A curiosum concerning discrete time convolution
نویسندگان
چکیده
It is shown that the discrete time convolution of two absolutely summable nowhere zero sequences may be identically equal to zero. DEVELOPMENT Consider two absolutely summable sequences a = {a,,: n E Z } and b = { b,,: n E Z } of real numbers. Further, assume that the sequences a and b are nowhere zero. Does it follow that the discrete time convolution a * b is nowhere zero? Does it follow that a * b is nonzero on some nonempty subset of Z? From a linear systems viewpoint, does an absolutely summable nowhere zero input to a discrete time linear time-invariant system described via discrete time convolution with a fixed absolutely summable nowhere zero sequence result in an output which is nonzero somewhere? 'The following development, inspired by [I . pp. 354-3561. addresses these questions. To begin, we will use the following notation. For an absolutely summable sequence of real numbers a = { a,,: n E Z 1, let T, map absolutely summable sequences of real numbers into absolutely sumrnable sequences of real numbers via where a = {a, , : n E Z } is any absolutely summable sequence of real numbers. For any two absolutely summable sequences of real Manuscript received May 25. 1989; revised July 20. 1989. This work was supported in part by the Office of Naval Research under Grant NOOO1490-J17 12. and in pan by the Air Force Office of kientific Research under Grant AFOSR-86-0026. E. B. Hall is with the Department of Electrical Engineering. Southern Methodist University. Dallas. TX 75275. G. L. Wise is with the Department of Statistics. University of California. Berkeley. CA 94720. IEEE Log Number 9034983. 0096-35 18/90/06001058$01 .OO O 1990 IEEE lEEE TRANSACTIONS ON ACOUSTICS. SPEECH. AND SIGNAL PROCESSING. VOL. 38. NO. 6. JUNE I W ) 1 059 numbers a = ( a , , : n E Z } and f l = { fl,,: t l E Z ). it follows that I m where we define X, = E , apflq. Finally, note that for any two absolutely summable sequences of real numbers a and b . i t follows via Fubini's theorem that T , ( a ) * T p ( b ) = (T,, 0 T @ ) ( N * 6). Theorem I : Letting the above paragraph set notation. there exist two nonidentically zero absolutely summable sequences of real numbers a and fl such that for any absolutely summable sequences of real numbers a and b , T , ( u ) * T @ ( b ) = 0 . Proof: Recall that the function I cos(x) I is expressible as a Fourier series given by C,",-, c , exp (inx) where i denotes the imaginary unit and where it follotvs easily that c,, = 0 if n is odd and c,, = ( 2 / + ) [ ( 1 ) " I 2 / ( 1 n 2 ) 1 if n is even. Further, if we define fi (x) = ( I COS (x ) I + cos (x)) and
منابع مشابه
A Discrete Singular Convolution Method for the Seepage Analysis in Porous Media with Irregular Geometry
A novel discrete singular convolution (DSC) formulation is presented for the seepage analysis in irregular geometric porous media. The DSC is a new promising numerical approach which has been recently applied to solve several engineering problems. For a medium with regular geometry, realizing of the DSC for the seepage analysis is straight forward. But DSC implementation for a medium with ir...
متن کاملFree Vibration of Annular Plates by Discrete Singular Convolution and Differential Quadrature Methods
Plates and shells are significant structural components in many engineering and industrial applications. In this study, the free vibration analysis of annular plates is investigated. To this aim, two different numerical methods including the differential quadrature and the discrete singular convolution methods are performedfor numerical simulations. Moreover, the Frequency values are obtained v...
متن کاملدرکنش سد و مخزن در دامنه زمان به وسیله روش انتگرال حلقوی منفرد مجزا (DSC)
In this paper, time-domain dynamic analysis of dam - infinite reservoir is studied. A numerical method, discrete singular convolution (DSC), which seems to be efficient and simple, has been used to model the mixed boundary conditions. After a brief introduction, DSC is applied to model the equation of motion of fluid-structure with constant cross section subjected to El-Centro (1940) earthquake...
متن کاملWeighted Convolution Measure Algebras Characterized by Convolution Algebras
The weighted semigroup algebra Mb (S, w) is studied via its identification with Mb (S) together with a weighted algebra product *w so that (Mb (S, w), *) is isometrically isomorphic to (Mb (S), *w). This identification enables us to study the relation between regularity and amenability of Mb (S, w) and Mb (S), and improve some old results from discrete to general case.
متن کاملStrong log-concavity is preserved by convolution
We review and formulate results concerning strong-log-concavity in both discrete and continuous settings. Although four different proofs of preservation of strong log-concavity are known in the discrete setting (where strong log-concavity is known as “ultra-log-concavity”), preservation of strong log-concavity under convolution has apparently not been investigated previously in the continuous c...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- IEEE Trans. Acoustics, Speech, and Signal Processing
دوره 38 شماره
صفحات -
تاریخ انتشار 1990