6 Introduction
DNLS equation was obtained [7,37,38] in the form of the set of coupled equa-
tions
∂
x
∂
t
h
±
+ ∂
x
|h|
2
h
±
± i∂
2
x
h
±
= −∇
2
⊥
h
+
+ h
−
, (0.15)
where h
±
= h
y
± ih
z
. In the case of the negative dispersion, solutions in the
form of the one-dimensional solitons were numerically found [7,37,38]; and no
numerical study of non-one-dimensional solutions was done. The exact two-
dimensional solitary solutions of the set (0.15) can in principle be obtained
by the IST method, similar to the case of the KP equation; it is necessary to
construct the corresponding Lax
ˆ
L–
ˆ
A pair [24].
Successes in the theoretical investigation of nonlinear physical systems
having soliton solutions has stimulated experimental studies in various ar-
eas of the physics of nonlinear waves. Here, we would like to mention the
laboratory and space experiments on excitation, evolution, and interaction
dynamics of the ion-acoustic and Alfv´en solitons as well as the experimental
studies of the structure of plasma shock waves [39–50], the modeling of the
solitons’ dynamics in the electric circuits [51,52], the experiments on surface
waves and internal waves in rotating vessels and hydro-trays [53,54], etc. This
in turn has raised new questions and demonstrated the urgency of the theo-
retical investigation of such problems as the stability of non-one-dimensional
solitons, the dynamics of their interaction, the nonlinear resonances and for-
mation of the bound states, the effects of small corrections in equations of
the class (0.3), the influence of dissipation on the structure and evolution
of the non-one-dimensional and nonlinear waves and solitons, the wave self-
influence effects (the wave collapse and the wave self-focusing), etc.
It was shown, in Refs. [16,55-57] for some particular cases (concrete values
of the coefficients) of the equation (0.10) with ∆
⊥
= ∂
2
y
, and in Ref. [58] for
arbitrary α, β, and κ, that for the negative dispersion (β/κ < 0) the one-
dimensional solitons are stable whereas for the positive dispersion (β/κ > 0)
they are unstable with respect to growth of infinitesimal perturbations. The
case of very small k was studied [16] by the Krylov–Bogolyubov method;in
Ref. [55], the term ∂
4
yyxx
u was added on the right-hand side of the equation
and the problem was solved by the method of the variation of the action with
the Lagrangian integrated over x and the test functions in the solitary-like
form; in Ref. [56], the IST method was used, and in Ref. [57], the Lyapunov’s
functional method was used. In Ref. [58], the numerical investigation was
done. It was demonstrated that for β/κ < 0 the perturbation can easily
transfer from the soliton into the medium and spread in all directions while
for β/κ > 0 it is “locked” in the soliton. In the region of localization of the
perturbation, the soliton’s velocity differs from the velocity of an unperturbed
soliton. This, taking into account the spreading, leads to the growth of the
perturbation.
For two-dimensional and three-dimensional solitons of the KP equa-
tion, the stability analysis is highly non-trivial. The most illustrative study