第一讲随机过程.docVIP

Lecture One Random Process 13. Probabilty distribution of narrow band process envoloping square Tips: We have talked about the probabilty distribution of envoloping and phase of narrow-band process in both cases of pure noise and sinusoidal signal plus noise and understood the distribution form of the signal, which can determine the detection threshold specifically and find signal from the noise background. Detection method can be used for phase or envelope, but the envelope detection was used commonly because it is relatively easy to implement. Initial phase of incoherent signals which we usually used can’t use phase detection because of its random change. Therefore, the envelope detection is divided into linear detection and square-law detection. The former has more complex theoretical analysis, but practices more and it is based on the probability distribution of the envelope of narrowband process. In contrast to the former,the latter uses multi-square-law detection in the theoretical analysis and it is based on the probability distribution of the square of the envelope of narrowband process. In fact, these two detectors have little difference in performance. When the information processing needs to know the probability distribution density after detection, the results discussed above is very practical. 1. (Pure noise) probability distribution of the envelope square of narrowband normal noise narrowband normal noise can be represented as: narrowband normal noise envelope: The probability density of narrowband normal noise envelope is 棗 Rayleigh distribution If envelope detector has half-wave square-law detection characteristics, that is, the output voltage According to the transform of random variable function, the envelope square of the probability density is 棗e xponential distribution Make the normalized random variable is , the probability density is 棗e xponential distribution standar