及自然标记法与人工标记法在系统理论意义下的异同等 ,都得到了定量的解释。结论是除了自然标记法的信噪比 (SNR)较人工标记法要低之外 ,两者并无其它本质区别。

For the correlative flow measurement with artificially injected traces, there has been a linear system theory for a long time. But inconsistency would happen if the same theory were used to the correlative flow measurement with natural traces. A new physical model with distributed injecting sources is presented. Based on this model, the reason why the auto-correlation-function (ACF) of the sensor signal is independent of the sensor location, the real meaning of the cross-correlation-function (CCF) by the correlative flow measurement with natural traces, and the difference in the sense of system theories between the system with natural traces and the system with artificially injected traces are quantitatively interpreted. It is concluded that no other distinctions exist between the method with natural traces and the method with artificially injected traces, except that the signal-noise-ratio (SNR) of the former is lower.