Pseudo-Exponential-Type Solutions of Wave Equations Depending on Several Variables

Using matrix identities, we construct explicit pseudo-exponential-type solutions of linear Dirac, Loewner and Schrödinger equations depending on two variables and of nonlinear wave equations depending on three variables.

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Symmetry, Integrability and Geometry: Methods and Applications
Datum:2015
Hauptverfasser: Fritzsche, B., Kirstein, B., Roitberg, I.Y., Sakhnovich, A.L.
Format: Artikel
Sprache:English
Veröffentlicht: Інститут математики НАН України 2015
Online Zugang:https://nasplib.isofts.kiev.ua/handle/123456789/146994
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Назва журналу:Digital Library of Periodicals of National Academy of Sciences of Ukraine
Zitieren:Pseudo-Exponential-Type Solutions of Wave Equations Depending on Several Variables / B. Fritzsche, B. Kirstein, I.Y. Roitberg, A.L. Sakhnovich // Symmetry, Integrability and Geometry: Methods and Applications. — 2015. — Т. 11. — Бібліогр.: 53 назв. — англ.

Institution

Digital Library of Periodicals of National Academy of Sciences of Ukraine
id nasplib_isofts_kiev_ua-123456789-146994
record_format dspace
spelling Fritzsche, B.
Kirstein, B.
Roitberg, I.Y.
Sakhnovich, A.L.
2019-02-12T18:05:38Z
2019-02-12T18:05:38Z
2015
Pseudo-Exponential-Type Solutions of Wave Equations Depending on Several Variables / B. Fritzsche, B. Kirstein, I.Y. Roitberg, A.L. Sakhnovich // Symmetry, Integrability and Geometry: Methods and Applications. — 2015. — Т. 11. — Бібліогр.: 53 назв. — англ.
1815-0659
2010 Mathematics Subject Classification: 35C08; 35Q41; 15A24
DOI:10.3842/SIGMA.2015.010
https://nasplib.isofts.kiev.ua/handle/123456789/146994
Using matrix identities, we construct explicit pseudo-exponential-type solutions of linear Dirac, Loewner and Schrödinger equations depending on two variables and of nonlinear wave equations depending on three variables.
The research of I.Ya. Roitberg was supported by the German Research Foundation (DFG) under grant No. KI 760/3-1. The research of A.L. Sakhnovich was supported by the Austrian Science Fund (FWF) under Grant No. P24301.
en
Інститут математики НАН України
Symmetry, Integrability and Geometry: Methods and Applications
Pseudo-Exponential-Type Solutions of Wave Equations Depending on Several Variables
Article
published earlier
institution Digital Library of Periodicals of National Academy of Sciences of Ukraine
collection DSpace DC
title Pseudo-Exponential-Type Solutions of Wave Equations Depending on Several Variables
spellingShingle Pseudo-Exponential-Type Solutions of Wave Equations Depending on Several Variables
Fritzsche, B.
Kirstein, B.
Roitberg, I.Y.
Sakhnovich, A.L.
title_short Pseudo-Exponential-Type Solutions of Wave Equations Depending on Several Variables
title_full Pseudo-Exponential-Type Solutions of Wave Equations Depending on Several Variables
title_fullStr Pseudo-Exponential-Type Solutions of Wave Equations Depending on Several Variables
title_full_unstemmed Pseudo-Exponential-Type Solutions of Wave Equations Depending on Several Variables
title_sort pseudo-exponential-type solutions of wave equations depending on several variables
author Fritzsche, B.
Kirstein, B.
Roitberg, I.Y.
Sakhnovich, A.L.
author_facet Fritzsche, B.
Kirstein, B.
Roitberg, I.Y.
Sakhnovich, A.L.
publishDate 2015
language English
container_title Symmetry, Integrability and Geometry: Methods and Applications
publisher Інститут математики НАН України
format Article
description Using matrix identities, we construct explicit pseudo-exponential-type solutions of linear Dirac, Loewner and Schrödinger equations depending on two variables and of nonlinear wave equations depending on three variables.
issn 1815-0659
url https://nasplib.isofts.kiev.ua/handle/123456789/146994
citation_txt Pseudo-Exponential-Type Solutions of Wave Equations Depending on Several Variables / B. Fritzsche, B. Kirstein, I.Y. Roitberg, A.L. Sakhnovich // Symmetry, Integrability and Geometry: Methods and Applications. — 2015. — Т. 11. — Бібліогр.: 53 назв. — англ.
work_keys_str_mv AT fritzscheb pseudoexponentialtypesolutionsofwaveequationsdependingonseveralvariables
AT kirsteinb pseudoexponentialtypesolutionsofwaveequationsdependingonseveralvariables
AT roitbergiy pseudoexponentialtypesolutionsofwaveequationsdependingonseveralvariables
AT sakhnovichal pseudoexponentialtypesolutionsofwaveequationsdependingonseveralvariables
first_indexed 2025-11-24T15:49:12Z
last_indexed 2025-11-24T15:49:12Z
_version_ 1850473573890654208
fulltext Symmetry, Integrability and Geometry: Methods and Applications SIGMA 11 (2015), 010, 13 pages Pseudo-Exponential-Type Solutions of Wave Equations Depending on Several Variables Bernd FRITZSCHE †, Bernd KIRSTEIN †, Inna Ya. ROITBERG † and Alexander L. SAKHNOVICH ‡ † Fakultät für Mathematik und Informatik, Universität Leipzig, Augustusplatz 10, D-04009 Leipzig, Germany E-mail: fritzsche@math.uni-leipzig.de, kirstein@math.uni-leipzig.de, innaroitberg@gmail.com ‡ Fakultät für Mathematik, Universität Wien, Oskar-Morgenstern-Platz 1, A-1090 Vienna, Austria E-mail: oleksandr.sakhnovych@univie.ac.at Received September 04, 2014, in final form January 23, 2015; Published online January 29, 2015 http://dx.doi.org/10.3842/SIGMA.2015.010 Abstract. Using matrix identities, we construct explicit pseudo-exponential-type solutions of linear Dirac, Loewner and Schrödinger equations depending on two variables and of nonlinear wave equations depending on three variables. Key words: Bäcklund–Darboux transformation; matrix identity; S-node; S-multinode; expli- cit solution; non-stationary Dirac equation; non-stationary Schrödinger equation; Loewner system; pseudo-exponential-type potential; integrable nonlinear equations 2010 Mathematics Subject Classification: 35C08; 35Q41; 15A24 1 Introduction The term pseudo-exponential potentials was introduced in [20] (see Remark 1.2 on interrela- tions between pseudo-exponential-type potentials and multi-soliton solutions). Ordinary linear differential equations with the so called pseudo-exponential-type potentials were actively studied (see [14, 15, 20, 21, 22, 38, 41] and references therein), since their solutions could be constructed explicitly (and inverse problems to recover these equations from rational Weyl functions or reflection coefficients could be solved explicitly). Thus, pseudo-exponential-type potentials and solutions, that is, potentials and solutions, which, roughly speaking, rationally depend on ma- trix exponentials, are of a special interest. When matrices in the matrix exponentials (from the rational functions of matrix exponentials) are nilpotent, purely rational functions (potentials) appear as an important subcase of the pseudo-exponential-type potentials. For a more rigorous definition of the term pseudo-exponential potential see, for example, [15, 20]. Explicit solutions of linear and nonlinear wave equations are important both in theory and applications. The theory is well-developed for the case of linear equations depending on one variable and nonlinear integrable equations depending on two variables and includes, in particular, algebro-geometric methods and several versions of the commutation methods and of Bäcklund–Darboux transformations (BDTs), see some results and various references in [10, 12, 16, 17, 18, 23, 32, 41, 52]. In spite of numerous interesting results on the cases of more variables (see, e.g., [1, 5, 7, 8, 13, 31, 33, 34, 37, 46, 48, 49]), these cases are more complicated and contain also more open problems. Matrix identities are actively used in this theory for the cases of one and several space variables starting from the seminal work [30]. By matrix (or operator) identities we mean an important subclass of so called Sylvester equations AX − Y B = Q, which are considered, for mailto:fritzsche@math.uni-leipzig.de mailto:kirstein@math.uni-leipzig.de mailto:innaroitberg@gmail.com mailto:oleksandr.sakhnovych@univie.ac.at http://dx.doi.org/10.3842/SIGMA.2015.010 2 B. Fritzsche, B. Kirstein, I.Ya. Roitberg and A.L. Sakhnovich instance, in control theory. Namely, matrix identities are equations of the form AR − RB = Q or, more often, AR − RB = Π1Π∗2 (see, e.g., [35, 42, 44]) with Πk of comparatively small rank. V.A. Marchenko [30] was the first to apply matrix and operator identities in this topic (see [46] and references therein for further developments of his approach). In another way (more precisely, for the construction of τ -functions) matrix identities were used in [25]. Our approach is based on the GBDT (generalized BDT) approach, which was introduced in [35, 36] (see further results and many references in [14, 15, 20, 39, 41]). Although the papers [35, 36] were initiated by [30], matrix identities in [30] and in GBDT are used in quite different ways. Moreover, solutions of the nonlinear equations are constructed in [30] as reductions of expressions of the form Γ−1Γx whereas GBDT is a kind of a binary Darboux transformation and solutions are expressed via matrix functions Φ∗2S −1Φ1. (Here Γ, Φ1 and Φ2 satisfy some simple auxiliary linear systems.) See, for instance, (4.5) for solutions in terms of Φ∗2S −1Φ1. Matrices of a much lesser order have to be inverted in GBDT when constructing, for instance, matrix solutions of nonlinear equations. In addition, Darboux matrices and wave functions are constructed explicitly using GBDT. The method develops during the last 20 years. Moreover, after the publication of [35, 36] a very close approach was used by M. Manas (see some comparative analysis in [10]) and related formulas are now successfully used by Mueller-Hoissen and coauthors (see, e.g., [13]). In our paper we apply multidimensional versions of the GBDT. That is, we follow [37] (where S-nodes introduced in [42, 43, 44] were applied to matrix Kadomtsev–Petviashvili equa- tions) and the S-multinodes approach from [40] in order to construct explicitly pseudo-exponen- tial-type potentials and solutions of some important equations of mathematical physics depen- ding on several variables. The transfer to S-multinodes is required in many examples because the same matrix should satisfy several matrix identities. S-multinodes first appeared in [40] as a certain generalization of the S-nodes on one hand and commutative colligations (introduced by M.S. Livšic [27]) on the other hand. A symmetric S-multinode (r-node) is a set of matrices{ A1, . . . , Ar; ν1, . . . , νr;R; Ĉ } such that for 1 ≤ i, k ≤ r the relations AiAk = AkAi, AkR+RA∗k = ĈνkĈ ∗, R = R∗, νk = ν∗k (1.1) hold. Here we shall deal with the cases r = 1, 2, 3. In the case r = 1 we have the well-known symmetric S-node introduced by L.A. Sakhnovich (see, e.g., [41, 42, 43, 44, 45] for various applications). For r > 1 the situation is more complicated, since R in general position is defined already by one of the identities AkR+RA∗k = ĈνkĈ ∗. However, the construction of S-multinodes proves both possible and useful. Remark 1.1. In our further considerations the matrices in the S-multinode or S-node (i.e., matrices in (1.1)) are constant and each S-multinode generates a potential and solution of a linear (or solution of a nonlinear) equation. Remark 1.2. We note that pseudo-exponential-type solutions are close to multi-soliton solu- tions and their analogues. However, multi-soliton solutions are usually generated when matri- ces Ai are diagonal, whereas we do not require Ai to be necessarily diagonal. This correspondence for the solutions of sine-Gordon and sinh-Gordon equations was studied in [35, Section 4]. In particular, it was shown in [35] that solutions of sine-Gordon equation from [9, 24] are derived in this way (i.e., using S-nodes with diagonal matrices A1). Explicit solutions of linear equations (especially, of non-stationary Dirac and Schrödinger equations) are of wide interest, and in Section 2 we use 2-nodes in order to study the case of Pseudo-Exponential-Type Solutions of Wave Equations Depending on Several Variables 3 the non-stationary Dirac system HΨ = 0, H := ∂ ∂t + σ2 ∂ ∂y − iV (t, y), σ2 = [ 0 −i i 0 ] , V = V ∗, (1.2) which presents more difficulties than the non-stationary (time-dependent) Schrödinger equation considered in [40]. Some new results for the non-stationary Schrödinger equation are derived in Section 3. Thus, we fill in the gap between papers [37] and [40], consider a class of solutions of the Schrödinger equation, which is wider than the one discussed in [1], and construct interesting examples. Section 4 is dedicated to the nonlinear integrable equations. As examples we consider ma- trix Davey–Stewartson I (DS I) and generalized nonlinear optics equations. In particular, our approach allows to construct a wide class of rational solutions of matrix DS I (see Remark 4.3). Remark 1.3. GBDT results for DS I and generalized nonlinear optics equation were obtained in [36, Section 3] but no examples were given. Here we construct wide classes of solutions using the S-node (S-multinode) approach, see Propositions 4.2 and 4.6. We note that GBDT results in [36, Section 3] include the case of nonzero background (in which situation auxiliary linear systems play a more essential role) and it would be very interesting to generalize S-multinode approach for that case. As usual, N denotes the set of natural numbers, const stands for a constant (number or matrix), Im(A) stands for the image of the matrix A, σ(D) stands for the spectrum of D, [G,F ] stands for the commutator GF − FG, ⊗ stands for Kronecker product, Ip is the p× p identity matrix, and Ψtx := ∂ ∂x ( ∂ ∂tΨ ) = ∂2 ∂x∂tΨ. By diag{b1, b2, . . . , bm} we denote the diagonal matrix with the entries b1, b2, . . . on the main diagonal. 2 Dirac and Loewner equations: explicit solutions 2.1 Non-stationary Dirac equation We note that in the GBDT version of the Bäcklund–Darboux transformation the solution of the transformed equation is represented in the form Π∗S−1, where Π∗ is a matrix solution of the initial equation and the matrix function S is constructed using the S-node (see, e.g., [39, 41] and references therein). Here we construct solutions of (1.2) in the same form. Namely, we set Π = CEA(t, y)Ĉ, EA = exp{tA1 + yA2}, A1A2 = A2A1, Ĉ = [ g∗1 g∗2 ] , (2.1) where Ĉ is an N × 2 matrix, g∗1 and g∗2 are columns of Ĉ, A1 and A2 are N ×N matrices and C is an n×N matrix (n,N ∈ N). We emphasize that the matrices A1, A2, Ĉ and C are constant (see also Remark 1.1). We assume that the equalities g1A ∗ 1 − ig2A ∗ 2 = 0, g2A ∗ 1 + ig1A ∗ 2 = 0 (2.2) hold. From (2.1) and (2.2), we easily see that H0Π∗ = 0, H0 := ∂ ∂t + σ2 ∂ ∂y , (2.3) where H0 is applied to Π∗ columnwise. Recall that matrices A1, A2, R, ν1, ν2 and Ĉ form a symmetric 2-node if A1 and A2 commute and the following identities are valid: AkR+RA∗k = ĈνkĈ ∗, k = 1, 2, R = R∗, νk = ν∗k . (2.4) 4 B. Fritzsche, B. Kirstein, I.Ya. Roitberg and A.L. Sakhnovich It is immediate that the matrix function S(t, y) = S0 + CEA(t, y)REA(t, y)∗C∗, S0 = S∗0 ≡ const, (2.5) satisfies equations ∂ ∂tS = Πν1Π∗ and ∂ ∂yS = Πν2Π∗. These equations and equation (2.3) yield the proposition below. Proposition 2.1. Let relations (2.1), (2.2), (2.4) and (2.5) hold and assume that ν1 = σ2, ν2 = −I2. Then, in the points of invertibility of S, we have H ( Π(t, y)∗S(t, y)−1 ) = 0, where H has the form (1.2) with V defined by V := i ( Π∗S−1Πσ2 − σ2Π∗S−1Π ) . The important part of the problem is to find the cases where the conditions of Proposition 2.1 hold. Then we obtain families of explicitly constructed potentials V and solutions Π∗S−1 of the corresponding Dirac systems. Example 2.2. Set g2 = −ig1jn, A1 = D = diag{D1, D2} (where D1 and D2 are n1 × n1 and n2×n2 diagonal blocks of the diagonal matrix D, n1+n2 = n, σ(Dk)∩σ(−D∗k) = ∅ for k = 1, 2), A2 = Djn and jn := [ In1 0 0 −In2 ] , R = [ R11 0 0 R22 ] . We uniquely define R11 and R22 by the matrix identities D1R11 +R11D ∗ 1 = −g∗1(In + jn)g1, D2R22 +R22D ∗ 2 = g∗1(In − jn)g1. Then the conditions of Proposition 2.1 hold. Thus, according to Proposition 2.1 and Example 2.2, each vector g1 and diagonal matrix D (such that σ(Dk) ∩ σ(−D∗k) = ∅) determine a set (depending on the choice of C and S0) of pseudo-exponential-type potentials and explicit solutions of (1.2). 2.2 Loewner’s system Loewner’s system has the form Ψx = L(x, y)Ψy, (2.6) where L is an m×m matrix function. For the case m = 2, this system was studied by C. Loewner in the seminal paper [28] and applications to the hodograph equation were obtained. In [29], C. Loewner rewrote in this way the system xη − yξ = 0, (ρx)ξ + (ρy)η = 0, which describes a steady compressible and irrotational flow of an ideal fluid. For the Loewner’s system, its transformations, generalizations and applications, see also [47, 50] and references therein. (For some special kinds of similarity transformations of L see also [28, formulas (5.10a) and (5.27)].) Direct calculation proves the following proposition. Proposition 2.3. Let m×m and m×n, respectively, matrix functions Λ1 and Λ2 satisfy a linear differential equation (Λi)x = q1(x, y)(Λi)y + q0(x, y)Λi, i = 1, 2, where the coefficients q0 and q1 are some m × m matrix functions. Then, in the points of invertibility of Λ1, the matrix function Ψ = Λ−1 1 Λ2 satisfies the Loewner equation (2.6), where L = Λ−1 1 q1Λ1. Pseudo-Exponential-Type Solutions of Wave Equations Depending on Several Variables 5 Pseudo-exponential-type Ψ and L are constructed in the next proposition. Proposition 2.4. Introduce m×m and m×n, respectively, matrix functions Λ1 and Λ2 by the equalities Λi = CiEA(x, y, i)Ĉi, i = 1, 2, (2.7) EA(x, y, i) := exp{xĂi + yÃi}, Ăi := D ⊗Ai, Ãi := Im ⊗Ai, D = diag{d1, . . . , dm}, Ci := m∑ k=1 (eke ∗ k)⊗ (e∗kci), where Ai are li × li matrices, ci are m × li matrices, Ĉ1 is an N1 ×m matrix, Ĉ2 is an N2 × n matrix, Ni = mli and li ∈ N. Here ⊗ is Kronecker product, ek is a column vector given by ek = {δjk}mj=1 and δjk is Kronecker’s delta. Then, in the points of invertibility of Λ1, the matrix functions Ψ = Λ−1 1 Λ2 and L = Λ−1 1 DΛ1 satisfy (2.6). Proof. It is easy to see that Λ1 and Λ2 given by (2.7) satisfy equation (Λi)x = D(Λi)y. Now, Proposition 2.4 follows from Proposition 2.3. � In a similar (to the construction of Λi in the proposition above) way, matrix functions Π satis- fying (4.19) are constructed in (4.21)–(4.23). 3 Non-stationary Schrödinger equation: explicit solutions and examples We consider the subcase of [40, Theorem 3.2], where S0 = S∗0 , and use notations Π instead of Ψ0, S instead of S and S0 instead of S0. We substitute α = i, k = 1, A1 = A, B1 = −A∗, ν1 = Ip, CΦ = Ĉ, CΨ = Ĉ∗, ĈΦ = C, ĈΨ = C∗ into [40, formula (3.1) and Theorem 3.2]. For this particular case, Theorem 3.2 from [40] takes the following form. Proposition 3.1. Fix some p, n,N ∈ N, an N ×N matrix A, an n ×N matrix C, an N × p matrix Ĉ and an n× n matrix S0 = S∗0 . Let R = R∗ satisfy the matrix identity AR+RA∗ = ĈĈ∗, (3.1) and put Π(x, t) = CeA(x, t)Ĉ, eA(x, t) := exp{xA− itA2}, (3.2) S(x, t) = S0 + CeA(x, t)ReA(x, t)∗C∗. (3.3) Then, the matrix function Π̃∗ := Π∗S−1 satisfies the vector non-stationary Schrödinger equation H ( Π̃∗ ) = 0, H := i ∂ ∂t + ∂2 ∂x2 − q̃(x, t), (3.4) where H is applied to Π̃∗ columnwise and q̃ is the p× p matrix function: q̃(x, t) = −2 ( Π(x, t)∗S(x, t)−1Π(x, t) ) x . (3.5) 6 B. Fritzsche, B. Kirstein, I.Ya. Roitberg and A.L. Sakhnovich Our approach allows to consider the cases of non-diagonal matrices A, and we adduce below several examples, where A is a 2 × 2 Jordan cell. Using some simple calculations, we easily construct eA, Π, S and, finally, solution Π̃∗ and potential q̃ in the following example of a scalar Schrödinger equation. Example 3.2. Let us put p = 1, N = n = 2, A = [ µ0 1 0 µ0 ] , Ĉ = [ ĉ1 ĉ2 ] , S0 = [ 0 b b d ] . (3.6) Formulas (3.1) and (3.6) yield (for R = {rij}2i,j=1) the equality AR+RA∗ = κR+ [ r12 + r21 r22 r22 0 ] , κ := µ0 + µ0. (3.7) From the definition of A we also obtain eA(x, t) = eµ0x−iµ20t ( I2 + [ 0 x− 2iµ0t 0 0 ]) . (3.8) Assume (in addition to (3.6)) that κ := µ0 + µ0 = 0, ĉ1 = 1, ĉ2 = 0, C = I2. (3.9) Taking into account (3.7) and the first three equalities in (3.9), we see that the relations R = R∗ and (3.1) are equivalent to the equalities r11 = r11, r21 = r12, r12 + r12 = 1, r22 = 0. (3.10) In view of (3.2), (3.3), (3.8) and (3.9), we have Π(x, t) = eµ0x−iµ20t [ 1 0 ] , S(x, t) = S0 + [ 1 x− 2iµ0t 0 1 ] R [ 1 0 x− 2iµ0t 1 ] . (3.11) Here we took into account that κ = 0 yields |eµ0x−iµ20t| = 1. From (3.5), (3.6), (3.10) and (3.11), after some simple calculations we derive Π̃(x, t)∗ = Π(x, t)∗S(x, t)−1 = ( c+ d(x− 2iµ0t) )−1 eiµ20t−µ0x [ d −r12 − b ] , c := dr11 − |r12 + b|2, (3.12) Π(x, t)∗S(x, t)−1Π(x, t) = d ( c+ d(x− 2iµ0t) )−1 , q̃(x, t) = 2d2 ( c+ d(x− 2iµ0t) )−2 . Clearly, this potential q̃ is rational, depends on one variable x − 2iµ0t and has singularity at certain values of x, t ∈ R. According to Proposition 3.1, each entry of Π̃∗ of the form (3.12) (in our case these entries are collinear) satisfies the Schrödinger equation with the potential q̃, which is given above. In the following example, the potential q̃ is rational and depends on two real-valued variab- les x and t or, equivalently, on one complex-valued variable P := x − iµ0t (and its complex conjugate P ). Pseudo-Exponential-Type Solutions of Wave Equations Depending on Several Variables 7 Example 3.3. Put p = n = 1, N = 2, S0 = 0, A = [ µ0 1 0 µ0 ] , κ := µ0 + µ0 > 0, Ĉ = [ 0 1 ] , C = [ 1 1 ] . (3.13) Using (3.7), we immediately check that R = κ−1 [ 2κ−2 −κ−1 −κ−1 1 ] . (3.14) Taking into account (3.3), (3.8), (3.13) and (3.14), we easily calculate S(x, t) = κ−1 ∣∣eµ0P (x,t) ∣∣2(2κ−2 − κ−1(P (x, t) + P (x, t) + 2) + |P (x, t) + 1|2 ) . (3.15) We sometimes omit the variables x, t in our further formulas. In view of (3.2), (3.8), (3.13) and (3.15) we derive Π∗S−1Π = κ|P + 1|2 2κ−2 − κ−1(P + P + 2) + |P + 1|2 . The rational potential q̃, which is given by (3.5), takes the form q̃ = 2 ( (P + 1)2 + (P + 1)2 − 2κ−1(P + P + 2) ) (2κ−2 − κ−1(P + P + 2) + |P + 1|2)2 . (3.16) Finally, the solution Π̃∗ = Π∗S−1 of the Schrödinger equation, where the potential q̃ has the form (3.16), is given by the formula: Π̃∗ = κe−µ0P (x,t)(x+ 2iµ0t+ 1) 2κ−2 − κ−1(P (x, t) + P (x, t) + 2) + |P (x, t) + 1|2 . It was shown in [37] that if σ(iA) ⊂ C+ and the pair A, Ĉ is full range, i.e., span N−1⋃ `=0 Im ( A`Ĉ ) = CN , then the solution R of (3.1) is unique and positive-definite, that is, R > 0. Hence, we obtain our next proposition. Proposition 3.4. Assume that σ(iA) ⊂ C+, the pair A, Ĉ is full range, rankC = n and S0 ≥ 0. Then we have S(x, t) > 0. Therefore, S(x, t) is invertible and the potential q̃ is nonsingular. In our next example we deal with a nonsingular pseudo-exponential potential depending on two variables. Example 3.5. Let the parameter matrices A, Ĉ and S0 have the form (3.6). Instead of the relations (3.9), we assume now that κ := µ0 + µ0 > 0, ĉ1 = 0, ĉ2 = 1, b = 0, d > 0, C = I2. (3.17) Like in Example 3.3, formula (3.7) again yields (3.14). Taking into account (3.2), (3.3), (3.6), (3.8), (3.14) and (3.17) we calculate Π∗S−1Π = Z1/Z2, Z1 = 2κ−3 + d ∣∣eµ0P ∣∣−2|P |2, Z2 = κ−4 + κ−1d ∣∣eµ0P ∣∣−2(|P |2 − κ−1(P + P ) + 2κ−2 ) , P := x− iµ0t. 8 B. Fritzsche, B. Kirstein, I.Ya. Roitberg and A.L. Sakhnovich Next, one easily obtains the derivatives of Z1 and Z2 with respect to x: (Z1)x = −κ(Z1 − 2κ−3) + d ∣∣eµ0P ∣∣−2 (P + P ), (Z2)x = −κ(Z2 − κ−4) + κ−1d ∣∣eµ0P ∣∣−2( P + P − 2κ−1 ) . Hence, in view of (3.5) and formulas for Zk and (Zk)x above, we have q̃ = −2 ( Π∗S−1Π ) x = 2κ−2Z2 − κ−3Z1 + 2κ−2d|e(µ0)|−2Z1 + d|e(µ0)|−2(P + P ) ( Z2 − κ−1Z1 ) = − 2d ∣∣eµ0P ∣∣−2 Z2 2 ( 8κ−5 − 3κ−4(P + P ) + κ−3 ( |P |2 + 2d ∣∣eµ0P ∣∣−2 (P + P ) ) + κ−2d ∣∣eµ0P ∣∣−2( 2|P |2 − (P + P )2 )) . The solution Π̃∗ = Π∗S−1 of (3.4) is given (in our case) by the formula Π̃∗ = ( e−µ0P /Z2 ) [ κ−2 + d ∣∣eµ0P ∣∣−2 P 2κ−3 − κ−2P ] . 4 Nonlinear integrable equations Among (2 + 1)-dimensional integrable equations, Kadomtsev–Petviashvili, Davey–Stewartson (DS) and generalized nonlinear optics (also called N -wave) equations are, perhaps, the most actively studied systems. S-nodes were applied to the construction and study of the pseudo- exponential, rational and nonsingular rational (so called multi-lump) solutions of the Kadomtsev– Petviashvili equations in [37]. Here we investigate the remaining two equations from the three above. 4.1 Davey–Stewartson equations The Davey–Stewartson equations are well-known in wave theory (see, e.g., [6, 11, 23, 26] and references therein). Since Davey–Stewartson equations (DS I and DS II) are natural multidi- mensional generalizations of the nonlinear Schrödinger equations (NLS), their matrix versions should also be of interest (similar to matrix versions of NLS, see, e.g., [4]). 1. The matrix DS I has the form iut − (uxx + uyy)/2 = uq1 − q2u, (4.1) (q1)x − (q1)y = 1 2 ( (u∗u)y + (u∗u)x ) , (q2)x + (q2)y = 1 2 ( (uu∗)y − (uu∗)x ) , (4.2) where u, q1 and q2 are m2 ×m1, m1 ×m1 and m2 ×m2 matrix functions, respectively (m1 ≥ 1, m2 ≥ 1). We note that another matrix version of the Davey–Stewartson equation, where m1 = m2, was dealt with in [26]. It is easy to see that in the scalar case m1 = m2 = 1 equations (4.1) and (4.2) are equivalent, for instance, to [23, p. 70, system (2.23)] (after setting in (2.23) ε = α = 1). GBDT version of the Bäcklund–Darboux transformation for the matrix DS I was constructed in [36]. When the initial DS I equation (in GBDT for DS I, see [36, Theorem 5]) is trivial, that is, when we set (in [36]) u0 ≡ 0 and Q0 ≡ 0, Theorem 5 from [36] takes the form: Proposition 4.1. Let an n×m (n ∈ N, m = m1 +m2) matrix function Π and an n×n matrix function S satisfy equations Πx = Πyj, Πt = −iΠyyj, j := [ Im1 0 0 −Im2 ] , (4.3) Pseudo-Exponential-Type Solutions of Wave Equations Depending on Several Variables 9 Sy = −ΠΠ∗, Sx = −ΠjΠ∗, St = i(ΠyjΠ ∗ −ΠjΠ∗y). (4.4) Partition Π into n×m1 and n×m2, respectively, blocks Φ1 and Φ2 (i.e., set Π =: [ Φ1 Φ2 ] ). Then, the matrix functions u = 2Φ∗2S −1Φ1, q1 = 1 2 u∗u− 2 ( Φ∗1S −1Φ1 ) y , q2 = −1 2 uu∗ + 2 ( Φ∗2S −1Φ2 ) y (4.5) satisfy (in the points of invertibility of S) DS I system (4.1), (4.2). Introduce Φ1, Φ2 and S via relations Φ1(x, t, y) = C1E1(x, t, y)Ĉ1, E1(x, t, y) := exp { (x+ y)A1 − itA2 1 } , (4.6) Φ2(x, t, y) = C2E2(x, t, y)Ĉ2, E2(x, t, y) := exp { (x− y)A2 + itA2 2 } , (4.7) S(x, t, y) = S0 + C1E1(x, t, y)R1E1(x, t, y)∗C∗1 − C2E2(x, t, y)R2E2(x, t, y)∗C∗2 , S0 = S∗0 , (4.8) where C1 and C2 are n × N matrices, A1, A2, R1 = R∗1 and R2 = R∗2 are N × N matrices, Ĉ1 and Ĉ2 are N ×m1 and N ×m2, respectively, matrices, S0 is an n×n matrix and the following identities hold: A1R1 +R1A ∗ 1 = −Ĉ1Ĉ ∗ 1 , A2R2 +R2A ∗ 2 = −Ĉ2Ĉ ∗ 2 . (4.9) It is immediate from (4.6)–(4.9) that Π = [ Φ1 Φ2 ] and S satisfy relations (4.3) and the first two relations in (4.4). In order to prove the third equality in (4.4), we note that( C1E1R1E ∗ 1C ∗ 1 ) t = −iC1E1 ( A2 1R1 −R1 ( A2 1 )∗) E∗1C ∗ 1 = −iC1E1 ( A1(A1R1 +R1A ∗ 1)− (A1R1 +R1A ∗ 1)A∗1 ) E∗1C ∗ 1 = i ( (Φ1)yΦ ∗ 1 − Φ1(Φ∗1)y ) . (4.10) Here we used (4.6) and the first identity in (4.9). In a similar way we show that( C2E2R2E ∗ 2C ∗ 2 ) t = i ( (Φ2)yΦ ∗ 2 − Φ2(Φ∗2)y ) . (4.11) Equalities (4.8), (4.10) and (4.11) yield the last equality in (4.4). Hence, the conditions of Proposition 4.1 are valid, and so we proved the following proposition. Proposition 4.2. Let Φ1, Φ2 and S be given by the formulas (4.6)–(4.8) and assume that (4.9) holds. Then, the matrix functions u, q1 and q2 given by (4.5) satisfy (in the points of invertibility of S) DS I system (4.1), (4.2). Remark 4.3. It is easy to see that if σ(A1) = σ(A2) = 0, then Φ1, Φ2 and S are rational matrix functions. Thus, if σ(A1) = σ(A2) = 0, the solutions u, q1 and q2 of the DS I system, which are constructed in Proposition 4.2, are also rational matrix functions. Remark 4.4. Note that matrices considered in (4.9) form two separate S-nodes or, equivalently, an S-node, where R is a block diagonal matrix and the matrix identity[ A1 0 0 A2 ] R+R [ A∗1 0 0 A∗2 ] = − [ Ĉ1Ĉ ∗ 1 0 0 Ĉ2Ĉ ∗ 2 ] , R := [ R1 0 0 R2 ] is valid. Another example of a block diagonal matrix R is dealt with in Subsection 4.2. It would also be of interest to compare solutions of the same system constructed using r1-nodes and r2-nodes (r1 6= r2). 10 B. Fritzsche, B. Kirstein, I.Ya. Roitberg and A.L. Sakhnovich 2. The compatibility condition wtx = wxt of the auxiliary systems wx = ±ijwy + jV w, wt = 2ijwyy ± 2jV wy ± jQw, (4.12) where V = [ 0 u u∗ 0 ] , Q = [ q1 uy ∓ iux u∗y ± iu∗x −q2 ] , (4.13) qk(x, t) = −qk(x, t)∗, k = 1, 2, (4.14) is equivalent (for the case that the solution w is a non-degenerate matrix function) to the matrix DS II equation ut + i(uxx − uyy) = ±(q1u− uq2), (4.15) (q1)x ∓ i(q1)y = (uu∗)y ∓ i(uu∗)x, (q2)x ± i(q2)y = (u∗u)y ± i(u∗u)x. (4.16) As we see from (4.12)–(4.16), there are two versions of auxiliary systems and corresponding DS II equations. After setting m1 = m2 = 1 (and setting also ε = 1, α = ∓i in [23, p. 70, system (2.23)]), like for the scalar DS I case, equations (4.15) and (4.16) are equivalent to [23, p. 70, (2.23)]. Open problem. Use the approach from Proposition 4.1 in order to construct explicit pseudo- exponential solutions of the matrix DS II. We note that various results on DS II, including BDT results, are not quite analogous to the results on DS I (see, e.g., [23]). A quasi-determinant approach to explicit solution of noncom- mutative DS equations is presented in [19]. 4.2 Generalized nonlinear optics equation The integrability of the generalized nonlinear optics equation (GNOE) [D, ξt]− [D̃, ξx] = [ [D, ξ], [D̃, ξ] ] +DξyD̃ − D̃ξyD, (4.17) ξ(x, t, y)∗ = Bξ(x, t, y)B, B = diag{b1, b2, . . . , bm}, bk = ±1, (4.18) D = diag{d1, d2, . . . , dm} > 0, D̃ = diag{d̃1, d̃2, . . . , d̃m} > 0 was dealt with in [2, 53]. This system is a generalization of the well-known N -wave (nonlinear optics) equation [D, ξt]−[D̃, ξx] = [[D, ξ], [D̃, ξ]] first studied in [51] (see also [3]). GBDT version of the Bäcklund–Darboux transformation for GNOE was constructed in [36]. When the initial system in GBDT for GNOE [36, Theorem 4] is trivial (i.e., ξ0 ≡ 0), Theorem 4 from [36] takes the form: Proposition 4.5. Let an n × m matrix function Π and an n × n matrix function S satisfy equations Πx = ΠyD, Πt = ΠyD̃, (4.19) Sy = −ΠBΠ∗, Sx = −ΠBDΠ∗, St = −ΠBD̃Π∗. (4.20) Then the matrix function ξ = Π∗S−1ΠB satisfies (in the points of invertibility of S) GNOE (4.17) and reduction condition (4.18). Pseudo-Exponential-Type Solutions of Wave Equations Depending on Several Variables 11 In order to construct pseudo-exponential-type solutions ξ, we will consider matrix functions Π and S of the form (2.1) and (2.5), respectively, where EA will depend on three variables and N = ml, l ∈ N. Namely, we set Π(x, t, y) = CEA(x, t, y)Ĉ, EA(x, t, y) = exp{xA1 + tA2 + yA3}, (4.21) A1 = D ⊗A, A2 = D̃ ⊗A, A3 = Im ⊗A, (4.22) Ĉ = m∑ k=1 (eke ∗ k)⊗ (ĉek), ek = {δik}mi=1 ∈ Cm, (4.23) where C is an n×N matrix, A is an l× l matrix, N = ml, ⊗ is Kronecker product, ĉ is an l×m matrix, ek is a column vector and δik is Kronecker’s delta. It is immediate that the matrices Ak (k = 1, 2, 3) commute. Hence, we see that matrices A, C and ĉ determine (via (4.21)–(4.23)) matrix function Π satisfying (4.19). Proposition 4.6. Let relations (4.21)–(4.23) hold and set S(x, t, y) = S0 + CEA(x, t, y)REA(x, t, y)∗C∗, S0 = S∗0 , (4.24) where the N ×N matrix R (N = ml, R = R∗) satisfies matrix identities A1R+RA∗1 = −ĈBDĈ∗, A2R+RA∗2 = −ĈBD̃Ĉ∗, (4.25) A3R+RA∗3 = −ĈBĈ∗. (4.26) Then, the matrix function ξ = Π∗S−1ΠB satisfies (in the points of invertibility of S) GNOE (4.17) and reduction condition (4.18). Proof. We mentioned above that Π given by (4.21)–(4.23) satisfies (4.19). Moreover, re- lations (4.21) and (4.24)–(4.26) yield (4.20). Thus, the conditions of Proposition 4.5 are ful- filled. � We note that, according to (4.23), the right-hand sides of the equalities in (4.25) and (4.26) are block diagonal matrices with l× l blocks. Therefore, we will construct block diagonal matrix R, the blocks Rkk of which are also l × l matrices: R = diag{R11, R22, . . . , Rmm}. (4.27) Taking into account (4.22), we see that for R of the form (4.27) identities ARkk +RkkA ∗ = −bk(ĉek)(ĉek)∗, 1 ≤ k ≤ m, (4.28) imply that identities (4.25) and (4.26) hold. Corollary 4.7. Let relations (4.21)–(4.23) and (4.28) hold. Then, the matrix function ξ = Π∗S−1ΠB, where S is given by (4.24) and (4.27), satisfies (in the points of invertibility of S) GNOE (4.17) and reduction condition (4.18). Remark 4.8. If σ(A) ∩ σ(−A∗) = ∅, there exist unique solutions Rkk satisfying (4.28). For that case we have also Rkk = R∗kk (i.e., R = R∗). Clearly, Rkk is immediately recovered if σ(A) ∩ σ(−A∗) = ∅ and A is a diagonal matrix. Acknowledgements The research of I.Ya. Roitberg was supported by the German Research Foundation (DFG) under grant No. KI 760/3-1. The research of A.L. Sakhnovich was supported by the Austrian Science Fund (FWF) under Grant No. P24301. 12 B. Fritzsche, B. Kirstein, I.Ya. Roitberg and A.L. Sakhnovich References [1] Ablowitz M.J., Chakravarty S., Trubatch A.D., Villarroel J., A novel class of solutions of the non-stationary Schrödinger and the Kadomtsev–Petviashvili I equations, Phys. Lett. A 267 (2000), 132–146. [2] Ablowitz M.J., Haberman R., Nonlinear evolution equations – two and three dimensions, Phys. Rev. Lett. 35 (1975), 1185–1188. [3] Ablowitz M.J., Haberman R., Resonantly coupled nonlinear evolution equations, J. Math. Phys. 16 (1975), 2301–2305. [4] Ablowitz M.J., Prinari B., Trubatch A.D., Discrete and continuous nonlinear Schrödinger systems, London Mathematical Society Lecture Note Series, Vol. 302, Cambridge University Press, Cambridge, 2004. [5] Andrianov A.A., Borisov N.V., Ioffe M.V., The factorization method and quantum systems with equivalent energy spectra, Phys. Lett. A 105 (1984), 19–22. [6] Anker D., Freeman N.C., On the soliton solutions of the Davey–Stewartson equation for long waves, Proc. Roy. Soc. London Ser. A 360 (1978), 529–540. [7] Bagrov V.G., Samsonov B.F., Supersymmetry of a nonstationary Schrödinger equation, Phys. Lett. A 210 (1996), 60–64. [8] Bergvelt M., Gekhtman M., Kasman A., Spin Calogero particles and bispectral solutions of the matrix KP hierarchy, Math. Phys. Anal. Geom. 12 (2009), 181–200, arXiv:0806.2613. [9] Caudrey P.J., Gibbon J.D., Eilbeck J.C., Bullough R.K., Exact multisoliton solutions of the self-induced transparency and sine-Gordon equations, Phys. Rev. Lett. 30 (1973), 237–238. [10] Cieśliński J.L., Algebraic construction of the Darboux matrix revisited, J. Phys. A: Math. Theor. 42 (2009), 404003, 40 pages, arXiv:0904.3987. [11] Davey A., Stewartson K., On three-dimensional packets of surface waves, Proc. Roy. Soc. London Ser. A 338 (1974), 101–110. [12] Dimakis A., Müller-Hoissen F., Solutions of matrix NLS systems and their discretizations: a unified treat- ment, Inverse Problems 26 (2010), 095007, 55 pages, arXiv:1001.0133. [13] Dimakis A., Müller-Hoissen F., Binary Darboux transformations in bidifferential calculus and integrable reductions of vacuum Einstein equations, SIGMA 9 (2013), 009, 31 pages, arXiv:1207.1308. [14] Fritzsche B., Kirstein B., Roitberg I.Ya., Sakhnovich A.L., Weyl theory and explicit solutions of direct and inverse problems for Dirac system with a rectangular matrix potential, Oper. Matrices 7 (2013), 183–196, arXiv:1105.2013. [15] Fritzsche B., Kirstein B., Sakhnovich A.L., Completion problems and scattering problems for Dirac type differential equations with singularities, J. Math. Anal. Appl. 317 (2006), 510–525, math.SP/0409424. [16] Gesztesy F., Holden H., Soliton equations and their algebro-geometric solutions. Vol. I. (1 + 1)-dimensional continuous models, Cambridge Studies in Advanced Mathematics, Vol. 79, Cambridge University Press, Cambridge, 2003. [17] Gesztesy F., Holden H., Michor J., Teschl G., Soliton equations and their algebro-geometric solutions. Vol. II. (1 + 1)-dimensional discrete models, Cambridge Studies in Advanced Mathematics, Vol. 114, Cambridge University Press, Cambridge, 2008. [18] Gesztesy F., Teschl G., On the double commutation method, Proc. Amer. Math. Soc. 124 (1996), 1831–1840. [19] Gilson C.R., Macfarlane S.R., Dromion solutions of noncommutative Davey–Stewartson equations, J. Phys. A: Math. Theor. 42 (2009), 235202, 20 pages, arXiv:0901.4918. [20] Gohberg I., Kaashoek M.A., Sakhnovich A.L., Canonical systems with rational spectral densities: explicit formulas and applications, Math. Nachr. 194 (1998), 93–125. [21] Gohberg I., Kaashoek M.A., Sakhnovich A.L., Pseudo-canonical systems with rational Weyl functions: explicit formulas and applications, J. Differential Equations 146 (1998), 375–398. [22] Gohberg I., Kaashoek M.A., Sakhnovich A.L., Scattering problems for a canonical system with a pseudo- exponential potential, Asymptot. Anal. 29 (2002), 1–38. [23] Gu C., Hu H., Zhou Z., Darboux transformations in integrable systems. Theory and their applications to geometry, Mathematical Physics Studies, Vol. 26, Springer, Dordrecht, 2005. [24] Hirota R., Exact solution of the sine-Gordon equation for multiple collisions of solitons, J. Phys. Soc. Japan 33 (1972), 1459–1463. [25] Kasman A., Gekhtman M., Solitons and almost-intertwining matrices, J. Math. Phys. 42 (2001), 3540–3551, math-ph/0011011. http://dx.doi.org/10.1016/S0375-9601(00)00020-7 http://dx.doi.org/10.1103/PhysRevLett.35.1185 http://dx.doi.org/10.1063/1.522460 http://dx.doi.org/10.1016/0375-9601(84)90553-X http://dx.doi.org/10.1098/rspa.1978.0083 http://dx.doi.org/10.1098/rspa.1978.0083 http://dx.doi.org/10.1016/0375-9601(95)00832-2 http://dx.doi.org/10.1007/s11040-009-9058-y http://arxiv.org/abs/0806.2613 http://dx.doi.org/10.1103/PhysRevLett.30.237 http://dx.doi.org/10.1088/1751-8113/42/40/404003 http://arxiv.org/abs/0904.3987 http://dx.doi.org/10.1098/rspa.1974.0076 http://dx.doi.org/10.1088/0266-5611/26/9/095007 http://arxiv.org/abs/1001.0133 http://dx.doi.org/10.3842/SIGMA.2013.009 http://arxiv.org/abs/1207.1308 http://dx.doi.org/10.7153/oam-07-10 http://arxiv.org/abs/1105.2013 http://dx.doi.org/10.1016/j.jmaa.2005.05.044 http://arxiv.org/abs/math.SP/0409424 http://dx.doi.org/10.1017/CBO9780511546723 http://dx.doi.org/10.1017/CBO9780511543203 http://dx.doi.org/10.1090/S0002-9939-96-03299-6 http://dx.doi.org/10.1088/1751-8113/42/23/235202 http://arxiv.org/abs/0901.4918 http://dx.doi.org/10.1002/mana.19981940108 http://dx.doi.org/10.1006/jdeq.1998.3428 http://dx.doi.org/10.1007/1-4020-3088-6 http://dx.doi.org/10.1143/JPSJ.33.1459 http://dx.doi.org/10.1063/1.1379313 http://arxiv.org/abs/math-ph/0011011 Pseudo-Exponential-Type Solutions of Wave Equations Depending on Several Variables 13 [26] Leznov A.N., Yuzbashyan E.A., Multi-soliton solutions of the two-dimensional matrix Davey–Stewartson equation, Nuclear Phys. B 496 (1997), 643–653, hep-th/9612107. [27] Livšic M.S., Operator waves in Hilbert space and related partial differential equations, Integral Equations Operator Theory 2 (1979), 25–47. [28] Loewner C., A transformation theory of the partial differential equations of gas dynamics, Tech. Notes Nat. Adv. Comm. Aeronaut. 1950 (1950), no. 2065, 1–56. [29] Loewner C., Generation of solutions of systems of partial differential equations by composition of infinitesimal Baecklund transformations, J. Analyse Math. 2 (1953), 219–242. [30] Marchenko V.A., Nonlinear equations and operator algebras, Mathematics and its Applications (Soviet Series), Vol. 17, D. Reidel Publishing Co., Dordrecht, 1988. [31] Matveev V.B., Darboux transformation and explicit solutions of the Kadomtcev–Petviaschvily equation, depending on functional parameters, Lett. Math. Phys. 3 (1979), 213–216. [32] Matveev V.B., Salle M.A., Darboux transformations and solitons, Springer Series in Nonlinear Dynamics, Springer-Verlag, Berlin, 1991. [33] Pecheritsyn A.A., Pozdeeva E.O., Samsonov B.F., The Darboux transformation for the nonstationary Dirac equation, Russ. Phys. J. 48 (2005), 365–374. [34] Sabatier P.C., On multidimensional Darboux transformations, Inverse Problems 14 (1998), 355–366. [35] Sakhnovich A.L., Exact solutions of nonlinear equations and the method of operator identities, Linear Algebra Appl. 182 (1993), 109–126. [36] Sakhnovich A.L., Dressing procedure for solutions of nonlinear equations and the method of operator iden- tities, Inverse Problems 10 (1994), 699–710. [37] Sakhnovich A.L., Matrix Kadomtsev–Petviashvili equation: matrix identities and explicit non-singular so- lutions, J. Phys. A: Math. Gen. 36 (2003), 5023–5033. [38] Sakhnovich A.L., Bäcklund–Darboux transformation for non-isospectral canonical system and Riemann– Hilbert problem, SIGMA 3 (2007), 054, 11 pages, math-ph/0703072. [39] Sakhnovich A.L., On the GBDT version of the Bäcklund–Darboux transformation and its applications to linear and nonlinear equations and Weyl theory, Math. Model. Nat. Phenom. 5 (2010), 340–389, arXiv:0909.1537. [40] Sakhnovich A.L., The time-dependent Schrödinger equation of dimension k+1: explicit and rational solutions via GBDT and multinodes, J. Phys. A: Math. Theor. 44 (2011), 475201, 12 pages, arXiv:1104.2554. [41] Sakhnovich A.L., Sakhnovich L.A., Roitberg I.Ya., Inverse problems and nonlinear evolution equations. Solutions, Darboux matrices and Weyl–Titchmarsh functions, De Gruyter Studies in Mathematics, Vol. 47, De Gruyter, Berlin, 2013. [42] Sakhnovich L.A., On the factorization of the transfer matrix function, Sov. Math. Dokl. 17 (1976), 203–207. [43] Sakhnovich L.A., Integral equations with difference kernels on finite intervals, Operator Theory: Advances and Applications, Vol. 84, Birkhäuser Verlag, Basel, 1996. [44] Sakhnovich L.A., Spectral theory of canonical differential systems. Method of operator identities, Operator Theory: Advances and Applications, Vol. 107, Birkhäuser Verlag, Basel, 1999. [45] Sakhnovich L.A., Levy processes, integral equations, statistical physics: connections and interactions, Ope- rator Theory: Advances and Applications, Vol. 225, Birkhäuser/Springer Basel AG, Basel, 2012. [46] Schiebold C., Explicit solution formulas for the matrix-KP, Glasg. Math. J. 51 (2009), 147–155. [47] Schief W.K., Rogers C., Loewner transformations: adjoint and binary Darboux connections, Stud. Appl. Math. 100 (1998), 391–422. [48] Skljanin E.K., A class of potentials for the nonstationary Dirac equation, Zap. Nauchn. Sem. Leningrad. Otdel. Mat. Inst. Steklov. (LOMI) 77 (1978), 214–226. [49] Suzko A.A., Schulze-Halberg A., Darboux transformations and supersymmetry for the generalized Schrödinger equations in (1 + 1) dimensions, J. Phys. A: Math. Theor. 42 (2009), 295203, 14 pages. [50] Takebe T., Dispersionless BKP hierarchy and quadrant Löwner equation, SIGMA 10 (2014), 023, 13 pages, arXiv:1308.4584. [51] Zakharov V.E., Manakov S.V., The theory of resonance interaction of wave packets in nonlinear media, Soviet Phys. JETP 69 (1975), 1654–1673. [52] Zakharov V.E., Mikhailov A.V., On the integrability of classical spinor models in two-dimensional space- time, Comm. Math. Phys. 74 (1980), 21–40. [53] Zakharov V.E., Shabat A.B., A scheme for integrating the nonlinear equations of mathematical physics by the method of the inverse scattering problem. I, Funct. Anal. Appl. 8 (1974), 226–235. http://dx.doi.org/10.1016/S0550-3213(97)00264-2 http://arxiv.org/abs/hep-th/9612107 http://dx.doi.org/10.1007/BF01729359 http://dx.doi.org/10.1007/BF01729359 http://dx.doi.org/10.1007/BF02825638 http://dx.doi.org/10.1007/978-94-009-2887-9 http://dx.doi.org/10.1007/978-94-009-2887-9 http://dx.doi.org/10.1007/BF00405295 http://dx.doi.org/10.1007/978-3-662-00922-2 http://dx.doi.org/10.1007/s11182-005-0134-x http://dx.doi.org/10.1088/0266-5611/14/2/010 http://dx.doi.org/10.1016/0024-3795(93)90495-A http://dx.doi.org/10.1016/0024-3795(93)90495-A http://dx.doi.org/10.1088/0266-5611/10/3/013 http://dx.doi.org/10.1088/0305-4470/36/18/307 http://dx.doi.org/10.3842/SIGMA.2007.054 http://arxiv.org/abs/math-ph/0703072 http://dx.doi.org/10.1051/mmnp/20105415 http://arxiv.org/abs/0909.1537 http://dx.doi.org/10.1088/1751-8113/44/47/475201 http://arxiv.org/abs/1104.2554 http://dx.doi.org/10.1515/9783110258615 http://dx.doi.org/10.1007/978-3-0348-8986-5 http://dx.doi.org/10.1007/978-3-0348-8986-5 http://dx.doi.org/10.1007/978-3-0348-8713-7 http://dx.doi.org/10.1007/978-3-0348-8713-7 http://dx.doi.org/10.1007/978-3-0348-0356-4 http://dx.doi.org/10.1007/978-3-0348-0356-4 http://dx.doi.org/10.1017/S0017089508004862 http://dx.doi.org/10.1111/1467-9590.00082 http://dx.doi.org/10.1111/1467-9590.00082 http://dx.doi.org/10.1088/1751-8113/42/29/295203 http://dx.doi.org/10.3842/SIGMA.2014.023 http://arxiv.org/abs/1308.4584 http://projecteuclid.org/euclid.cmp/1103907930 http://dx.doi.org/10.1007/BF01075696 1 Introduction 2 Dirac and Loewner equations: explicit solutions 2.1 Non-stationary Dirac equation 2.2 Loewner's system 3 Non-stationary Schrödinger equation: explicit solutions and examples 4 Nonlinear integrable equations 4.1 Davey–Stewartson equations 4.2 Generalized nonlinear optics equation References