International Advanced of Technology Congress (ATCi),PWTC, Malaysia.

November 3-5, 2009

Asymptotic properties of solutions to hyperbolic equations


Michael Ruzhansky

Department Mathematics Department, Imperial College London, 180 Queens Gate,
London SW7 2AZ, United Kingdom.
Email: [email protected]
_______________________________________________________________________________________________

Abstract
In this note we describe time-decay rates of solutions to hyperbolic equations with constant
coefficients with arbitrary lower order terms.
__________________________________________________________________________

I. Introduction

In this note we give an overview of microlocal time-decay rates of oscillatory integrals appearing in the solution
to the Cauchy problem for hyperbolic partial differential equations with constant coefficients and arbitrary lower order
terms. We will be most interested in microlocal decay rates corresponding to different regions in the phase space. A
slightly more detailed overview can be found in [8]. These decay rates can be then put together in the standard way as
explained, for example, in [11].
We consider equations of the form

e e
+ +

. 1, , 0, = ), ( ) ( = ) (0,
0, > 0, = ) ( ) , (
0
,
= | |
1
0 = 1 =
n n
l
l
t
r
t x r
l r
m
l
j m
t x j
m
j
m
t x t
x m l C x f x u D
t u D D c u D D P u D D D L
P P
o
o
o
(1)
Results on the decay rates of its solutions can be expressed in rems of its characteristic roots ) ( , ), (
1
t t
m
,
which are solutions to the characteristic polynomial equation 0 = ) , ( t L with respect to t . Symbol ) (
j
P of ) (
x j
D P
(where as usual
x x
i D c = ) is a homogeneous polynomial of order j , and the
r
c
, o
are complex constants. We assume
that equation (1) is strictly hyperbolic in order not to worry about its well-posedness.
Throughout, we assume the stability condition
, , 1, = for 0 ) ( m k
k
> t (2)
for all
n
P e . Solution to the Cauchy problem (1) can be written in the form
), ( ) ( = ) , (
1
0 =
x f t E x t u
j j
m
j

where propagators ) (t E
j
are defined by
, ) ( )) , (
) (
( = ) ( ) (
1 =

t

d f t A
t
e e x f t E
k
j
k
i
m
k
ix
n j

}

P
(3)
with suitable amplitudes ) , ( t A
k
j
. In the areas where roots are simple, phases and amplitudes are smooth, and
we can analyse the sum (3) termwise, reducing the analysis to single integrals. In the case of multiple characteristics we
can group terms in (3) in a special way to obtain suitable decay estimates. We note that properties of characteristics may
be different in different regions. To isolate different types of behavior in different areas we can microlocalise with a cut-off
function ) ( _ , to study integrals of the form
2


International Advanced of Technology Congress (ATCi),PWTC, Malaysia. November 3-5, 2009

. ) ( ) ( )) , (
) (
( = ) ( ) ( ) (
1 =
_

_
t

d f t A
t
e e x f t E D
k
j
k
i
m
k
ix
n j

}

P
(4)

We will denote various constants throughout the paper by the same letter C . Balls with radius R centred at
n
P e will be denoted by ) (
R
B . We will use the notation A + 1 = , | | 1 =
2
D and . | |=| |
1/2
A D The
Sobolev space
l
p
W is then defined as the space of measurable functions for which ) (
n
x
p
l
L f D P e .
We will also use the standard notation for the symbol class

1,0
= S S , as a space of smooth functions
) ( ) , ( =
n n
C x a a P P e

satisfying symbolic estimates
| |
|) | (1 | ) , ( |
o
o|
o

|


+ s c c C x a
x
, for all
n
x P e , , and
all multi-indices | o, .
If function ) ( = a a is independent of x , we will sometimes also write ) (
1,0
U S a

e for an open set
n
U P c ,
if ) ( ) ( = U C a a

e satisfies
| |
|) | (1 | ) ( |
o
o
o

+ s c C a , for all U e , and all multi-indices o .
Estimates for wave type equations have been thoroughly analysed in [1,2,4,6,12,15]. The case of dissipative wave
equations was treated in [5] and for the analysis of the Klein--Gordon equation we refer to [3] and references therein.
Equations with homogeneous symbols were analysed in [13,14].
The present note is based on papers [10,11,7,9]. For the detailed proofs we refer to [8]. The author was
supported in part by EPSRC grants EP/E062873/01 and EP/G007233/1.

Away from the real axis
We begin by looking at the zone where roots are separated from the real axis. If the roots are smooth, we can
analyse solution (3) termwise:

Theorem 1 2.1. Let X U : t be a smooth function,
n
U P c open. Let ) (
1,0
U S a

e , i.e. assume that
) ( ) ( = U C a a

e satisfies , |) | (1 | ) ( |
| |o
o
o

+ s c C a for all U e and all multi-indices o . Let ) (
0
1,0
n
S P e _
be such that 0 = _ outside U . Assume further that: [leftmargin=*,label=()]
there exists 0 > o such that o t > ) ( for all U e ;
|) | (1 | ) ( | t + s C for all U e .
Then for all 0 > t we have
,
| | ) (
) ) ( ) ( ) ( (
) ) ( (
o
_
o t o
+ +
s
+
}
r
f Ce d f a e D D
p
N
p
W
t
n
x
q
L
t x i
n x
r
t
H H H H
P P

(5)
where 1 =
1 1
q p
+ , 2 < 1 s p , )
1 1
(
q p
n N
p
> , 0 > r , o a multi-index and ) (
0
n
C f P

e . If 1 = p , we
take n N >
1
.
Moreover, let us assume that equation 0 = ) , ( t L has only simple roots ) ( t
k
which satisfy condition (i)
above, in the open set
n
U P c , for all . , 1, = m k Then solution u to (Error! Reference source not found.) satisfies
,
| |
|| ||
) (
|| ) , ( ) ( ||
1
0 =
l r
f Ce t u D D D
p
N
p
W
l
m
l
t
n
x
q
L
x
r
t
+ +
s

o
_
o o
P
(6)
3


International Advanced of Technology Congress (ATCi),PWTC, Malaysia. November 3-5, 2009

where 2 1 s s p , 1 =
1 1
q p
+ , and o , , r N
p
are as above.

We note that we may have different norms on the right hand side of (6) . For example, we also have the
following estimate:
,
| |
|| ||
) (
|| ) , ( ) ( ||
2
1
0 =
l r
f Ce t u D D D
'
q
N
W
l
m
l
t
n
x
q
L
x
r
t
+ +
s

o
_
o o
P
(7)
where 2 < 1 s p , 1 =
1 1
q p
+ , )
1 1
(
2 q p
n
N
'
q
> , and
2
>
n
N
'

for 1 = p .
To be able to derive time decay in the case of multiple roots, we will group terms in (3) in the following way.
Assume that roots ) ( , ), (
1
t t
L
coincide on a set contained in some M , that is { }. ) ( = = ) (
1
t t
L
M For
0 > c , we define { }. < ) , ( : := c
c
M M dist
n
P e Choose 0 > c so that these roots ) ( , ), (
1
t t
L
do not
intersect with any of the other roots ) ( , ), (
1
t t
m L

+
in
c
M . If different numbers of roots intersect in different sets,
we can apply the following theorem to such sets one by one. We note that by the strict hyperbolicity
c
M is bounded.
Here we will estimate the sum
. ) ( ) ( )) , (
) (
(
1 =
_
t

c
d f t A
t
e e
k
j
k
i
L
k
ix

}

M
(8)


Theorem 2 2.2. Let the sum (3) be the solution to the Cauchy problem (1) . Assume that roots ) ( , ), (
1
t t
L

coincide in a set contained in M and do not intersect other roots in the set
c
M . Let ) (
0
c
_ M

eC . Assume that there

exists 0 > o such that o t > ) (
k
for all
c
M e and . , 1, = L k
Then for all 0 > t we have
, ) (1
) (
) ) ( ) ( )) , (
) (
( (
1
1 =
p
L
t L
n
x
q
L
k
j
k
i
L
k
ix
x
r
t
f e t C dx f t A
t
e e D D H H H H
o
t

c
o
_

+ s

}
P

M

where 1 =
1 1
q p
+ , 2 1 s s p .
Thus, if characteristic roots are separated from the real axis on the support of some ) (
0
n
C P

e _ , we can
separate the solution (3) into groups of multiple roots for which the
q p
L L norms still decay exponentially as stated in
Theorem 2.2. We also note that since
c
M is bounded, assumption (ii) of Theorem 1 is automatically satisfied and,
therefore, it is omitted in the formulation of Theorem 2.1.

Roots with non-degeneracies
The following case that we consider is the one of roots satisfying certain non-degeneracy conditions. These may
be conditions on the Hessian, convexity conditions, or simply the information on the index of the corresponding level
surfaces. In this section we will give the corresponding statements. We always assume the stability condition
(Error! Reference source not found.) but no longer assume that roots are separated from the real axis.
First we state the result for phases with the non-degenerate Hessian. The behavior depends on critical points
0

with 0 = ) (
0
t V and the behavior of the Hessian at such points. As usual, we say that the critical point
0
is non-
4


International Advanced of Technology Congress (ATCi),PWTC, Malaysia. November 3-5, 2009

degenerate if the Hessian ) (
0
t Hess is non-degenerate.
Theorem 3 Let
n
U P c be a bounded open set, and let X U : t be smooth and such that 0 ) ( > t for
all U e . Assume that there are some constants
0
C and M such that
M
C Hess

+ > |) | (1 | ) ( det |
0
t for all
U e . Let ) (
0
1,0
n
S P e _ be such that 0 = _ outside U and let ) (
1,0
U S a

e .
Assume that t has only one non-degenerate critical point in U , and that U is sufficiently small. Then there is a
constant 0 > C independent of the position of U such that for all 0 > t we have
,
) (
|| || ) (1
) (
) ( ) ( ) (
)
1 1
(
2 ) ) ( (
n
x
p
L
q p
n
n
x
q
L
t x i
n
f t C d f a e
P
P
P

+
+ s
}
_
t

(9)
with 2 1 s s p , 1 =
1 1
q p
+ , . )
1 1
(
2
=
q p
M
N
p

For example, the case of the Klein--Gordon equation corresponds to 2 = + n M in this theorem. If we want to
have estimate (Error! Reference source not found.) uniformly over all U of fixed volume but independent of its position,
we must replace the
p
L
f || || norm on the right hand side of (Error! Reference source not found.) by
p
N
p
W
f || || . For
details of this we refer to [11]. The condition that critical points are isolated and therefore can be localised by different
sets U may follow from certain properties of t .
If we apply different versions of the stationary phase method under different conditions, we can reach different
conclusions here. For example, we also have:
Theorem 4 Let
n
U P c be a bounded open and let X U : t be smooth and such that 0 ) ( > t for all
U e . Let ) (
0
1,0
n
S P e _ be such that 0 = _ outside U and let ) (
1,0
U S a

e . Assume that t has only one critical
point
0
in U , and that U is sufficiently small.
Suppose that there are constants 0 > ,
0
M C independent of the size and position of U and of
0
, with the
following conditions. Suppose that k rankHess = ) (
0
t , that this rank is attained on an k k submatrix ) (
0
A and
that
M
C A

+ > |) | (1 | ) ( det |
0
0
0
. Then for all 0 > t we have
, || || ) (1
) (
) ( ) ( ) (
)
1 1
(
2 ) ) ( (
p
N
p
W
q p
k
n
x
q
L
t x i
n
f t C d f a e

+
+ s
}
P
P
_
t

(10)
with 2 1 s s p , 1 =
1 1
q p
+ , . )
1 1
(
2
=
q p
M
N
p

The proof of this theorem is similar to the proof of Theorem 3 once we restrict to the set of k variables (possibly
after a suitable change) on which the rank of the Hessian is attained on ) (
0
A .
The next theorem is an estimate of oscillatory integrals with real-valued phases under convexity condition. The
convexity condition is weaker than (but does not contain) the condition that the Hessian of t is positive definite and the
result can be compared with Theorem 3, dependent on suitable properties of roots.
Let us first give the necessary definitions. Given a smooth function P P
n
: t and P e , set
{ }. = ) ( : := ) ( t t

n
P e E E
A smooth function P P
n
: t is said to satisfy the convexity condition if surface

E is convex for each

5


International Advanced of Technology Congress (ATCi),PWTC, Malaysia. November 3-5, 2009

P e . Note that the empty set and the point set are considered to be convex. If the Gaussian curvatures of

E never
vanish,

E is automatically convex (the converse is not true). This curvature condition corresponds to the case 1 = n k
in Theorem 4. Another important notion is that of the maximal order of contact of a hypersurface:
Definition 5 Let E be a hypersurface in
n
P . let E e o , and denote the tangent plane at o by
o
T . Now let
P be a 2--dimensional plane containing the normal to E at o and denote the order of the contact between the line
P T
o
and the curve P E by ) , ; ( P o E . Then set
. ) , ; ( sup sup := ) ( P
P
o
o
E E
E e


We note that 2 = ) (
n
E since 2 = ) , ; ( P
n
o E for all
n
E e o and all planes P containing o and the origin.
If ) (
l
is a characteristic root of an
th
m order homogeneous strictly hyperbolic constant coefficient operator, then
m
l
s E ) (

([14]). Now we can formulate the corresponding theorem.
Theorem 6 Suppose P P
n
: t satisfies the convexity condition and let ) (
n
C P

e _ ; furthermore, on
_ supp , we assume: [label=(),leftmargin=*]
1 for all multi-indices o there exists a constant 0 >
o
C such that
; |) | (1 | ) ( |
| | 1 o
o
o

t

+ s c C

2 there exist constants 0 > , C M such that for all M > | | we have | | | ) ( | t C > ;
3 there exists a constant 0 > C such that C > c | ) ( | e t
e
for all
1
e
n
E e , 0 > ; in particular,
C > V | ) ( | t for all { } 0 \
n
P e ;
4 there exists a constant 0 >
1
R such that, for all 0 > ,
{ } . (0) = ) ( :
1
) (
1
1
R
n
B c e E t

P

Also, set )) ( ( sup :=
0 >
t

E and assume this is finite. Let
j
j j
S a a

e
1,0
) ( = be a symbol of order j of
type (1,0) on
n
P . Then for all 0 > t we have the estimate
, ) (1
) (
) ( ) ( ) (
, ,
)
1 1
(
1
) ) ( (
t j p
N
p
W
q p
n
n
x
q
L
j
t x i
n
f t C d f a e H H H H

+
+ s
}
t
_
P P

(11)
where 1 =
1 1
q p
+ , 2 < 1 s p , and the Sobolev order satisfies j
q p
n N
t j p
> )
1 1
(
, ,
for 1 < 0 t s , and
j
q p
n
n N
t j p

|
|
.
|

\
|
> )
1 1
(
1
, ,

for 1 > t .

In the case without convexity, we also introduce an analog of the order of contact. Thus, if E is a hypersurface in
n
P , not necessarily convex, we define
), ( ) , ; (
inf
sup := ) (
0
E s E E
E e
o
o
P
P

where ) , ; ( P o E is as in Definition 5. If ) ( p is a polynomial of order m, { } 0 = ) ( : = p
n
P e E is
6


International Advanced of Technology Congress (ATCi),PWTC, Malaysia. November 3-5, 2009

compact and 0 = ) ( / V p on E , then m s E s E ) ( ) (
0
([14]).

Theorem 7 Suppose P P
n
: t is a smooth function. Let ) (
n
C P

e _ ; furthermore, on _ supp , we
assume: [label=()]
5 for all multi-indices o there exist constants 0 >
o
C such that
; |) | (1 | ) ( |
| | 1 o
o
o

t

+ s c C

6 there exist constants 0 > , C M such that for all M > | | we have | | | ) ( | t C > ;
7 there exists a constant 0 > C such that C > c | ) ( | e t
e
for all
1
e
n
E e and 0 > ;
8 there exists a constant 0 >
1
R such that, for all 0 > ,
{ } . (0) = ) ( :
1
1
R
n
B c e t

P

Set )) ( ( sup :=
0 0 > 0
t

E and assume it is finite. Let
j
j j
S a a

e
1,0
) ( = be a symbol of order j of type
(1,0) on
n
P . Then for all 0 > t we have the estimate
,
)
1 1
(
) (1
) (
) ( ) ( ) (
, ,
0
1
) ) ( (
t j p
N
p
W
n
x
q
L
j
t x i
n
f
q p
t C d f a e H H H H

+ s

+
}

t
_
P P

(12)
where 1 =
1 1
q p
+ , 2 < 1 s p , and the Sobolev order satisfies j
q p
n N
t j p
> )
1 1
(
, ,
for 1 < 0 t s , and
j
q p
n N
t j p

|
|
.
|

\
|
> )
1 1
(
1
0
, ,

for 1 > t .


Roots meeting the real axis
In this section we will present the results for characteristic roots (or phase functions) in the upper complex plane
near the real axis, that become real at some point or in some set.
For
n
P c M , denote { } c
c
< ) , ( : = M M dist
n
P e as before. The largest number N e v such that
v c
c C meas s ) (M for all sufficiently small 0 > c , will be denoted by M codim , and we will call it the codimension of
M .
We will say that the root
k
t meets the real axis at
0
with order
k
s if 0 = ) (
0
t
k
and if there exists a
constant 0 >
0
c such that
, ) ( | |
0
0
t
k
k
s
c s
for all sufficiently near
0
. More generally, if the root
k
t meets the axis on the set
{ } 0 = ) ( : = t
k
n
k
Z eP , we will say that it meets the axis with order s if
. ) ( ) , (
0
t
k
s
k
Z dist c s
We will localise around each connected component of
k
Z , e.g. around each point of
k
Z , if it is a union of
7


International Advanced of Technology Congress (ATCi),PWTC, Malaysia. November 3-5, 2009

isolated points. As usual, when we talk about multiple roots intersecting in a set M , we adopt the terminology
introduced earlier. Since we are dealing with strictly hyperbolic equations, roots can meet each other only for bounded
frequencies, so we may assume that set M is bounded.

Theorem 8 Assume that the characteristic roots ) ( , ), (
1
t t
L
intersect in the
1
C set M of codimension
. Assume also that they meet the real axis in M with the finite orders s s , i.e. that
, ) ( ) , (
0
t
k
s
dist c s M
for some 0 >
0
c and all L k , 1, = . Assume that (Error! Reference source not found.) is the solution of the
Cauchy problem (Error! Reference source not found.) and we look at its part (Error! Reference source not found.) . Let
) (
0
c
_ M

eC for sufficiently small 0 > c . Then for all 0 > t we have

, ) (1
) (
) ) ( ) ( )) , (
) (
( (
1 )
1 1
(
1 =
p
L
L
q p s
n
x
q
L
k
j
k
i
L
k
ix
x
r
t
f t C
d f t A
t
e e D D
H H
H H
+

+ s

P
_
t

c
o
M
(13)
where 1 =
1 1
q p
+ , 2 1 s s p .
We assume 0 > c to be small enough to make sure that the type of behaviour assumed in the theorem is the
only one that takes place in
c
M . In the complement of
c
M we may use other theorems to analyse the decay rate.
Moreover, we assume that set M is
1
C . In fact, it is usually Lipschitz, so in order to avoid to go into depth about its
structure and existence of almost everywhere differentiable coordinate systems, we make the technical
1
C assumption.
Let us now give a special case of this theorem where simple roots meet the axis at a point, so that we have 1 = L
and n = . The following statement is also global in frequency, so we have the result in Sobolev spaces.

Theorem 9 Consider the
th
m order strictly hyperbolic Cauchy problem (Error! Reference source not found.) for
operator ) , (
x t
D D L , with initial data
j r
W f
p
N
p j
+ +
e
| |o
, for 1 , 0, = m j , where 2 1 s s p and s s q 2
are such that 1 =
1 1
q p
+ , 0 > r and o is a multi-index. We assume that the Sobolev index
p
N satisfies
)
1 1
(
q p
n N
p
> for 2 < 1 s p and n N >
1
for 1 = p .
Assume that the characteristic roots ) ( , ), (
1
t t
m
of 0 = ) , ( t L satisfy 0 >
k
t for all k , and also the
following conditions: [leftmargin=*,label=(H)]
for all m k , 1, = , we have
; 0 > ) (
liminf
| |
t

for each
n
P e
0
there is at most one index k for which 0 = ) (
0
t
k
and there exists a constant 0 > c
such that
), ( | |
0
t
k
s
c s
for in some neighbourhood of
0
. Assume also that there are finitely many points
0
with 0 = ) (
0
t
k
.
8


International Advanced of Technology Congress (ATCi),PWTC, Malaysia. November 3-5, 2009

Then the solution ) , ( = x t u u to Cauchy problem (Error! Reference source not found.) satisfies the following
estimate for all 0 > t :
.
| |
) (1 ) , (
1
0 =
)
1 1
(
,
j r
f t C t u D D
p
N
p
W
j
m
j
q p s
n
r q
L
x
r
t
+ +
+ s

o
o
o
H H H H (14)

As a special case, such estimate together with (Error! Reference source not found.) below (used with
2 = =
1
s s ), we improve the indices in Sobolev spaces over
2
L for the dissipative wave equation compared to [5].
If conditions of Theorem 9 hold only with 0 =
0
, namely if 0 = ) (
0
t
k
implies 0 =
0
, we will call the
polynomial ) , ( t L strongly stable. Now we will give some improvements of (Error! Reference source not found.) under
additional assumptions on the roots:

Remark 10 The order of time decay in Theorem 9 may be improved in the following cases, if we make additional
assumptions. If, in addition, we assume that 0 = ) (
0
t
k
in (H2) implies that 0 =
0
, then we actually get the estimate
,
| |
) (1
) (
) , (
1
0 =
2
| |
)
1 1
(
j r
f t C t u D D
p
N
p
W
j
m
j
q p s
n
n
x
q
L
x
r
t
+ +
+ s

o
o
o
H H H H
P

where here and further in this remark
p
N is as in Theorem 9.
Now, assume further that for all
0
in (H2) we also have the estimate
, | | | ) ( |
1
0
1
s
k
c t s (15)
with some constant 0 >
1
c , for all sufficiently close to
0
.
If we have that 0 = ) (
0
t
k
in (H2) implies that we have (Error! Reference source not found.) around such
0
, then we actually get
.
| |
) (1
) (
) , (
1
0 =
1
)
1 1
)( (
j r
f
s
t C t u D D
p
N
p
W
j
m
j
rs
q p s
n
n
x
q
L
x
r
t
+ +
+ s

o
o
H H H H
P


And finally, assume that for all
0
such that 0 = ) (
0
t
k
in (H2), we also have 0 =
0
and
(Error! Reference source not found.) around such
0
. Then we actually get
.
| |
) (1
) (
) , (
1
0 =
1
| |
)
1 1
(
j r
f
s
t C t u D D
p
N
p
W
j
m
j
rs
s q p s
n
n
x
q
L
x
r
t
+ +
+ s

o
o
o
H H H H
P
(16)

Estimate (Error! Reference source not found.) with 2 = =
1
s s gives the decay estimate for the dissipative
wave equation.
Moreover, there are other possibilities of multiple roots intersecting each other while lying entirely on the real
axis. For example, this is the case for the wave equation or for more general equations with homogeneous symbols, when
several roots meet at the origin. In this case roots always lie on the real axis, but they become irregular at the point of
multiplicity, which is the origin for homogeneous roots. In such cases we have to look at the structure of such multiple
points by making cut-offs around them and studying their structure in more detail. In particular, there is an interaction
between low frequencies and large times, which does not take place for homogeneous symbols. The detailed discussion of
9


International Advanced of Technology Congress (ATCi),PWTC, Malaysia. November 3-5, 2009

this topic and corresponding decay rates can be found in [11].

References

[1] Brenner, P., On
'
p
p
L L estimates for the wave-equation, Math. Z. 145 (1975), 251--254.

[2] Brenner, P.,
p p
L L
'
-estimates for Fourier integral operators related to hyperbolic equations, Math. Z. 152
(1977), 273--286.

[3] Hrmander L., Lectures on nonlinear hyperbolic differential equations, Mathmatiques & Applications (Berlin),
vol. 26, Springer-Verlag, Berlin, 1997.

[4] Littman, W.,
q p
L L -estimates for singular integral operators arising from hyperbolic equations, Partial
differential equations (Proc. Sympos. Pure Math., Vol. XXIII, Univ. California, Berkeley, Calif., 1971), Amer. Math. Soc.,
Providence, R.I., 1973, pp. 479--481.

[5] Matsumura, A., On the asymptotic behavior of solutions of semi-linear wave equations, Publ. Res. Inst. Math.
Sci., Kyoto Univ. 12 (1976/77), 169--189.

[6] Pecher, H.,
p
L -Abschtzungen und klassische Lsungen fr nichtlineare Wellengleichungen. I, Math. Z. 150
(1976), 159--183.

[7] Ruzhansky, M., On some properties of Galerkin approximations of solutions to Fokker--Planck equations, in
Proceedings of the 4th International Conference ``Analytical Methods in Analysis and Differential Equations (AMADE-
2006), Vol.3, Differential Equations, Minsk: Institute of Mathematics of NAS of Belarus, 133--139, 2006.

[8] Ruzhansky, M. Microlocal time decays for hyperbolic equations with lower order terms, Novi Sad J. Math, 38
(2008), 15--26.

[9] Ruzhansky, M., Pointwise van der Corput lemma for functions of several variables, Funct. Anal. Appl., 43
(2009), 75--77.

[10] Ruzhansky, M. and Smith, J., Global time estimates for higher order hyperbolic equations, Journees
``Equations aux Derivees Partielles", Exp. No. XII, 29 pp., Ecole Polytech., Palaiseau, 2005.

[11] Ruzhansky, M. and Smith, J., Dispersive and Strichartz estimates for hyperbolic equations with constant
coefficients, arXiv:0711.2138v1, to appear in MSJ Memoirs.

[12] Strichartz, R. S., A priori estimates for the wave equation and some applications, J. Funct. Analysis 5 (1970),
218--235.

[13] Sugimoto, M., A priori estimates for higher order hyperbolic equations, Math. Z. 215 (1994), 519--531.

[14] Sugimoto, M., Estimates for hyperbolic equations with non-convex characteristics, Math. Z. 222 (1996), 521--
531.

[15] Von Wahl, W.,
p
L -decay rates for homogeneous wave-equations, Math. Z. 120 (1971), 93--106.


10


International Advanced of Technology Congress (ATCi),PWTC, Malaysia. November 3-5, 2009

Department of Mathematics
Imperial College London
180 Queen's Gate, London SW7 2AZ
United Kingdom
E-mail address: [email protected]