Decay estimates for the linear damped wave equation on the Heisenberg group

Alessandro Palmieri

REVIEW 1 major objections 4 minor 1 cited by

Decay estimates for the linear damped wave equation on the Heisenberg group

T0 review · 1 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read For the damped wave equation on the Heisenberg group, L2 norms and their derivatives satisfy heat-like decay rates with homogeneous dimension Q=2n+2.

desk verdict The L1-L2 decay estimates for the damped wave on H_n are a genuine and useful new result, but Theorem 1.1's L2-only estimates (5)-(6) are overstated and false as written. read the letter →

arxiv 1908.02657 v1 pith:SE6D3B73 submitted 2019-08-07 math.AP

Alessandro Palmieri This is my paper · ORCID

classification math.AP MSC 35L1035R0358J4533C4543A3043A80

keywords dampedwaveequationdecayestimatesHeisenberggroupFouriertransformHermitefunctionssub-LaplacianhomogeneousdimensionL1-L2

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper claims and argues that the damped wave equation $\partial_t^2 u - \Delta_H u + \partial_t u = 0$ on the Heisenberg group $\mathbf{H}_n$ obeys heat-like $L^2$ decay laws, with the homogeneous dimension $Q = 2n+2$ playing the role that the Euclidean dimension plays in the classical setting. For data in $H^1 \times L^2$, the horizontal gradient decays as $(1+t)^{-1/2}$ and the time derivative as $(1+t)^{-1}$; if the data also lie in $L^1$, the solution decays as $(1+t)^{-Q/4}$, the gradient as $(1+t)^{-Q/4-1/2}$, and the time derivative as $(1+t)^{-Q/4-1}$. The proof proceeds through the group Fourier transform and Hermite functions, diagonalizing the sub-Laplacian mode by mode and splitting frequencies into a low region with heat-like polynomial decay and a high region with exponential decay. These rates match, with $Q$ in place of $n$, the known Euclidean estimates that the paper sets out to generalize.

What carries the argument

The key machinery is the group Fourier transform of $\mathbf{H}_n$ paired with the Hermite basis of $L^2(\mathbb{R}^n)$. Under a Schr\"odinger representation $\pi_\lambda$, the sub-Laplacian becomes $d\pi_\lambda(\Delta_H) = -|\lambda| H_w$, where $H_w$ is the harmonic oscillator with Hermite eigenfunctions $e_k$ and eigenvalues $\mu_k = 2|k| + n$. Each Fourier coefficient $(\hat{u}(t,\lambda)e_k, e_\ell)_{L^2}$ solves the scalar damped oscillator equation $\partial_t^2 v + \partial_t v + \mu_k |\lambda| v = 0$. The proof splits $|\lambda|$ about the threshold $1/(8\mu_k)$: for small $|\lambda|$, the factor $e^{-\mu_k |\lambda| t}$ together with the bound $\|\hat{u}(\lambda)\| \le \|u\|_{L^1}$ and the convergence of $\sum_k \mu_k^{-(n+1)}$ yields the polynomial heat-like decay; for large $|\lambda|$, uniform exponential decay takes over. Plancherel's formula with measure $c_n |\lambda|^n d\lambda$ converts these mode-by-mode estimates into the stated $L^2(\mathbf{H}_n)$ bounds.

What would settle it

Test the identity after (42) on data whose group Fourier transform at one fixed $\lambda$ is the rank-one operator $(\cdot, e_0)e_0$. The left-hand side equals $|\lambda| |\hat{u}_{0,0}|^2$, while the right-hand side equals $4|\lambda| |\hat{u}_{0,0}|^2$, so the asserted equality fails; this calculation settles that the derivation as written is invalid, while leaving the rates themselves to be established through the two-sided inequality.

Watch

Extended reading notes

Core claim

The central assertion is that the heat-like decay law of the Euclidean damped wave equation extends to the Heisenberg group after replacing the Euclidean dimension $n$ by the homogeneous dimension $Q = 2n+2$. Theorem 1.1 states that, for data in $H^1(\mathbf{H}_n) \times L^2(\mathbf{H}_n)$, the solution satisfies $\|u(t,\cdot)\|_{L^2} \lesssim \|u_0\|_{L^2} + \|u_1\|_{L^2}$, while $\|\nabla_H u(t,\cdot)\|_{L^2} \lesssim (1+t)^{-1/2} (\|\nabla_H u_0\|_{L^2} + \|u_1\|_{L^2})$ and $\|\partial_t u(t,\cdot)\|_{L^2} \lesssim (1+t)^{-1} (\|\nabla_H u_0\|_{L^2} + \|u_1\|_{L^2})$. With additional $L^1$ regularity of the data, those rates improve to $(1+t)^{-Q/4}$, $(1+t)^{-Q/4-1/2}$, and $(1+t)^{-Q/4-1}$, respectively. The argument is a phase-space analysis: the group Fourier transform sends the sub-Laplacian to $|\lambda|$ times a harmonic oscillator, Hermite functions diagonalize that oscillator, and each resulting scalar equation is a damped oscillator whose low-frequency part behaves like a heat kernel and whose high-frequency part decays exponentially.

Load-bearing premise

The proof's gradient estimate depends on the equality displayed after (42), which identifies a weighted sum of Fourier coefficients of $u_0$ with twice the sum of Hilbert\,--\,Schmidt norms of the Fourier transforms of $(X_j + iY_j)u_0$ and $(X_j - iY_j)u_0$; that equality is false as stated, and only a comparable two-sided inequality is available, so the derivation needs an extra constant.

Editorial extensions

If this is right

If Theorem 1.1 is correct, $L^1(\mathbf{H}_n)$ data produce the same decay rates as the heat semigroup on $\mathbf{H}_n$: the solution decays as $(1+t)^{-Q/4}$ with $Q = 2n+2$.
In the $L^1$-improved regime, the time derivative decays faster than the horizontal gradient by a factor $(1+t)^{-1/2}$, matching the pattern of Euclidean damped waves.
Even without $L^1$ data, the horizontal gradient and time derivative decay polynomially, at rates $(1+t)^{-1/2}$ and $(1+t)^{-1}$, with constants depending only on $H^1 \times L^2$ norms.
The vertical derivative $Tu$ satisfies a decay estimate with the same rate as $\partial_t u$, at the price of requiring $T^{1/2} u_1 \in L^2$, as stated in Section 5.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Editorial inference: because only a two-sided comparison holds at the step after (42), the theorem's rates are probably still true with a larger constant; the proof requires an inequality in place of the displayed equality, not a conceptual change.
Editorial inference: if these decay rates are optimal, they should produce a Fujita-type critical exponent $1 + 2/Q$ for the semilinear damped wave equation with power nonlinearity on $\mathbf{H}_n$, matching the heat equation's critical exponent on the group; the paper's announced application points in this direction.
Editorial inference: the low-frequency/$L^1$ splitting should extend to other stratified Lie groups with an explicit Plancherel formula and Hermite-type eigenbases, yielding the same $Q$-based exponents whenever such bases are available.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The L1-L2 decay estimates (7)-(9) are the real content of this paper, and they look right. The Heisenberg group with the sub-Laplacian is the natural setting, and the homogeneous dimension Q=2n+2 enters exactly as in the Euclidean analogue. The proof is a serious adaptation of Matsumura's phase-space analysis: group Fourier transform, Hermite functions, low/high frequency splitting, and careful eigenvalue-series convergence. That part deserves credit and will be useful to people working on semilinear damped waves on sub-Riemannian structures.

The soft spots are concentrated in the L2-only estimates (5) and (6). They are stated with only ||∇_H u0|| + ||u1|| on the right, but the proofs in Remarks 4 and 5 actually produce low-frequency bounds with ||u0||. To remove that term you would need a global Poincaré inequality on H_n, which does not hold: data concentrated at low group-Fourier frequencies have fixed L2 norm and arbitrarily small horizontal gradient. The Euclidean analogue already shows the problem. Take u0 with Fourier frequency ε, unit L2 norm, ||∇u0||=ε. Then ||∇u(t)|| behaves like εe^{-ε^2t}, which for ε << t^{-1/2} is much larger than (1+t)^{-1/2}ε. So (5) and (6) fail as stated. The fix is straightforward: put ||u0|| back into the right-hand sides, as the paper's own intermediate inequalities already allow.

There is also a false displayed equality after (42). The sum of weighted Fourier coefficients is comparable, not equal, to twice the sum of the Hilbert-Schmidt norms of (X_j±iY_j)u0. The proof only needs the ≲ direction, so this is a presentational error rather than a gap in the gradient estimates, but it should be corrected.

The citation pattern is fine; [21] is used as a benchmark, not as a crutch, and the self-citations concern applications rather than the proof. No circularity.

Who is this for? People working on decay estimates for evolution equations on Lie groups and on Fujita-type exponents for semilinear damped waves. The main L1-L2 result is likely correct and worth having. I would send the paper to a serious referee, but with the clear expectation of major revision: the overclaimed L2-only estimates must be fixed and the false equality repaired. After that, this should be a solid contribution.

Referee Report

1 major / 4 minor

Summary. The paper studies the linear damped wave equation on the Heisenberg group H_n, namely ∂_t^2 u − Δ_H u + ∂_t u = 0, and derives L^2 decay estimates for the solution u, its horizontal gradient ∇_H u, and its time derivative ∂_t u. The main tool is the group Fourier transform combined with the Hermite spectral decomposition of the sub-Laplacian symbol, exactly in the spirit of WKB/phase-space analysis. Theorem 1.1 states L^2-only bounds (4)-(6) and, under additional L^1(H_n) regularity of the data, the heat-like decay bounds (7)-(9) with rates determined by the homogeneous dimension Q = 2n+2.

Significance. The L1-L2 estimates (7)-(9) are, if correct, a meaningful extension of Matsumura's Euclidean decay laws to a sub-Riemannian setting; the derivation is concrete, uses no fitted parameters, and the rates (1+t)^{-Q/4}, (1+t)^{-Q/4-1/2}, (1+t)^{-Q/4-1} match the expected diffusion asymptotics on H_n. The heavy use of Plancherel and Hermite functions is appropriate and the high-frequency part is handled by a clean exponential argument. However, the L2-only estimates (5)-(6), which the abstract advertises as a central goal, are false as stated; the proof in Remarks 4-5 does not establish them, and a concrete low-frequency datum shows the asserted rates fail. The false equality displayed after (42) is a further gap, although the needed inequality direction is true. Overall, the L1-L2 part appears defensible, but the main theorem as stated is not.

major comments (1)

[Theorem 1.1, (5)-(6); Remarks 4-5, Sections 4.2-4.3] The displayed identity after (42) is false as written. For a datum u0 with a single Hermite component e_0, the left-hand side equals |λ| ‖u0(λ)e_0‖^2, while the right-hand side equals 4|λ| ‖u0(λ)e_0‖^2 (each of (X_j±iY_j)u0 contributes 2|λ|‖u0e_0‖^2). Only the ≲ direction is needed for the K_high estimate, so the argument can be repaired by replacing the equality with a two-sided inequality after inserting the appropriate constants. The author should correct this step explicitly, since the displayed equality is incorrect and used without comment.

minor comments (4)

[Title and abstract] The title contains a typo: 'He isenberg group' should be 'Heisenberg group', and the abstract has a similar spacing issue. A careful proofreading is needed throughout.
[Introduction, paragraph 1] The word 'liturature' should be 'literature'.
[Section 4.1, after (26)] The sentence 'Note thatF (t,λ,k ) = ∂tG(t,λ,k )' is missing a space after 'that'. Please rewrite for clarity.
[References [26]-[28]] References [26] and [27] are preprints and [28] is an unpublished preprint of the author. If the final version is published, please update these entries; otherwise, the use of unpublished work for the Fujita-type critical exponent should be marked as forthcoming.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the decay estimates are derived from the Plancherel formula, Hermite spectral decomposition, and elementary ODE estimates, with no fitted parameter and no load-bearing self-citation.

full rationale

The paper's derivation is self-contained rather than circular. Theorem 1.1 is proved by applying the group Fourier transform, diagonalizing the sub-Laplacian through Hermite functions, solving the resulting parameter-dependent ODE for each Fourier coefficient, and then splitting the Plancherel integral into low- and high-frequency regions. The low-frequency estimates use the Riemann-Lebesgue-type bound (10) under additional L1 data, while the high-frequency estimates use Plancherel and L2 regularity. No quantity appearing in the conclusions is used as an input to obtain itself, and no parameter is fitted to the decay rates being predicted. The Euclidean result [21] is cited only in Remark 1 as a comparison, not as a premise of the proof. The self-references [26]-[28] concern heat-equation lifespan bounds and future semilinear applications, and none is load-bearing for the proof of Theorem 1.1. Two mathematical weaknesses do appear, but neither is circularity: the displayed equality after (42) is not exact as stated, although the needed argument only requires a comparable upper bound; and the L2-only estimates (5)-(6) as stated may require an unavailable Poincare-type inequality to remove an ||u0|| term. These are correctness or presentation concerns, not reductions of the derived estimates to their own inputs. Hence the circularity score is 0.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

No free parameters or invented entities: the proof is a parameter-free derivation using standard harmonic analysis on H_n and Hermite functions. The axioms are the Plancherel formula, the Hermite spectral decomposition, the infinitesimal representation formulas, and the well-posedness assumption. One additional input, the comparability of Fourier-weighted sums to gradient HS norms, is stated as an exact identity that is actually false.

assumptions (6)

standard math Plancherel formula on H_n: ‖f‖_L2(H_n)^2 = cn ∫_{R*} ‖f^(λ)‖_HS^2 |λ|^n dλ (Eq. (11)).
Used throughout to convert L2 norms of u, ∇_H u and ∂_t u into Hilbert-Schmidt norms of their group Fourier transforms.
standard math Hermite functions {e_k} form an orthonormal basis of L2(R^n) and diagonalize the harmonic oscillator: H_w e_k = (2|k|+n) e_k (Prop. 3.1 and (20)).
Provides the discrete eigenvalues µ_k that turn the sub-Laplacian symbol into |λ|µ_k and enable the mode-by-mode ODE analysis.
standard math Schrödinger representation formulas dπ_λ(X_j)=√|λ|∂_{w_j}, dπ_λ(Y_j)=i sign(λ)√|λ| w_j, dπ_λ(∆_H)=-|λ|H_w (Eqs. (13)-(15)).
Links the horizontal derivative operators to creation/annihilation on the Hermite basis, which is essential for the gradient estimates.
domain assumption The Cauchy problem (1) is well-posed in the stated spaces, so a solution u with the assumed regularity exists.
The theorem assumes u ∈ C([0,∞),H^1)∩C^1([0,∞),L^2); the paper does not prove well-posedness.
standard math The high-frequency factor e^{-t/2}|F|, e^{-t/2}|G| decays exponentially with a rate δ uniform in k and λ for |λ|>1/(8µ_k) (used in (27),(28),(38),(41),(42)).
Follows from the roots of τ^2+τ+µ_k|λ|=0 having real part -1/2 and the boundedness of the trigonometric/hyperbolic factors, but the uniformity is asserted, not shown in detail.
domain assumption The weighted Fourier coefficient sums in §4.3 are comparable to the Hilbert-Schmidt norms of the Fourier transforms of (X_j±iY_j)u0.
The paper states this as an exact identity, which is false; the true two-sided inequality is what the gradient estimates need.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Decay estimates for the linear damped wave equation on the Heisenberg group." pith.science (2026). https://pith.science/paper/SE6D3B73

@misc{pith2026190802657,
  author       = {Pith},
  title        = {Pith review of: Decay estimates for the linear damped wave equation on the Heisenberg group},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SE6D3B73}},
  note         = {Machine review of arXiv:1908.02657}
}

abstract

This paper is devoted to the derivation of $L^2$ - $L^2$ decay estimates for the solution of the homogeneous linear damped wave equation on the Heisenberg group $\mathbf{H}_n$, for its time derivative and for its horizontal gradient. Moreover, we consider the improvement of these estimates when further $L^1(\mathbf{H}_n)$ regularity is required for the Cauchy data. Our approach will rely strongly on the group Fourier transform of $\mathbf{H}_n$ and on the properties of the Hermite functions that form a maximal orthonormal system for $L^2(\mathbb{R}^n)$ of eigenfunctions of the harmonic oscillator.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

Critical exponent of Fujita-type for the semilinear damped wave equation on the Heisenberg group with power nonlinearity
math.AP 2019-08 conditional novelty 6.0 of 10

On the Heisenberg group, the semilinear damped wave equation with |u|^p has Fujita critical exponent p_Fuj(Q)=1+2/Q: global small-data solutions for p above it, blow-up for p at or below it.

Reference graph

Works this paper leans on

36 extracted references · 30 canonical work pages · cited by 1 Pith paper

[21]

Matsumura A, On the asymptotic behavior of solutions of semi-linear wave equations, Publ. Res. Inst. Math. Sci. 12(1) (1976/77), 169-189

1976
[1]

Bahouri H, G´ erard P, Xu C-J, Espaces de Besov et estimati ons de Strichartz g´ en´ eralis´ ees sur le groupe de Heisenberg, J. Anal. Math. 82 (2000), 93-118

work page 2000
[2]

Beals R, Gaveau B, Greiner PC, Hamilton-Jacobi theory an d the heat kernel on Heisenberg groups, J. Math. Pures Appl. 79 (2000), 633-689. 15

work page 2000
[3]

Berti M, Procesi M, Nonlinear wave and Schr¨ odinger equa tions on compact Lie groups and homogeneous spaces, Duke Math. J. 159(3) (2011), 479-538

work page 2011
[4]

Bruno T, Calzi M, Asymptotics for the heat kernel on H-typ e groups, Ann. Mat. Pura Appl. 197 (2018), 1017-1049

work page 2018
[5]

Part I: Basic theory and examples

Corwin L J, Greenleaf F P, Representations of nilpotent Lie groups and their applicat ions. Part I: Basic theory and examples. Volume 18 of Cambridge Studies in Advanced Mathematics. Cam bridge University Press, Cambridge, 1990

work page 1990
[6]

North-Holland Publishing Co., Amsterdam, 1977

Dixmier J, C∗ -algebras. North-Holland Publishing Co., Amsterdam, 1977. Translate d from the French by Francis Jellett, North-Holland Mathematical Library, Vol. 15

work page 1977
[7]

Volume 27 of North-Holland Mathematical Library

Dixmier J, von Neumann algebras. Volume 27 of North-Holland Mathematical Library. North-Ho lland Publishing Co., Amsterdam, 1981. With a preface by E. C. Lance, Translated fr om the second French edition by F. Jellett

work page 1981

Show all 36 references

[8]

Qualitative pr operties of solutions, phase space analysis, semilinear models, Birkh¨ auser, Basel, 2018

Ebert M R, Reissig M, Methods for partial diﬀerential equations. Qualitative pr operties of solutions, phase space analysis, semilinear models, Birkh¨ auser, Basel, 2018

2018
[9]

Fegan H D, The heat equation on a compact Lie group, Trans. Amer. Math. Soc. 246 (1978), 339-357

1978
[10]

Fegan H D, Introduction to Compact Lie Groups. Ser. Pure Math., vol. 13, W orld Scientiﬁc Publishing Co. Inc ., River Edge, NJ, 1991

1991
[11]

Math., vol

Fischer V, Ruzhansky M, Quantization on Nilpotent Lie Groups , Progr. Math., vol. 314, Birkh¨ auser/Springer, 2016

2016
[12]

Folland G B, Subelliptic estimates and function spaces on nilpotent Lie groups, Ark. Mat. 13(2) (1975), 161-207

1975
[13]

Studies in Advanced Mathematics

Folland G B, A course in abstract harmonic analysis. Studies in Advanced Mathematics. CRC Press, Boca Raton, FL, 1995

1995
[14]

Diﬀerential Equations 258 )(12) (2015), 4324-4347

Garetto C, Ruzhansky M, W ave equation for sums of square s on compact Lie groups, J. Diﬀerential Equations 258 )(12) (2015), 4324-4347

2015
[15]

139 (1977), 95-153

Gaveau B, Principe de moindre action, propagation de la chaleur et estimÃľes sous elliptiques sur certains groupes nilpotents, Acta Math. 139 (1977), 95-153

1977
[16]

Greiner P C, A Hamiltonian approach to the heat kernel of a sublaplacian on S2n+1, Anal. Appl. (Singap.) 11(6) (2013), 1350035, 62 pp

2013
[17]

Greiner P C, Holcman D, Kannai Y, W ave kernels related to second-order operators, Duke Math. J. 114(2) (2002), 329-386

2002
[18]

Greiner P, Li Y, Heat kernels, old and new, Bull. Inst. Math. Acad. Sin. (N.S.) 12 (2017), 1-37

2017
[19]

61(7) (2005), 1189-1208

Ikehata R, Tanizawa K, Global existence of solutions fo r semilinear damped wave equations in RN with noncompactly supported initial data, Nonlinear Anal. 61(7) (2005), 1189-1208

2005
[20]

Volume 64 of Graduate Studies in Mathematics

Kirillov A A, Lectures on the orbit method. Volume 64 of Graduate Studies in Mathematics. American Math ematical Society, Providence, RI, 2004

2004
[22]

PDE 8(5) (2015), 1051-1100

M¨ uller D, Seeger A, Lp bounds for the wave equation on groups of Heisenberg type, Anal. PDE 8(5) (2015), 1051-1100

2015
[23]

M¨ uller D, Stein E M, Lp-estimates for the wave equation on the Heisenberg group, Rev. Mat. Iberoamericana 15(2) (1999), 297-334

1999
[24]

Partial Diﬀerential Equations 7(6) (1982), 675-714

Nachman A I, The wave equation on the Heisenberg group, Comm. Partial Diﬀerential Equations 7(6) (1982), 675-714

1982
[25]

Pseudo Diﬀ

Nicola F, Rodino L, Global Pseudo-Diﬀerential Calculus on Euclidean Spaces . Pseudo Diﬀ. Oper., vol. 4, Birkh¨ auser Verlag, Basel, 2010

2010
[26]

Georgiev V, Palmieri A, Lifespan estimates for local in time solutions to the semilinear heat equation on the Heisen berg group, preprint, arxiv:1905.05696 (2019)

2019 arXiv
[27]

Georgiev V, Palmieri A, Upper bound estimates for local in time solutions to the semilinear heat equation on Carnot groups in the sub-Fujita case, preprint, (2019)

2019
[28]

Georgiev V, Palmieri A, Critical exponent of Fujita-ty pe for the semilinear damped wave equation on the Heisenberg group with power nonlinearity, preprint, (2019)

2019
[29]

Ruzhansky M, Taranto C, Time-dependent wave equations on graded groups preprint, arXiv:1705.03047v2 (2017)

2017 arXiv
[30]

Diﬀerential Equations 265 (2018), 5212-5236

Ruzhansky M, Tokmagambetov N, Nonlinear damped wave eq uations for the sub-Laplacian on the Heisenberg group and for Rockland operators on graded Lie groups J. Diﬀerential Equations 265 (2018), 5212-5236

2018
[31]

Ruzhansky M, Yessirkegenov N, Very weak solutions to hy poelliptic wave equations, preprint, arxiv:1810.11864 (2018)

2018 arXiv
[32]

Ruzhansky M, Yessirkegenov N, Existence and non-exist ence of global solutions for semilinear heat equations and i n- equalities on sub-Riemannian manifolds, and Fujita expone nt on unimodular Lie groups, preprint, arxiv:1812.01933v2 (2018)

2018 arXiv
[33]

Thesis, 2018, arXiv:1804.03544

Taranto C, Wave equations on graded groups and hypoelliptic Gevrey spa ces, Imperial College London Ph.D. Thesis, 2018, arXiv:1804.03544

2018 arXiv
[34]

Volume 22 of Mathematical Surveys and Monographs

Taylor M E, Noncommutative harmonic analysis. Volume 22 of Mathematical Surveys and Monographs. American Mathematical Society, Providence, RI, 1986

1986
[35]

Diﬀerential Equations 174(2) (2001), 464-489

Todorova G, Yordanov B, Critical exponent for a nonline ar wave equation with damping, J. Diﬀerential Equations 174(2) (2001), 464-489

2001
[36]

Cambridge University Press, Cambridge, 1992

Varopoulos N Th, Saloﬀ-Coste L, Coulhon T, Analysis and geometry on groups , Cambridge Tracts in Mathematics, 100. Cambridge University Press, Cambridge, 1992. 16

1992

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [21]

Matsumura A, On the asymptotic behavior of solutions of semi-linear wave equations, Publ. Res. Inst. Math. Sci. 12(1) (1976/77), 169-189

1976

[2] [1]

Bahouri H, G´ erard P, Xu C-J, Espaces de Besov et estimati ons de Strichartz g´ en´ eralis´ ees sur le groupe de Heisenberg, J. Anal. Math. 82 (2000), 93-118

work page 2000

[3] [2]

Beals R, Gaveau B, Greiner PC, Hamilton-Jacobi theory an d the heat kernel on Heisenberg groups, J. Math. Pures Appl. 79 (2000), 633-689. 15

work page 2000

[4] [3]

Berti M, Procesi M, Nonlinear wave and Schr¨ odinger equa tions on compact Lie groups and homogeneous spaces, Duke Math. J. 159(3) (2011), 479-538

work page 2011

[5] [4]

Bruno T, Calzi M, Asymptotics for the heat kernel on H-typ e groups, Ann. Mat. Pura Appl. 197 (2018), 1017-1049

work page 2018

[6] [5]

Part I: Basic theory and examples

Corwin L J, Greenleaf F P, Representations of nilpotent Lie groups and their applicat ions. Part I: Basic theory and examples. Volume 18 of Cambridge Studies in Advanced Mathematics. Cam bridge University Press, Cambridge, 1990

work page 1990

[7] [6]

North-Holland Publishing Co., Amsterdam, 1977

Dixmier J, C∗ -algebras. North-Holland Publishing Co., Amsterdam, 1977. Translate d from the French by Francis Jellett, North-Holland Mathematical Library, Vol. 15

work page 1977

[8] [7]

Volume 27 of North-Holland Mathematical Library

Dixmier J, von Neumann algebras. Volume 27 of North-Holland Mathematical Library. North-Ho lland Publishing Co., Amsterdam, 1981. With a preface by E. C. Lance, Translated fr om the second French edition by F. Jellett

work page 1981