Test Functions and Distributions
[27/0%]Suppose is a complex continuous function in , with compact support. Prove that uniformly on , for some and for some sequence of polynomials.
Show that the metrizable topology for that was rejected in Section 6.2 is not complete for any .
If is an arbitrary closed subset of , show that there is an such that for every and for every other .
Suppose and whenever and . Prove that is then a positive measure in (which is finite on compact sets).
Prove that the numbers in the Leibniz formula are
Suppose . Does the series
converge for every ?
Let be open in , suppose , and suppose that all have their supports in some fixed compact . Prove that the sequence cannot converge in unless the orders of the are bounded. Hint: Use the Banach-Steinhaus theorem.
Can the assumption about the supports be dropped in (b)?
Let . Define
Prove that is a distribution of infinite order in . Prove that cannot be extended to a distribution in ; that is, there exists no such that in .
Characterize all distributions whose supports are finite sets.
Prove that a set is bounded if and only if
for every .
Suppose is a sequence in such that is a bounded sequence of numbers, for every . Prove that some subsequence of converges, in the topology of .
Suppose is a sequence in such that is bounded, for every . Prove that some subsequence of converges in and that the convergence is uniform on every bounded subset of . Hint: By the Banach-Steinhaus theorem, the restrictions of the to are equicontinuous. Apply Ascoli's theorem.
Suppose is a sequence of locally integrable functions in (an open set in ) and
for every compact . Prove that then in , as , for every multi-index .
Suppose is open in , and is a sequence of harmonic functions in that converges in the distribution sense to some ; explicitly, the assumption is that
Prove then that converges uniformly on every compact subset of and that is a harmonic function. Hint: If is harmonic, is the average of over small circles centered at .
Recall that (the Dirac measure) is the distribution defined by , for . For which is it true that ? Answer the same question for . Conclude that a function may vanish on the support of a distribution although .
If and , does either of the statements
imply the other?
Suppose is the closed unit ball in has its support in , and vanishes on . Prove that . Find other sets for which this is true. (Compare with Exercise 12).
Suppose is compact, and are open in has its support in , and satisfies
for every multi-index . Prove that then
The preceding statement becomes false if is replaced by in the hypothesis (a). Show this by means of the following example, in which . Choose , such that ; define
and consider functions such that if if . Show also that this is a distribution of order 1.
However, for certain can be replaced by in the hypothesis (a) of Exercise 15. Show that this is so when is the closed unit ball of . Find other sets for which this is true.
If has order , show that , for some continuous function . If , what are the possibilities for ?.
Express in the form given by Theorem 6.27, as explicitly as you can.
Suppose , and for every in the support of and for every multi-index . Prove that . Suggestion: Do it first for distributions with compact support, by the method used in Theorem 6.25.
Prove that every continuous linear functional on is of the form , where is a distribution with compact support in ; this is a converse to of Theorem 6.24.
Let be the space of all infinitely differentiable complex functions on the unit circle in . One may regard as the subspace of consisting of those functions that have period . Suppose
converges in the open unit disc in . Prove that each of the following three properties of implies the other two:
There exist and such that
There exist and such that
exists (as a complex number) for every .
For , show that
as . (The derivative of may thus still be regarded as a limit of quotients).
Suppose is a sequence of locally integrable functions in , such that
exists, for each and each . Prove that then converges uniformly on compact sets, for each multi-index .
Let be the Heaviside function on , defined by
and let be the Dirac measure.
Show that , if .
Show that .
Show that . (Here 1 denotes the locally integrable function whose value is 1 at every point and which is thought of as a distribution).
It follows that the associative law fails:
but
Here is another characterization of convolutions analogous to Theorem 6.33. Suppose is a continuous linear mapping of into which commutes with every , that is, (a)
Then there is a such that
Suggestion: Fix , put
let be the directional derivative used in the proof of Theorem 6.30, and show that
which is 0 if (a) holds. Thus , which implies that . Can the assumption that the range of is in be weakened?
If for every , define its principal value integral to be
if the limit exists. For , put
Show that
Find all distributions that satisfy at least one of the following two conditions:
for every ,
for every with .