In mathematics, the Bussgang theorem is a theorem of stochastic analysis. The theorem states that the cross-correlation between a Gaussian signal before and after it has passed through a nonlinear operation are equal to the signals auto-correlation up to a constant. It was first published by Julian J. Bussgang in 1952 while he was at the Massachusetts Institute of Technology.[1]
Statement
Let be a zero-mean stationary Gaussian random process and where is a nonlinear amplitude distortion.
If is the autocorrelation function of , then the cross-correlation function of and is
where is a constant that depends only on .
It can be further shown that
Derivation for One-bit Quantization
It is a property of the two-dimensional normal distribution that the joint density of and depends only on their covariance and is given explicitly by the expression
where and are standard Gaussian random variables with correlation .
Assume that , the correlation between and is,
- .
Since
- ,
the correlation may be simplified as
- .
The integral above is seen to depend only on the distortion characteristic and is independent of .
Remembering that , we observe that for a given distortion characteristic , the ratio is .
Therefore, the correlation can be rewritten in the form
.
The above equation is the mathematical expression of the stated "Bussgang‘s theorem".
If , or called one-bit quantization, then .
Arcsine law
If the two random variables are both distorted, i.e., , the correlation of and is
.
When , the expression becomes,
where .
Noticing that
,
and , ,
we can simplify the expression of as
Also, it is convenient to introduce the polar coordinate . It is thus found that
.
Integration gives
,
This is called "Arcsine law", which was first found by J. H. Van Vleck in 1943 and republished in 1966.[2][3] The "Arcsine law" can also be proved in a simpler way by applying Price's Theorem.[4][5]
The function can be approximated as when is small.
Price's Theorem
Given two jointly normal random variables and with joint probability function
,
we form the mean
of some function of . If as , then
.
Proof. The joint characteristic function of the random variables and is by definition the integral
.
From the two-dimensional inversion formula of Fourier transform, it follows that
.
Therefore, plugging the expression of into , and differentiating with respect to , we obtain
After repeated integration by parts and using the condition at , we obtain the Price's theorem.
Proof of Arcsine law by Price's Theorem
If , then where is the Dirac delta function.
Substituting into Price's Theorem, we obtain,
.
When , . Thus
,
which is Van Vleck's well-known result of "Arcsine law".
Application
This theorem implies that a simplified correlator can be designed. Instead of having to multiply two signals, the cross-correlation problem reduces to the gating of one signal with another.
References
- 1 2 J.J. Bussgang,"Cross-correlation function of amplitude-distorted Gaussian signals", Res. Lab. Elec., Mas. Inst. Technol., Cambridge MA, Tech. Rep. 216, March 1952.
- 1 2 3 Vleck, J. H. Van. "The Spectrum of Clipped Noise". Radio Research Laboratory Report of Harvard University. No. 51.
- 1 2 3 Vleck, J. H. Van; Middleton, D. (January 1966). "The spectrum of clipped noise". Proceedings of the IEEE. 54 (1): 2–19. doi:10.1109/PROC.1966.4567. ISSN 1558-2256.
- 1 2 3 Price, R. (June 1958). "A useful theorem for nonlinear devices having Gaussian inputs". IRE Transactions on Information Theory. 4 (2): 69–72. doi:10.1109/TIT.1958.1057444. ISSN 2168-2712.
- 1 2 Papoulis, Athanasios (2002). Probability, Random Variables, and Stochastic Processes. McGraw-Hill. p. 396. ISBN 0-07-366011-6.
Further reading
- E.W. Bai; V. Cerone; D. Regruto (2007) "Separable inputs for the identification of block-oriented nonlinear systems", Proceedings of the 2007 American Control Conference (New York City, July 11–13, 2007) 1548–1553