AJMS_27.1._edb_proof 1..1A rab
Jo u
Volume 27 Issue 1 2021
Number 1
1 Editorial advisory board
2 On warped product bi-slant submanifolds of Kenmotsu manifolds
Siraj Uddin, Ion Mihai and Adela Mihai
15 On the number of edges of a graph and its complement Jamel
Dammak, Gerard Lopez and Hamza Si Kaddour
20 On Hilbert class fi eld tower for some quartic number fi elds
Abdelmalek Azizi, Mohamed Talbi and Mohammed Talbi
26 A Method of approximation for a zero of the sum of maximally
monotone mappings in Hilbert spaces Getahun Bekele Wega and Habtu
Zegeye
41 Reconstruction of a homogeneous polynomial from its additive
decompositions when identifi ability fails E. Ballico
53 Warped product pointwise semi-slant submanifolds of cosymplectic
space forms and their applications Lamia Saeed Alqahtani
73 Toeplitz and slant Toeplitz operators on the polydisk Munmun
Hazarika and Sougata Marik
94 Structure and substructure connectivity of circulant graphs and
hypercubes T. Tamizh Chelvam and M. Sivagami
104 Note on a W1, ∞ (L2)-error estimate of a nonlinear fi nite
volume scheme for a semi-linear heat equation Ahmed Berkane and
Abdallah Bradji
119 The Q 0 -matrix completion problem
Kalyan Sinha
Quarto trim size: 174mm x 240mm
The Arab Journal of Mathematical Sciences is the official science
journal of the Saudi Association for Mathematical Sciences. It is
dedicated to the publication of original and expository papers in
pure and applied mathematics, and is reviewed and edited by an
international group of scholars. The Arab Journal of Mathematical
Sciences will accept submissions in the mainstream areas of pure
and applied mathematics, including algebra, analysis, geometry,
diff erential equations, and discrete mathematics.
EDITOR Dr Tariq Alfhadel Department of Mathematics, King Saud
University, Saudi Arabia Homepage:
https://www.emeraldgrouppublishing.com/ journal/ajms
e-ISSN 2588-9214 p-ISSN 1319-5166 © King Saud University
Guidelines for authors can be found at:
https://www.emeraldgrouppublishing.com/journal/
ajms#author-guidelines
Emerald Publishing Limited Howard House, Wagon Lane, Bingley BD16
1WA, United Kingdom Tel +44 (0) 1274 777700; Fax +44 (0) 1274
785201 E-mail
[email protected] For more information about
Emerald’s regional offi ces please go to
http://www.emeraldgrouppublishing.com/offi ces Customer helpdesk:
Tel +44 (0) 1274 785278; Fax +44 (0) 1274 785201 E-mail
[email protected] The Publisher and Editors cannot be held
responsible for errors or any consequences arising from the use of
information contained in this journal; the views and opinions
expressed do not necessarily refl ect those of the Publisher and
Editors, neither does the publication of advertisements constitute
any endorsement by the Publisher and Editors of the products
advertised.
Emerald is a trading name of Emerald Publishing Limited
EDITORIAL BOARD
Professor Sharief Deshmukh King Saud University, Saudi Arabia
[email protected]
Professor Alberto Ferrero Universita del Piedmonte Oriental, Italy
[email protected]
Professor Filippo Gazzola Politecnico di Milano, Italy
[email protected]
Professor Mohammed Guediri King Saud University, Saudi Arabia
[email protected]
Professor Mourad Ismail University of Central Florida, USA
[email protected]
Dr Wissem Jedidi King Saud University, Saudi Arabia
[email protected]
Professor Abdellatif Laradji King Saud University, Saudi Arabia
[email protected]
Professor Elisabetta Maluta Politecnico di Milano, Italy
[email protected]
Professor Nabil Ourimi King Saud University, Saudi Arabia
[email protected]
Professor Vicentiu Radulescu University of Craiova, Romania
[email protected]
Professor Bassem Samet King Saud University, Saudi Arabia
[email protected]
EDITORIAL ADVISORY BOARD
Professor M. Al-Gwaiz King Saud University, Saudi Arabia
Professor V. Anandam Institute of Mathematical Science, India
[email protected]
Professor Vieri Benci University of Pisa, Italy
[email protected]
Professor Christian Berg University of Copenhagen, Denmark
[email protected]
Professor Bang-Yen Chen Michigan State University, USA
[email protected]
Professor S. Hedayat University of Illinois, USA
[email protected]
Professor S. Kabbaj King Fahad University, Saudi Arabia
[email protected]
Professor Mokhtar Kirane Universite de la Rochelle, France
[email protected]
Professor Zuhair Nashed University of Central Florida, USA
[email protected]
Professor Maurice Pouzet Universite Claude-Bernard, France
[email protected]
Professor David Yost Federation University, Australia
[email protected]
Editorial boards
Vol. 27 No. 1, 2021 p. 1
Emerald Publishing Limited e-ISSN: 2588-9214 p-ISSN:
1319-5166
Quarto trim size: 174mm x 240mm
On warped product bi-slant submanifolds of
Kenmotsu manifolds Siraj Uddin
Department of Mathematics, Faculty of Science, King Abdulaziz
University, Jeddah, Saudi Arabia
Ion Mihai Faculty of Mathematics and Computer Science, University
of Bucharest,
Bucharest, Romania, and
Adela Mihai Department of Mathematics and Computer Science,
Technical University of Civil Engineering Bucharest, Bucharest,
Romania
Abstract Chen (2001) initiated the study of CR-warped product
submanifolds in Kaehler manifolds and established a general
inequality between an intrinsic invariant (the warping function)
and an extrinsic invariant (second fundamental form).
In this paper, we establish a relationship for the squared norm of
the second fundamental form (an extrinsic invariant) of warped
product bi-slant submanifolds of Kenmotsu manifolds in terms of the
warping function (an intrinsic invariant) and bi-slant angles. The
equality case is also considered. Some applications of derived
inequality are given.
Keywords Warped products, Semi-slant, Pseudo-slant, Bi-slant
submanifolds, Warped product bi-slant
submanifolds, Kenmotsu manifolds
Paper type Original Article
1. Introduction In [18], K. Kenmotsu studied one class of almost
contact metric manifolds known as Kenmotsu manifolds. He proved
that:
AJMS 27,1
2
JEL Classification — 53C15, 53C40, 53C42, 53B25 © Siraj Uddin, Ion
Mihai and Adela Mihai. Published in the Arab Journal of
Mathematical Sciences.
Published by Emerald Publishing Limited. This article is published
under the Creative Commons Attribution (CCBY4.0) license. Anyonemay
reproduce, distribute, translate and create derivativeworks of this
article (for both commercial and non-commercial purposes), subject
to full attribution to the original publication and authors. The
full terms of this license may be seen at http://creativecommons.
org/licences/by/4.0/legalcode
Declaration of Competing Interest: The authors declare that they
have no known competing financial interests or personal
relationships that could have appeared to influence the work
reported in this paper.
The publisher wishes to inform readers that the article “Onwarped
product bi-slant submanifolds of Kenmotsu manifolds” was originally
published by the previous publisher of the Arab Journal of
Mathematical Sciences and the pagination of this article has been
subsequently changed. There has been no change to the content of
the article. This change was necessary for the journal to
transition from the previous publisher to the new one. The
publisher sincerely apologises for any inconvenience caused. To
access and cite this article, please use Uddin, S., Mihai, I.,
Mihai, A. (2019), “On warped product bi-slant submanifolds of
Kenmotsu manifolds”, Arab Journal of Mathematical Sciences, Vol. 27
No. 1, pp. 2-14. The original publication date for this paper was
05/07/2019.
The current issue and full text archive of this journal is
available on Emerald Insight at:
https://www.emerald.com/insight/1319-5166.htm
Arab Journal of Mathematical Sciences Vol. 27 No. 1, 2021 pp. 2-14
Emerald Publishing Limited e-ISSN: 2588-9214 p-ISSN: 1319-5166 DOI
10.1016/j.ajmsc.2019.06.001
1. Locally a Kenmotsu manifold is a warped product I 3 f M of an
interval I and a Kaehler manifold M, with warping function f ¼ cet,
where c is a nonzero constant.
2. A Kenmotsu manifold with constant sectional curvature is a space
of constant curvature −1, and so it is locally a hyperbolic
space.
A ð2mþ 1Þ-dimensional manifold ~M is said to be an almost contact
manifold if it admits an endomorphism w of its tangent bundle T ~M,
a vector field ξ and a 1-form η, which satisfy:
w2 ¼ −I þ η⊗ ξ;wξ ¼ 0; ηðξÞ ¼ 0; η+w ¼ 0: (1.1)
There exists a compatible metric g, which satisfies
gðwX ;wY Þ ¼ gðX ;Y Þ ηðXÞηðY Þ; ηðXÞ ¼ gðX ; ξÞ; (1.2)
for all vector fields X ;Y on ~M [6]. In addition, an almost
contact metric manifold ~M is said to be a Kenmotsu manifold [18]
if the relation
ð~∇XwÞY ¼ gðwX ;Y Þξ ηðY ÞwX (1.3)
holds, where ~∇ is the Levi-Civita connection of g. From (1.3), for
a Kenmotsu manifold ~M, we also have
~∇Xξ ¼ X ηðXÞξ: (1.4)
As Kenmotsu manifolds are warped product manifolds, therefore it is
interesting to investigate the geometry of its warped product
submanifolds. The notion of warped submanifolds was first
introduced by B.-Y. Chen as a CR-warped product submanifold of
Kaehler manifolds in his series of articles [11,12]. He established
a general sharp inequality between the main extrinsic invariant
(the second fundamental form) and an intrinsic invariant (the
warping function) of such submanifolds. Motivated by Chen’s work
many geometers studied warped product submanifolds for different
spaces (see, e.g., [4,17,19–25,29,30] among many others. For the
most up-to-date overview of this subject, see [13–15]).
On the other hand, J.L. Cabrerizro et al. studied in [7] bi-slant
submanifolds of almost contact metric manifolds. In [28], the first
author and B.-Y. Chen investigated warped product bi-slant
submanifolds in Kaehler manifolds. They proved that there do not
exist any warped product bi-slant submanifolds of Kaehler manifolds
other than hemi-slant warped products and CR-warped products. The
non-existence of warped product bi-slant submanifolds is proved in
[2] for cosymplectic manifolds.
In this paper, we study warped product bi-slant submanifolds of a
Kenmotsu manifold. The geometry of such submanifolds in Kenmotsu
manifolds is quite different from Sasakian and cosymplectic case
because in case of Kenmotsu manifolds such submanifolds exist while
there is no proper warped product bi-slant submanifolds in Sasakain
and cosymplectic as well. On their existence, we establish a
generalized Chen type sharp inequality for the squared norm of the
second fundamental form in terms of the warping function and
bi-slant angles. The equality case is considered. Some applications
are given in the last section.
2. Preliminaries Let ψ : Mn
→M ðnþdÞ be an isometric immersion of an n-dimensional Riemannian
manifold M into an ðnþ dÞ-dimensional Riemannian manifold ~M. We
denote by ∇ and ~∇ the Levi- Civita connections onM and ~M,
respectively. Then the Gauss and Weingarten formulas are
respectively given by [31]
~∇XY ¼ ∇XY þ hðX ;Y Þ (2.1)
Submanifolds of Kenmotsu
XV ; (2.2)
for any X ;Y ∈TM and V ∈T⊥M, where ∇⊥ is the normal connection in
the normal bundle T⊥M andAV is the shape operator ofM with respect
toV . Moreover, h : TM 3TM →T⊥M is the second fundamental form of M
in ~M. Furthermore, AV and h are related by
gðhðX ;Y Þ;V Þ ¼ gðAV X ;Y Þ; (2.3)
for any X ;Y ∈TM and V ∈T⊥M. A submanifold M of a
Riemannianmanifold ~M is said to be a totally umbilical
submanifold
if hðX ;Y Þ ¼ gðX ;Y ÞH, for any X ;Y ∈TM, where H ¼ 1 n
Pn i¼1 hðei; eiÞ is the mean
curvature vector ofM. A submanifold M is said to be totally
geodesic if hðX ;Y Þ ¼ 0. Also, one denotes
khk2 ¼ Xn
gðhðei; ejÞ; hðei; ejÞÞ; hrij ¼ gðhðei; ejÞ; erÞ; (2.4)
with i; j ¼ 1; . . . ; n; r ¼ nþ 1; . . . ; nþ d, where fe1; . . .
; eng is an orthonormal basis of the tangent space TpM and fenþ1; .
. . ; enþdg an orthonormal basis of the normal space T⊥
p , for any p∈M.
For a differentiable function f on a m-dimensional manifold ~M, the
gradient ∇ ! f of f is
defined as gð∇!f ;XÞ ¼ Xðf Þ, for any X tangent to ~M. For any
vector field X tangent to M, we write
wX ¼ TX þ FX ; (2.5)
whereTX is the tangential component and FX is the normal component
of wX. Thus,T is an endomorphism on the tangent bundle TM and F is
a normal bundle valued 1-form of TM. A submanifoldM is called
invariant if F is identically zero, that is, wX ∈TM for any X ∈TM;
while, M is anti-invariant if T is identically zero, that is, wX
∈T⊥M, for any X ∈TM.
Similarly, for any vector field V normal to M, we put
wV ¼ tV þ fV ; (2.6)
where tV and fV are the tangential and normal components of wV ,
respectively.
LetM be a submanifold of an almost contact metric manifold ð ~M ;w;
ξ; η; gÞ. Hence, if we denote D the orthogonal distribution to ξ in
TM, then TM ¼ D⊕ hξi, where hξi is the 1-dimensional distribution
spanned by ξ. For any nonzero vector X tangent toM at the point
p∈M, such that X is not proportional to ξp, we denote by θðXÞ, the
angle between wX and TpM. In fact, sincewξ ¼ 0, θðXÞagrees with the
angle betweenwX andDp. A submanifoldM
of an almost contact metric manifold ~M is said to be slant [8], if
for any non-zero vector X tangent toM at p such thatX is not
proportional to ξp, the angle θðXÞbetweenwX andTpM is constant,
i.e., it does not depend on the choice of p∈M andX ∈TpM − hξpi. In
this caseD is a slant distribution with slant angle θ.
A slant submanifold is said to be proper slant, if neither θ ¼ 0
nor θ ¼ π 2 [9,10]. We note
that on a slant submanifold if θ ¼ 0, then it is an invariant
submanifold and if θ ¼ π 2, then it is
an anti-invariant submanifold. A slant submanifold is said to be
proper slant if it is neither invariant nor anti-invariant.
A characterization of slant submanifolds was given in [8] as
follows:
Theorem 1 ([8]). Let M be a submanifold of an almost contact metric
manifold ~M such that ξ∈TM. Then M is slant if and only if there
exists a constant λ∈ ½0; 1 such that
AJMS 27,1
T2 ¼ λð−I þ η⊗ ξÞ: (2.7)
Furthermore, in such case, if θ is slant angle, then λ ¼ cos2 θ
.
The following relations are straightforward consequences of
(2.7)
gðTX ;TY Þ ¼ cos2 θ½gðX ;Y Þ ηðXÞηðY Þ; (2.8)
gðFX ;FY Þ ¼ sin2 θ½gðX ;Y Þ ηðXÞηðY Þ; (2.9)
for any X ;Y tangent to M. The following useful relation is
obtained as a consequence of (2.7) in [26].
Theorem2 ([26]). LetM be a proper slant submanifold of an almost
contact metric manifold ~M. Then
ðaÞ tFX ¼ sin2 θð−X þ ηðXÞξÞ; ðbÞ fFX ¼ −FTX ; (2.10)
for any X ∈TM.
Another characterization of slant submanifolds was given in
[7]:
Theorem3 ([7]). LetD be a distribution onM , orthogonal to ξ .
Then,D is slant if and only if there exists a constant λ∈ ½0; 1
such that ðPTÞ2 X ¼ −λX , for any X ∈Dp at p∈M , where P denotes
the orthogonal projection on D. Furthermore, in this case, λ ¼
cos2θD.
J.L. Cabrerizo et al. [7] defined bi-slant submanifolds as
follows:
Definition1. AsubmanifoldM of an almost contactmetricmanifold ~M is
said to be bi-slant if there exists a pair of orthogonal
distributions D1 and D2 on M such that:
(i) The tangent bundle TM admits the orthogonal direct
decomposition: TM ¼ D1 ⊕D2 ⊕ hξi.
(ii) Each Di ∀i ¼ 1; 2 is a slant distribution with slant angle
θi.
Given a bi-slant submanifold M, for any X ∈TM, we put
X ¼ P1X þ P2X þ ηðXÞξ; (2.11)
where PiX denotes the component ofX inDi, for any i ¼ 1; 2. In
particular, ifX ∈Di, then we obtain X ¼ PiX. If we define Ti ¼
Pi+T, then we have
wX ¼ T1X þ T2X þ FX ; (2.12)
for any X ∈TM. Given i ¼ 1; 2, from Theorem 3, we get
T2 i X ¼ −cos2θiX ; (2.13)
for any X ∈Di. Non-trivial examples of bi-slant submanifolds are
given in [7].
3. Warped product bi-slant submanifolds LetM1 3 f M2 be a warped
product manifold of two RiemannianmanifoldsM1 andM2. Then from a
result of [5], we have
∇XZ ¼ ∇Z X ¼ 1
f ðXf ÞZ (3.1)
for any vector fields X ; Z tangent to M1;M2, respectively.
Submanifolds of Kenmotsu
manifolds
5
Recently, B.-Y. Chen and the first author introduced the notion of
warped product bi-slant submanifolds of Kaehler manifolds [28].
They proved the non-existence of proper warped product bi-slant
submanifolds. Then, they introduced the notion of warped product
pointwise bi-slant submanifolds and obtained several fundamental
results [16]. In this section, we give some useful lemmas for
warped product bi-slant submanifolds of Kenmotsu manifolds. First
we define these submanifolds as follows:
A warped productM1 3 f M2 of two slant submanifoldsM1 andM2 with
slant angles θ1 and θ2 of a Kenmotsu manifold ~M is called a warped
product bi-slant submanifold.
A warped product bi-slant submanifold M ¼ M1 3 f M2 is called
proper if both M1 and M2 are proper slant submanifolds with slant
angles θ1; θ2 ≠ 0; π2 of
~M. A warped product bi- slant submanifold M1 3 f M2 is a contact
CR-warped product if θ1 ¼ 0; θ2 ¼ π
2 or θ2 ¼ 0; θ1 ¼ π
2 ; such submanifolds were discussed in [3,27]. Also, it is a
warped product pseudo-slant submanifold if θ1 ¼ θ and θ2 ¼ π
2 [1] or θ2 ¼ θ and θ1 ¼ π 2 [23]. The warped
product bi-slant submanifolds with slant angles θ1 ¼ 0; θ2 ¼ θ or
θ2 ¼ 0; θ1 ¼ θ were discussed in [25,30].
LetM ¼ M1 3 f M2 be awarped product bi-slant submanifold of a
Kenmotsumanifold ~M such that the structure vector field ξ is
tangent to M, where M1 and M2 are proper slant submanifolds of ~M.
Then, we distinguish 2 cases:
(i) ξ is tangent to M1;
(ii) ξ is tangent to M2.
From (1.4), (2.1) and (3.1), the second case is trivial i.e., there
does not exist any proper warped product bi-slant submanifold of a
Kenmotsu manifold when the structure vector field is tangent to the
fiber.
Now, we start with the case (i). Throughout this paper, we assume
that the tangent spaces ofM1 andM2, respectively areD1 andD2. From
now on, we use the following conventions: X1;Y1 are vector fields
in D1 and X2;Y2 are vector fields in D2.
Lemma 1. Let M ¼ M1 3 f M2 be a warped product bi-slant submanifold
of a Kenmotsu manifold ~M such that ξ∈D1 . Then, we have
(i) ξðln f Þ ¼ 1;
(ii) gðhðX1;Y1Þ; FX2Þ ¼ gðhðX1;X2Þ;FY1Þ;
for any X1;Y1 ∈D1 and X2 ∈D2 .
Proof. First part is trivial and can be obtained by using (1.4),
(2.1) and (3.1). For the second part, we have
gð~∇X1 Y1;X2Þ ¼ −gð∇X1
X2;Y1Þ ¼ −X1ðln f Þ g ðY1;X2Þ ¼ 0; (3.2)
for anyX1;Y1 ∈D1 andX2 ∈D2. Also, from (1.2) and the fact that ξ is
tangent toM1, we have
gð~∇X1 Y1;X2Þ ¼ gðw~∇X1
Y1;wX2Þ ¼ gð~∇X1 wY1;wX2Þ g
ð~∇X1 wÞY1;wX2Þ: (3.3)
wY1;wX2Þ ¼ gð~∇X1 T1Y1;T2X2Þ
þ gð~∇X1 T1Y1;FX2Þ þ gð~∇X1
FY1;wX2Þ: Then using the covariant derivative property of
Riemannian connection and (1.2), (1.3), (2.1) and (3.1), we
derive
AJMS 27,1
6
ηðY1Þ gðT1X1;T2X2Þ ¼ −X1ðln f Þ gðT1Y1;T2X2Þ þ gðhðX1;T1Y1Þ;FX2Þ þ
gðð~∇X1
wÞFY1;X2Þ gð~∇X1 wFY1;X2Þ:
By the orthogonality of vector fields, the left hand side and the
first term in the right hand side vanish identically. Then using
(1.3), (2.6) and (2.10), we find
0 ¼ gðhðX1;T1Y1Þ;FX2Þ þ sin2 θ1 g ð~∇X1 Y1;X2Þ sin2 θ1 g
ð~∇X1
ξ;X2Þ þ gð~∇X1
FT1Y1;X2Þ: Using (1.4), (2.2), (2.3), (3.1) and the orthogonality
of vector fields, we derive
gðhðX1;X2Þ;FT1Y1Þ ¼ gðhðX1;T1Y1Þ;FX2Þ: (3.4)
Interchanging Y1 by T1Y1 in (3.4), we obtain
cos2 θ1 gðhðX1;X2Þ;FY1Þ ¼ cos2 θ1gðhðX1;Y1Þ;FX2Þ cos2 θ1ηðY1Þ g
ðhðX1; ξÞ;X2Þ:
(3.5)
Since for a submanifold M of a Kenmotsu manifold ~M, hðX ; ξÞ ¼ 0;
∀X ∈TM, then the second part of the lemma follows from above
relation. Hence, the proof is complete. -
Lemma 2. Let M ¼ M1 3 f M2 be a warped product bi-slant submanifold
of a Kenmotsu manifold ~M such that ξ∈D1 , whereM1 andM2 are proper
slant submanifolds of ~M with slant angles θ1 and θ2 ,
respectively. Then, we have
gðhðX2;Y2Þ;FT1X1Þ gðhðX2;T1X1Þ;FY2Þ ¼ T1X1ðln f Þ gðX2;T2Y2Þ cos2
θ1ðX1ðln f Þ ηðX1ÞÞ gðX2;Y2Þ
(3.6)
Proof. For any X1 ∈D1 and X2;Y2 ∈D2, we have
gð~∇X2 X1;Y2Þ ¼ X1ðln f Þ gðX2;Y2Þ: (3.7)
On the other hand, we also have
gð~∇X2 X1;Y2Þ ¼ gðw~∇X2
X1;wY2Þ ¼ gð~∇X2 wX1;wY2Þ g
ð~∇X2 wÞX1;wY2Þ:
gð~∇X2 X1;Y2Þ ¼ gð~∇X2
T1X1;T2Y2Þ þ gð~∇X2 T1X1;FY2Þ
þ gð~∇X2 FX1;wY2Þ þ ηðX1Þ g ðwX2;wY2Þ:
From (1.2), (2.1) and (3.1), the above equation takes the
form
gð~∇X2 X1;Y2Þ ¼ T1X1ðln f ÞgðX2;T2Y2Þ þ gðhðX2;T1X1Þ;FY2Þ
gð~∇X2
wFX1;Y2Þ þ gðð~∇X2
wÞFX1;Y2Þ þ ηðX1Þ g ðX2;Y2Þ: Again, using (1.3) and (2.6), we
derive
gð~∇X2 X1;Y2Þ ¼ T1X1ðln f Þ g ðX2;T2Y2Þ þ gðhðX2;T1X1Þ;FY2Þ
gð~∇X2
t FX1;Y2Þ gð~∇X2
Submanifolds of Kenmotsu
gð~∇X2 X1;Y2Þ ¼ T1X1ðln f Þ gðX2;T2Y2Þ þ gðhðX2;T1X1Þ;FY2Þ
þ sin2 θ1 X1ðln f Þ gðX2;Y2Þ sin2 θ1 ηðX1Þ gðX2;Y2Þ þ gð~∇X2
FT1X1;Y2Þ þ ηðX1Þ gðX2;Y2Þ ¼ T1X1ðln f Þ gðX2;T2Y2Þ þ
gðhðX2;T1X1Þ;FY2Þ þ sin2 θ1X1ðln f Þ gðX2;Y2Þ þ cos2 θ1ηðX1Þ
gðX2;Y2Þ gðhðX2;Y2Þ;FT1X1Þ:
(3.8)
Thus, the result follows from (3.7) and (3.8), which proves the
lemma completely. -
The following useful relations are easily derived by
interchangingX1 byT1X1,X2 byT2X2
and Y2 by T2Y2 in Lemma 2.
gðhðX2;Y2Þ;FX1Þ gðhðX1;X2Þ;FY2Þ ¼ T1X1ðln f Þ gðX2;Y2Þ ðX1ðln f Þ
ηðX1ÞÞ gðT2X2;Y2Þ;
(3.9)
gðhðT2X2;Y2Þ;FT1X1Þ gðhðT2X2;T1X1Þ;FY2Þ ¼ cos2 θ2T1X1ðln f Þ
gðX2;Y2Þ cos2 θ1 ðX1ðln f Þ ηðX1ÞÞ gðT2X2;Y2Þ;
(3.10)
gðhðX2;T2Y2Þ;FT1X1Þ gðhðX2;T1X1Þ;FT2Y2Þ ¼ − cos2 θ2T1X1ðln f Þ
gðX2;Y2Þ þ cos2 θ1ðX1ðln f Þ ηðX1ÞÞ gðT2X2;Y2Þ;
(3.11)
cos2 θ1 cos 2 θ2 ðX1ðln f Þ
ηðX1ÞÞ gðX2;Y2Þ;
(3.12)
gðhðT2X2;Y2Þ;FX1Þ gðhðT2X2;X1Þ;FY2Þ ¼ T1X1ðln f Þ gðT2X2;Y2Þ þ cos2
θ2 ðX1ðln f Þ ηðX1ÞÞ gðX2;Y2Þ;
(3.13)
gðhðX2;T2Y2Þ;FX1Þ gðhðX2;X1Þ;FT2Y2Þ ¼ T1X1ðln f Þ gðX2;T2Y2Þ cos2
θ2 ðX1ðln f Þ ηðX1ÞÞ gðX2;Y2Þ
(3.14)
and gðhðT2X2;T2Y2Þ;FX1Þ gðhðT2X2;X1Þ;FT2Y2Þ ¼ cos2 θ2T1X1ðln f Þ g
ðX2;Y2Þ
cos2 θ2 ðX1ðln f Þ ηðX1ÞÞ gðT2X2;Y2Þ:
(3.15)
8
4. An inequality for warped product bi-slant submanifolds LetM ¼ M1
3 f M2 be a warped product bi-slant submanifolds of a Kenmotsu
manifold ~M; we decompose the normal bundle of M as follows
T⊥M ¼ FD1 ⊕FD2 ⊕ μ; μ ⊥ FD1 ⊕FD2; (4.1)
where μ is a w-invariant normal subbundle of T⊥M. A warped product
bi-slant submanifold M ¼ M1 3 f M2 of a Kenmotsu manifold ~M
is
said to bemixed totally geodesic, if hðX ; ZÞ ¼ 0, for anyX ∈D1 and
Z ∈D2, whereD1 andD2
are the tangent bundles of M1 and M2, respectively. Now, we set the
following frame fields for an n-dimensional warped product
bi-slant
submanifoldM ¼ M1 3 f M2 of a ð2mþ 1Þ-dimensional Kenmotsumanifold
~M such that ξ is
tangent toM1, whereM1 andM2 are proper slant submanifolds of ~M
with slant angles θ1 and θ2 respectively. Let us consider the
dimensions dimðM1Þ ¼ 2pþ 1 and dimM2 ¼ 2q, i.e., n ¼ 2pþ 1þ 2q.
Then the orthonormal frames of the corresponding tangent
spacesD1
andD2, respectively are given by fe1; . . . ; ep; epþ1 ¼ sec
θ1T1e1; . . . ; e2p ¼ sec θ1T1ep; e2pþ1 ¼ ξg and fe2pþ2 ¼ e1; . . .
; e2pþ1þq ¼ eq; e2pþqþ2 ¼ eqþ1 ¼ sec θ2T2e1; . . . ; en ¼ e2q ¼ sec
θ2T2eqg. Thus, the orthonormal frame fields of the normal
subbundles of FD1;FD2 and μ, respectively are fenþ1 ¼ ~e1 ¼ csc
θ1Fe1; . . . ; enþp ¼ ~ep ¼ csc θ1Fep; enþpþ1 ¼ ~epþ1 ¼ csc θ1 sec
θ1FT1e1; . . . ; enþ2p ¼ ~e2p ¼ csc θ1 sec θ1FT1epg; fenþ2pþ1 ¼
~e2pþ1 ¼ be1 ¼ csc θ2 Fe1; . . . ; enþ2pþq ¼ ~e2pþq ¼ beq ¼ csc θ2
Feq; enþ2pþqþ1 ¼ ~e2pþqþ1 ¼ beqþ1 ¼ csc θ2 sec θ2 FT2e1; . . . ;
enþ2pþ2q ¼ ~e2pþ2q ¼ be2q ¼ csc θ2 sec θ2 FT2eqg and fe2n ¼ ~en; .
. . ; e2mþ1 ¼ ~e2ðm−nþ1Þg.
Now, using the results of Section 3 and the above frame fields, we
give following main result of this paper.
Theorem 4. Let M ¼ M1 3 f M2 be a mixed totally geodesic warped
product bi-slant submanifold of a Kenmotsu manifold ~M such that ξ
is tangent to M1 , where M1 and M2 are proper slant submanifolds of
~M with slant angles θ1 and θ2 , respectively. Then
(i) The second fundamental form h of M satisfies the following
inequality
khk2 ≥ 2q csc2 θ1 cos2 θ1 þ cos2 θ2
k∇!ðln f Þk2 1 ; (4.2)
where 2q ¼ dimM2 and ∇ !ðln f Þ is the gradient of ln f along
M1.
(ii) If equality sign in (i) holds identically, then:
(a) M1 is a totally geodesic submanifold of ~M;
(b) M2 is a totally umbilical submanifold of ~M.
Proof. From (2.4), we have
khk2 ¼ Xn
r¼nþ1
gðhðei; ejÞ; erÞ2:
Splitting the above expression for the tangent bundles of M1 and
M2, we derive
khk2 ¼ X2mþ1
r¼1
r¼1
gðhðei; ejÞ; erÞ2: (4.3)
manifolds
9
Since M is mixed totally geodesic, then the second term in the
right hand side of (4.3) is identically zero. Thus, we find
khk2 ¼ X2p r¼1
X2pþ1
X2pþ1
gðhðei; ejÞ;~erÞ2 þ X2q r¼1
X2q i; j¼1
gðhðei; ejÞ;berÞ2
gðhðei; ejÞ;~erÞ2:
(4.4)
The third and sixth terms have μ-components and we could not find
any relation for warped products in terms of μ-components,
therefore we shall leave these two terms. Also, we could not find
any relations for gðhðei; ejÞ;~erÞ; i; j ¼ 1; . . . ; 2pþ 1; r ¼ 1;
. . . ; 2p and gðhðei; ejÞ;berÞ; i; j; r ¼ 1; . . . ; 2q.
Therefore, we also leave the first and fifth terms. By using Lemma
1(ii) for a mixed totally geodesic warped product the second term
in the right hand side is also zero. Thus, the evaluated term is
only fourth term which can be expressed by using the constructed
frame fields as follows
khk2 ≥ csc2 θ1 sec 2 θ2
Xp
Xp
2 θ2 Xp
r¼1
2 θ2 Xp
r¼1
Xp
Xp
4 θ2 Xp
r¼1
gðhðT2ei;T2ejÞ;FT1erÞ2:
(4.5)
Using Lemma 2 and the relations (3.9)–(3.15) for a mixed totally
geodesic warped product submanifold, we derive
AJMS 27,1
Xp
2 θ2 Xp
r¼1
ðT1erðln f ÞÞ2
r¼1
ðT1erðln f ÞÞ2 þ 2q csc2 θ1 cos 2 θ1
Xp
Xp
2 θ1 Xp
r¼1
ðT1erðln f ÞÞ2:
(4.6)
Since ηðerÞ ¼ 0; r ¼ 1; . . . ; 2p, then above expression will
be
khk2 ≥ 2q csc2 θ1 cos2 θ1 þ cos2 θ2
Xp
2 θ1 X2pþ1
r¼1
ðT1erðln f ÞÞ2
2 θ1 X2p r¼pþ1
gðer;T1∇ !ðln f ÞÞ2
2 θ1 ðT1e2pþ1ðln f ÞÞ2:
(4.7)
Since T1e2pþ1 ¼ T1ξ ¼ 0, (4.7) can be written as
khk2 ≥ 2q csc2 θ1 cos2 θ1 þ cos2 θ2
Xp
2 θ1 kT1∇
!ðln f Þk2
1þ cos2 θ2 sec
kT1∇ !ðln f Þk2 ¼ gðT1∇
!ðln f Þ;T1∇ !ðln f ÞÞ:
Using (2.8), we get
kT1∇ !ðln f Þk2 ¼ cos2 θ1
:
kT1∇ !ðln f Þk2 ¼ cos2 θ1
k∇!ðln f Þk2 1 : (4.9)
Submanifolds of Kenmotsu
khk2 ≥ 2q csc2 θ1 cos2 θ1 þ cos2 θ2
Xp
k∇!ðln f Þk2 1
2q csc2 θ1 1þ cos2 θ2 sec
2 θ1 csc2 θ1
Xp
k∇!ðln f Þk2 1
2q csc2 θ1 cos2 θ1 þ cos2 θ2
Xp
k∇!ðln f Þk2 1 ;
which is inequality (4.2). For the equality case, since M is mixed
totally geodesic, then
hðD1;D2Þ ¼ f0g: (4.10)
From the leaving third and sixth terms in the right hand side of
(4.4), we have
hðX ;Y Þ ⊥ μ; (4.11)
for any X ;Y ∈TM. Also, from the leaving first term of (4.4), we
find
hðD1;D1Þ ⊥ FD1: (4.12)
hðD1;D1Þ ⊂ FD2: (4.13)
But for a mixed totally geodesic submanifold, from Lemma 1(ii), we
get
hðD1;D1Þ ⊥ FD2: (4.14)
hðD1;D1Þ ¼ f0g: (4.15)
SinceM1 is totally geodesic inM [5,11], then using this fact with
(4.10) and (4.15) we conclude
that M1 is a totally geodesic submanifold of ~M, which is the first
relation of inequality. Similarly, from the leaving fifth term in
the right hand side of (4.4), we get
hðD2;D2Þ ⊥ FD2: (4.16)
hðD2;D2Þ ⊂ FD1: (4.17)
SinceM2 is totally umbilical inM [5,11], using this fact with
(4.10) and (4.17), we conclude that
M2 is a totally umbilical submanifold of ~M, which proves the
statement ðbÞ. Hence, the proof is complete. -
AJMS 27,1
12
5. Some applications of derived inequality In this section, we give
some applications of the derived inequality (4.2).
1. In Theorem 4, if θ1 ¼ π 2 and θ2 ¼ 0, then the warped product
bi-slantM1 3 f M2 takes
the formM⊥3 f MT i.e., M is a contact CR-warped product studied in
[27]. In this case, the inequality (4.2) will be khk2 ≥ 2qðk∇!ðln f
Þk2 − 1Þ, which is the main Theorem 3.4 of [27].
2. If we consider θ1 ¼ θ and θ2 ¼ 0 in a warped product bi-slant
submanifold M ¼ M1 3 f M2, then M is a warped product semi-slant
submanifold of the form Mθ 3 f MT studied in [30]. Then, the
inequality (4.2) change into khk2 ≥ 2qðcsc2 θþ cot2 θÞðk∇!ðln f Þk2
− 1Þ, which is Theorem 4.2 of [30]. Thus, Theorem 4.2 of [30] is a
special
case of Theorem 4.
3. Also, if we consider θ1 ¼ θ and θ2 ¼ π 2 in a warped product
bi-slant submanifold
M ¼ M1 3 f M2 of a Kenmotsu manifold ~M, thenM turns into a warped
product pseudo- slant submanifold of the formM ¼ Mθ 3 f M⊥, whereMθ
andM⊥ are proper slant and anti- invariant submanifolds of ~M,
respectively. In this case, if we put θ1 ¼ θ and θ2 ¼ π
2 in Theorem 4, then inequality (4.2) will be khk2 ≥ 2q cot2
θðk∇!ðln f Þk2 − 1Þ, which the inequality (5.1) of [1]. Hence,
Theorem 5.1 of [1] is a special case of Theorem 4.
4. Similarly, if we assume that θ1 ¼ π 2 and θ2 ¼ θ in a warped
product bi-slant
submanifold M ¼ M1 3 f M2 of a Kenmotsu manifold ~M, then M is a
warped product pseudo-slant submanifold M ¼ M⊥ 3 f Mθ such that M⊥
is an anti-invariant submanifold and Mθ is a proper slant
submanifold of ~M. Thus inequality (4.2) takes the
form khk2 ≥ 2q cos2 θðk∇!ðln f Þk2 – 1Þ, which the inequality (4.1)
of [23]. Hence, Theorem
4.1 of [23] is a special case of Theorem 4.
References
[1] F.R. Al-Solamy, M.F. Naghi, S. Uddin, Geometry of warped
product pseudo-slant submanifolds of Kenmotsu manifolds, Quaest.
Math. 42 (3) (2019) 373–389, http://dx.doi.org/10.2989/16073606.
2018.1452800.
[2] L.S. Alqahtani, M.S. Stankovic, S. Uddin, Warped product
bi-slant submanifolds of cosymplectic manifolds, Filomat 31 (16)
(2017) 5065–5071, http://dx.doi.org/10.2298/FIL1716065A.
[3] K. Arslan, R. Ezentas, I. Mihai, C. Murathan, Contact CR-warped
product submanifolds in Kenmotsu space forms, J. Korean Math. Soc.
42 (2005) 1101–1110.
[4] M. Atceken, Warped product semi-slant submanifolds in Kenmotsu
manifolds, Turk. J. Math. 36 (2012) 319–330.
[5] R.L. Bishop, B. O’Neill, Manifolds of negative curvature,
Trans. Amer. Math. Soc. 145 (1969) 1–49.
[6] D.E. Blair, Contact Manifolds in Riemannian Geometry, in
Lecture Notes in Mathematics, vol. 509, Springer-Verlag, New York,
1976.
[7] J.L. Cabrerizo, A. Carriazo, L.M. Fernandez, M. Fernandez,
Semi-slant submanifolds of a Sasakian manifold, Geom. Dedicata 78
(2) (1999) 183–199.
[8] J.L. Cabrerizo, A. Carriazo, L.M. Fernandez, M. Fernandez,
Slant submanifolds in Sasakian manifolds, Glasg. Math. J. 42 (2000)
125–138.
[9] B.-Y. Chen, Geometry of Slant Submanifolds, Katholieke
Universiteit Leuven, 1990.
[10] B.-Y. Chen, Slant immersions, Bull. Aust. Math. Soc. 41 (1990)
135–147.
[11] B.-Y. Chen, Geometry of warped product CR-submanifolds in
Kaehler manifolds, Monatsh. Math. 133 (2001) 177–195.
Submanifolds of Kenmotsu
manifolds
13
[12] B.-Y. Chen, Geometry of warped product CR-submanifolds in
Kaehler manifolds II, Monatsh. Math. 134 (2001) 103–119.
[13] B.-Y. Chen, Pseudo-Riemannian Geometry δ-Invariants and
Applications, World Scientific, Hackensack, NJ, 2011.
[14] B.-Y. Chen, Geometry of warped product submanifolds: a survey,
J. Adv. Math. Stud. 6 (2) (2013) 1–43.
[15] B.-Y. Chen, Differential Geometry of Warped Product Manifolds
and Submanifolds, World Scientific, Hackensack, NJ, 2017.
[16] B.-Y. Chen, S. Uddin, Warped product pointwise bi-slant
submanifolds of Kaehler manifolds, Publ. Math. Debr. 92 (1–2)
(2018) 183–199.
[17] I. Hasegawa, I. Mihai, Contact CR-warped product submanifolds
in Sasakian manifolds, Geom. Dedicata 102 (2003) 143–150.
[18] K. Kenmotsu, A class of almost contact Riemannian manifolds,
Tohoku Math. J. 24 (1972) 93–103.
[19] K. Matsumoto, I. Mihai, Warped product submanifolds in
Sasakian space forms, SUT J. Math. 38 (2) (2002) 135–144.
[20] A. Mihai, Warped product submanifolds in complex space forms,
Acta Sci. Math. Szeged 70 (2004) 419–427.
[21] A. Mihai, Warped product submanifolds in quaternion space
forms, Rev. Roum. Math. Pures Appl. 50 (2005) 283–291.
[22] C. Murathan, K. Arslan, R. Ezentas, I. Mihai, Warped product
submanifolds in Kenmotsu space forms, Taiwanese J. Math. 10 (6)
(2006) 1431–1441.
[23] M.F. Naghi, S. Uddin, F.R. Al-Solamy, Warped product
submanifolds of Kenmotsu manifolds with slant fiber, Filomat 32 (6)
(2018) 2115–2126.
[24] B. Sahin, Warped product pointwise semi-slant submanifolds of
Kaehler manifolds, Port. Math. 70 (2013) 252–268.
[25] S. Uddin, Geometry of warped product semi-slant submanifolds
of Kenmotsu manifolds, Bull. Math. Sci. 8 (2018) 435–451,
http://dx.doi.org/10.1007/s13373-017-0106-9.
[26] S. Uddin, F.R. Al-Solamy, Warped product pseudo-slant
immersions in Sasakian manifolds, Publ. Math. Debr. 91 (3–4) (2017)
331–348, http://dx.doi.org/10.5486/PMD.2017.7640.
[27] S. Uddin, A. Alghanemi, M.F. Naghi, F.R. Al-Solamy, Another
class of warped product CR- submanifolds in Kenmotsu manifolds, J.
Math. Comput. Sci. 17 (2017) 148–157.
[28] S. Uddin, B.-Y. Chen, F.R. Al-Solamy, Warped product bi-slant
immersions in Kaehler manifolds, Mediterr. J. Math. 14 (2) (2017)
http://dx.doi.org/10.1007/s00009-017-0896-8, Art. 95, p. 11.
[29] S. Uddin, V.A. Khan, K.A. Khan, Warped product submanifolds of
a Kenmotsu manifold, Turk. J. Math. 36 (2012) 319–330.
[30] S. Uddin, M.F. Naghi, F.R. Al-Solamy, Another class of warped
product submanifolds of Kenmotsu manifolds, RACSAM 112 (4) (2018)
1141–1155, http://dx.doi.org/10.1007/s13398-017-0415-6.
[31] K. Yano, M. Kon, Structures on Manifolds, in: Series in Pure
Mathematics, World Scientific Publishing Co., Singapore,
1984.
Corresponding author Siraj Uddin can be contacted at:
[email protected]
For instructions on how to order reprints of this article, please
visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm Or contact us
for further details:
[email protected]
AJMS 27,1
Quarto trim size: 174mm x 240mm
On the number of edges of a graph and its complement
Jamel Dammak Department of Mathematics, Faculty of Sciences of
Sfax, Sfax, Tunisia
Gerard Lopez Institut de Mathematiques de Luminy, CNRS-UPR 9016,
Marseille, France, and
Hamza Si Kaddour Institut Camille Jordan, CNRS UMR 5208, Universite
Claude Bernard Lyon 1,
Univ Lyon, Villeurbanne, France
Abstract
Let G ¼ ðV ;EÞbe a graph. The complement of G is the graph G :¼ ðV
; ½V 2 nEÞwhere ½V 2 is the set of pairs fx; ygof distinct elements
ofV . IfK is a subset ofV , the restriction ofG toK is the
graphG
K :¼ ðK; ½K2 ∩EÞ. We prove that ifG ¼ ðV ;EÞ is a graph and k is an
integer, 2 ≤ k ≤ v− 2, then there is a k -element subsetK ofV
such that eðG KÞ≠ eðG
KÞ, moreover the condition k ≤ v− 2 is optimal. We also study the
case
eðG KÞ u eðG
KÞðmod pÞ where p is a prime number. Following a question from
M.Pouzet, we show this:
LetG ¼ ðV ;EÞbe a graphwith vvertices. If eðGÞ ≠ eðGÞ (resp. eðGÞ ¼
eðGÞ) then there is an increasing family
ðHnÞ2≤n≤v (resp. ðHnÞ2≤n≤v−2) of n -element subsets Hn of V such
that eðG Hn
Þ≠ eðG Hn
Þ for all n. Similarly if
eðGÞ u eðGÞ ðmod pÞwhere p is a prime number, p > v− 2, then
there is an increasing family ðHnÞ2≤n≤v of n -element subsets Hn of
V such that eðG
Hn Þ u eðG
Hn Þðmod pÞ for all integer n∈ f2; 3; . . . ; vg.
Keywords Set, Matrix, Graph, Edge, Prime number
Paper type Original Article
1. Introduction Our notations and terminology follow [2]. A graph
is an ordered pair G :¼ ðV ;EÞ (or
ðV ðGÞ;EðGÞÞ), where E is a subset of ½V 2, the set of pairs fx; yg
of distinct elements of V . Elements of V are the vertices of G and
elements of E are its edges. An edge fx; yg is also noted by x y.
The cardinality jV j of V is called the order of G. Two distinct
vertices x and y are adjacent if x y∈EðGÞ, otherwise x and y are
non-adjacent. We denote by eðGÞ :¼ jEðGÞj the number of edges of G.
The degree of a vertex x of G, denoted by dGðxÞ, is the number of
edges which contain x. The graph G is δ -regular (or regular) if
dGðxÞ ¼ δ for all x∈V ;
Edges of a graph and its complement
15
JEL Classification — 05C50, 05C60 © Jamel Dammak, Gerard Lopez and
Hamza Si Kaddour. Published in the Arab Journal of
Mathematical Sciences. Published by Emerald Publishing Limited.
This article is published under the Creative Commons Attribution
(CC BY 4.0) license. Anyone may reproduce, distribute, translate
and create derivative works of this article (for both commercial
and non-commercial purposes), subject to full attribution to the
original publication and authors. The full terms of this license
may be seen at http://
creativecommons.org/licences/by/4.0/legalcode
The publisher wishes to inform readers that the article “On the
number of edges of a graph and its complement” was originally
published by the previous publisher of the Arab Journal of
Mathematical Sciences and the pagination of this article has been
subsequently changed. There has been no change to the content of
the article. This change was necessary for the journal to
transition from the previous publisher to the new one. The
publisher sincerely apologises for any inconvenience caused. To
access and cite this article, please use Dammak, J., Lopez, G.,
Kaddour, H. S. (2019), “On the number of edges of a graph and its
complement”,Arab Journal of Mathematical Sciences, Vol. 27 No. 1,
pp. 15-19. The original publication date for this paper was
29/05/2019.
The current issue and full text archive of this journal is
available on Emerald Insight at:
https://www.emerald.com/insight/1319-5166.htm
Accepted 26 May 2019
Vol. 27 No. 1, 2021 pp. 15-19
Emerald Publishing Limited e-ISSN: 2588-9214 p-ISSN:
1319-5166
DOI 10.1016/j.ajmsc.2019.05.006
δ is called the degree of the regular graph G. The complement of G
is the graph
G :¼ ðV ; ½V 2 nEÞ. If K is a subset of V , the restriction of G to
K, also called the induced
subgraph of G on K, is the graph G K :¼ ðK; ½K2 ∩EÞ. For instance,
given a set V , ðV ; 0=Þ is
the empty graph on V whereas ðV ; fxy : x ≠ y∈VgÞ is the complete
graph. Our first result is Theorem 1.1, we prove that: given a
graph G ¼ ðV ;EÞ and k be an
integer, 2 ≤ k ≤ v− 2, we cannot have eðG KÞ ¼ eðG
KÞ for all k -element subsets K of V , moreover the condition k ≤
v− 2 is optimal, indeed for k ¼ v− 1 a counterexample is given by
(2) of Theorem 1.1.
Theorem 1.1. Let G ¼ ðV ;EÞ be a graph of order v.
(1) If v ≥ 4 , then for each integer kwith 2 ≤ k ≤ v− 2 , there is
a k-element subset K of V
such that eðG KÞ ≠ eðG
KÞ. (2) If G is regular with degree v− 1
2 then eðGÞ ¼ eðGÞ and eðG− xÞ ¼ eðG− xÞ for all vertex x.
(3) Let p be a prime number with p ≥ 3 such that 2dGðxÞ ≡ v− 1ðmod
pÞ for all x∈V .
Then eðGÞ ≡ eðGÞðmod pÞ and eðG− xÞ ≡ eðG− xÞðmod pÞ for all vertex
x. Our second result is Theorem 1.2. Given a graph G ¼ ðV ;EÞ, p a
prime number, and k an
integer, 2 ≤ k ≤ v− 2, under some conditions on k, we cannot have
eðG KÞ ≡ eðG
KÞ ðmod pÞ for all k-element subsets K of V .
Theorem 1.2. Let G ¼ ðV ;EÞ be a graph with v≥ 4 vertices and let p
be a prime number. Let k be an integer, 2 ≤ k ≤ v− 2.
(1) If ( p ¼ 2 and k ≡ 2ðmod 4Þ ) or ( p ≥ 3 and k u 0; 1ðmod pÞ ),
then there is a k-element subset K of V such that eðG
KÞ u eðG KÞðmod pÞ.
(2) If p ≥ 3 and k ≡ 0ðmod pÞ then there is a k-element subset K of
V such that
eðG KÞ u eðG
KÞðmod pÞ if and only if G is neither the complete graph nor the
empty graph.
Our third result is Theorem 1.3. It is related to a question
thatM.Pouzet asked us about the existence, in a graph G ¼ ðV ;EÞ,
of an increasing family ðHnÞn of n -element subsetsHn ofV
such that eðG Hn
Þ ≠ eðG Hn
Þ. Theorem 1.3. Let G ¼ ðV ;EÞ be a graph of v vertices with v≥
3.
(1) If eðGÞ ≠ eðGÞ then there is a vertex x such that eðG− xÞ ≠
eðG− xÞ. (2) If eðGÞ ≠ eðGÞ then there is an increasing family
ðHnÞ2≤n≤vof n-element subsets Hn of V
such that eðG Hn
Þ ≠ eðG Hn
Þ for all integer n∈ f2; 3; . . . ; vg. (3) If eðGÞ ¼ eðGÞ and v ≥
4 then there is an increasing family ðHnÞ2≤n≤v−2 of n-element
subsets Hn of V such that eðG Hn
Þ ≠ eðG Hn
Þ for all integer n∈ f2; 3; . . . ; v− 2g. (4) Let p be a prime
number, p > v− 2 . If eðGÞu eðGÞðmod pÞ then there is an
increasing
family ðHnÞ2≤n≤v of n-element subsets Hn of V such that eðG
Hn
Þu eðG Hn
Þðmod pÞ for all integer n∈ f2; 3; . . . ; vg.
2. Incidence matrices We consider the matrix Wt k defined as
follows: Let V be a finite set, with v elements. Given
non-negative integers t; k, letWt k be the
v
t
by
v
k
matrix of 0’s and 1’s, the rows ofwhich
AJMS 27,1
16
are indexed by the t-element subsets T of V , the columns are
indexed by the k-element subsets K of V , and where the entry
WtkðT;KÞ is 1. T ⊆K and is 0 otherwise. The matrix transpose of Wt
k is denoted
tWt k. A fundamental result, due to D.H. Gottlieb [4], and
independently W. Kantor [5], is this:
Theorem 2.1 (D.H. Gottlieb [4], W. Kantor [5]). For t ≤ minðk; v−
kÞ , Wt k has full row rank over the field of rational
numbers.
It is clear that t ≤ minðk; v− kÞ implies v
t
following result.
Corollary 2.2. For t ≤ minðk; v− kÞ , the rank of Wt k over the
field of rational numbers
is v
and thus KerðtWt kÞ ¼ f0g.
Corollary 2.2 and the following theorem are important tools in the
proof of our main results. In fact, Theorem 2.3 has made to
establish a version modulo a prime of Kelly’s combinatorial lemma
[6]; it also allows to obtain a version modulo a prime of the
particular version of Pouzet’s combinatorial lemma [7].
Let n; p be positive integers, the decomposition of n ¼ PnðpÞ i¼0
nip
i in the basis p is also denoted by ½n0; n1; . . . ; nnðpÞp where
nnðpÞ ≠ 0 if and only if n≠ 0.
Theorem 2.3 [1]. Let p be a prime number. Let v; t and k be
non-negative integers, k ¼ ½k0; k1; . . . ; kkðpÞp, t ¼ ½t0; t1; .
. . ; ttðpÞp, t ≤ minðk; v− kÞ . We have:
(1) kj ¼ tj for all j < tðpÞ and ktðpÞ ≥ ttðpÞ if and only if
KerðtWtkÞ ¼ f0g (mod p ).
(2) t ¼ ttðpÞptðpÞ and k ¼ PkðpÞ i¼tðpÞþ1kip
i if and only if dKerðtWtkÞ ¼ 1 (mod p ) and
fð1; 1; . . . ; 1Þg is a basis of KerðtWt kÞ.
The notation a jb (resp. a b) means a divide b (resp. a not divide
b).
Theorem 2.4 (Lucas’ Theorem [3] ). Let p be a prime number, t; k be
positive integers, t ≤ k, t ¼ ½t0; t1; . . . ; ttðpÞp and k ¼ ½k0;
k1; . . . ; kkðpÞp . Then
k
t
¼
The following result is a consequence of Lucas’ theorem.
Corollary 2.5 ([1]). Let p be a prime number, t; k be positive
integers, t ≤ k, t ¼ ½t0; t1; . . . ; ttðpÞp and k ¼ ½k0; k1; . . .
; kkðpÞp . Then:
pj k
t
if and only if there is i∈ f0; 1; . . . ; tðpÞg such that ti >
ki:
3. Proofs of main results
Let G ¼ ðV ;EÞ be a graph of order v. Let T1;T2; . . . ;T v 2
be an enumeration of the
2-element subsets ofV , letK1;K2; . . . ;K v k
be an enumeration of the k-element subsets ofV .
Let wG be the row matrix ðg1; g2; . . . ; g v 2
Þ where gi ¼ 1 if Ti is an edge of G, 0 otherwise.
We have wGW2 k ¼ ðeðG K1
Þ; eðG K2
Þ; . . . ; eðG K
17
2
Þ with gi ¼ 0 if Ti is an edge of G, 1 otherwise. We
have w G W2 k ¼ ðeðG
K1 Þ; eðG
K2 Þ; . . . ; eðG
K v k
3.1 Proof of Theorem 1.1 (1) Assume that eðG
KÞ ¼ eðG KÞ for all k-element subsetsK ofV , thenwGW2 k ¼ w
G W2 k.
By Corollary 2.2,KerðtWt kÞ ¼ f0g. ThenwG ¼ w G , soG ¼ G, which is
impossible. So there is
a k-element subset K of V such that eðG KÞ ≠ eðG
KÞ. (2) For x∈V , dGðxÞ þ d
G ðxÞ ¼ v− 1. Since dGðxÞ ¼ v− 1
2 then dGðxÞ ¼ d G ðxÞ. We haveP
x∈V dGðxÞ ¼ 2eðGÞ and P
x∈V d G ðxÞ ¼ 2eðGÞ, then eðGÞ ¼ eðGÞ. Now from eðGÞ ¼
eðG− xÞþ d G ðxÞ and eðGÞ ¼ eðG− xÞ þ dGðxÞ, we deduce that eðG− xÞ
¼ eðG− xÞ.
(3) For x∈V , dGðxÞ þ d G ðxÞ ¼ v− 1. Since 2dGðxÞ ≡ v− 1ðmod pÞ
then dGðxÞ ≡ d
G ðxÞ
ðmod pÞ. We conclude using similar arguments to those in item (2).
,
3.2 Proof of Theorem 1.2 We set t:¼ 2. We recall the notation k ¼
½k0; k1; . . . ; kkðpÞp. (1) Case 1. p ¼ 2 and k ≡ 2ðmod 4Þ. We
have k0 ¼ 0, k1 ¼ 1, t ¼ ½0; 12. Since k0 ¼ t0 and k1 ≥ t1 ¼ ttðpÞ
then, by Theorem 2.3,
KerðtWt kÞ ¼ f0g (mod p). Case 2. p ≥ 3 and k u 0; 1ðmod pÞ. We
have k0 ≥ 2, t ¼ t0 ¼ 2. Since k0 ≥ 2 ¼ ttðpÞ then, by Theorem 2.3,
KerðtWt kÞ ¼ f0g
(mod p).
In the two cases,KerðtWt kÞ ¼ f0g (mod p). Assume that eðG KÞ ≡
eðG
KÞðmod pÞ for all k-element subsets K of V . Then wGW2k ¼ w
G W2 kðmod pÞ. As KerðtWt kÞ ¼ f0g (mod p),
then wG ¼ w G ðmod pÞ, so G ¼ G, which is impossible. Then there is
a k-element subset K
of V such that eðG KÞ u eðG
KÞðmod pÞ. (2) If p ≥ 3 then t ¼ t0 ¼ 2 ¼ ttðpÞ. Since k ≡ 0 ðmod
pÞ then k0 ¼ 0, and thus
k ¼ PkðpÞ i¼tðpÞþ1 ki p
i. By Theorem 2.3, fð1; 1; . . . ; 1Þg is a basis of KerðtWt kÞ.
IfG is the complete graph or the empty graph then eðG
KÞ ≡ eðG KÞðmod pÞ. Indeed, ifG
is the complete graph then eðG KÞ ¼
k
2
, eðG
KÞ ¼ 0. Since t0 ¼ 2 > k0 ¼ 0 then, by
Corollary 2.5, pj k
. So eðG
KÞ ≡ 0 ≡ eðG KÞðmod pÞ. If G is the empty graph,
then eðG KÞ ¼ 0 ≡ eðG
KÞ ¼ k
ðmod pÞ.
Conversely if G is neither the complete graph nor the empty graph,
assume that
eðG KÞ ≡ eðG
KÞðmod pÞ for all k-element subsetsK ofV . ThenwGW2 k ¼ w G W2
kðmod pÞ.
So wG −w G ¼ λð1; . . . ; 1Þðmod pÞ with λ∈ f0; 1; − 1g. As G is
neither the complete graph
nor the empty graph, there are i; j such that gi ¼ 0and gj ¼ 1, so
gi − gi ¼ −1 and gj − gj ¼ 1.
Then λ ≠ 1 and λ ≠ − 1. Thus λ ¼ 0, and wG ¼ w G ðmod pÞ, so G ¼ G,
which is impossible.
Then there is a k-element subset K of V such that eðG KÞ u
eðG
KÞðmod pÞ. ,
3.3 Proof of Theorem 1.3 We need the following lemma.
AJMS 27,1
18
Lemma 3.1. Let G ¼ ðV ;EÞ be a graph of order v and let p be a
prime number, p ≥ 3.
(1) Let k be an integer, 2 ≤ k ≤ v− 2. If
v− 2
k− 2
eðGÞ u
v− 2
k− 2
eðGÞ ðmod pÞ then there
is a k-element subset K of V such that eðG KÞ u eðG
KÞðmod pÞ . (2) If ðv− 2ÞeðGÞ u ðv− 2ÞeðGÞ ðmod pÞ then there is
x∈V such that eðG− xÞu
eðG− xÞ ðmod pÞ . Proof. (1) It is an immediate consequence of the
following formula
v 2 k 2
eðG kÞ
(2) Follows from (1) by taking k ¼ v− 1. ,
Now we prove Theorem 1.3.
(1) Assume that eðG− xÞ ¼ eðG− xÞ for all x∈V . From eðGÞ ¼ eðG− xÞ
þ dGðxÞ, we obtain
P x∈V eðGÞ ¼ P
x∈V eðG− xÞ þP x∈V dGðxÞ. Since
P x∈V dGðxÞ ¼ 2eðGÞ then ðv− 2Þ
eðGÞ ¼ P x∈V eðG− xÞ. Similarly, ðv− 2ÞeðGÞ ¼ P
x∈V eðG− xÞ. Then eðGÞ ¼ eðGÞ. (2) We make the proof by induction
on v. We set Hv :¼ V . We assume that Hi is defined
for all k ≤ i ≤ v. Let us defineHk−1. As eðGHk Þ ≠ eðG
Hk Þ then by (1), there is x∈Hk such that
eðG Hk
− xÞ ≠ eðG Hk
− xÞ. We setHk−1 :¼ Hk n fxg. SoHk−1 ⊂Hk and eðGHk− 1 Þ ≠ eðG
Hk− 1 Þ.
(3) By applying (1) of Theorem 1.1 for the graphG and k ¼ v− 2we
obtain x ≠ y∈V such
that eðG− fx; ygÞ ≠ eðG− fx; ygÞ. If v ¼ 4 we are done. If v ≥ 5,
then v− 2 ≥ 3 and here we conclude using (2).
(4) By induction on v. Since eðGÞ u eðGÞðmod pÞ and p > v− 2
then ðv− 2ÞeðGÞ u ðv− 2ÞeðGÞ ðmod pÞ. By (2) of Lemma 3.1, there is
x ∈ V such that eðG− xÞu eðG− xÞ ðmod pÞ. We conclude by using the
induction hypothesis. ,
References
[1] A. Ben Amira, J. Dammak, H. Si Kaddour, On a generalization of
Kelly’s combinatorial lemma, Turk. J. Math. 36 (2014)
949–964.
[2] J.A. Bondy, Basic graph theory: paths and circuits, in: R.L.
Graham, M. Gr€otschel, L. Lovasz (Eds.), HandBook of Combinatorics,
Vol. 1, North-Holland, 1995, pp. 3–110.
[3] N.J. Fine, Binomial coefficients modulo a prime, Amer. Math.
Monthly 54 ((10) Part 1) (1947) 589–592.
[4] D.H. Gottlieb, A certain class of incidence matrices, Proc.
Amer. Math. Soc. 17 (1966) 1233–1237.
[5] W.M. Kantor, On incidence matrices of finite projective and
affine spaces, Math. Z. 124 (1972) 315–318.
[6] P.J. Kelly, A congruence theorem for trees, Pac. J. Math. 7
(1957) 961–968.
[7] M. Pouzet, Application d’une propriete combinatoire des parties
d’un ensemble aux groupes et aux relations, Math. Z. 150 (1976)
117–134.
Corresponding author Hamza Si Kaddour can be contacted at:
[email protected]
For instructions on how to order reprints of this article, please
visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm Or contact us
for further details:
[email protected]
Edges of a graph and its complement
19
On Hilbert class field tower for some quartic number fields
Abdelmalek Azizi Department of Mathematics, Faculty of Sciences,
Mohammed First University,
Oujda, Morocco, and
Mohamed Talbi and Mohammed Talbi Regional center of Education and
Training, Oujda, Morocco
Abstract We determine the Hilbert 2-class field tower for some
quartic number fields k whose 2-class group Ck;2 is isomorphic to
=2 3 =2. Keywords Cyclic quartic field, Class group, Hilbert class
field
Paper type Original Article
1. Introduction Let k be a number field, Ck;2 its 2-class group, it
is the 2-Sylow subgroup of the class group
(in the wide sense) of k; kð0Þ ¼ k ⊆ kð1Þ ⊆ kð2Þ . . . kðiÞ . . .,
the tower of Hilbert 2-class field of k which means that kð1Þ is
the Hilbert 2-class field of k (this is the maximal abelian
unramified
extension of k of degree a power of 2) and kðiþ1Þ is the Hilbert
2-class field of kðiÞ for i≥ 1.
Lemma 1. If G is a 2-group of order 2m , m ≥ 2 , such that G=G0 ’
=23=2 , then G is isomorphic to Qm ðrespectively Dm; Sm; ð2; 2ÞÞ
the quaternion (respectively dihedral, semidihedral, Klein) group
of order 2m . In particular G0, the commutator subgroup of G , is
cyclic.
Proof. See [3]. ,
Let G ¼ Galðkð2Þ=kÞ, be the Galois group of kð2Þ=k, then, with
class field theory we have
that G0 ¼ Galðkð2Þ=kð1ÞÞ ’ C kð1Þ;2 and G=G
0 ¼ Galðkð1Þ=kÞ ’ Ck;2. According to [3] and [8], if
Ck;2 is an elementary group of rank 2, then Ckð1Þ;2 is cyclic,
which implies that the Hilbert
2-class field tower stops at kð2Þ.
AJMS 27,1
JEL Classification — 11R27, 11R29, 11R37 © Abdelmalek Azizi,
Mohamed Talbi and Mohammed Talbi. Published in the Arab Journal
of
Mathematical Sciences. Published by Emerald Publishing Limited.
This article is published under the Creative Commons Attribution
(CC BY 4.0) license. Anyone may reproduce, distribute, translate
and create derivative works of this article (for both commercial
and non-commercial purposes), subject to full attribution to the
original publication and authors. The full terms of this license
may be seen at http://
creativecommons.org/licences/by/4.0/legalcode
Declaration of Competing Interest: Authors do not have any conflict
of interest. The publisher wishes to inform readers that the
article “On Hilbert class field tower for some quartic
number fields”was originally published by the previous publisher of
the Arab Journal of Mathematical Sciences and the pagination of
this article has been subsequently changed. There has been no
change to the content of the article. This change was necessary for
the journal to transition from the previous publisher to the new
one. The publisher sincerely apologises for any inconvenience
caused. To access and cite this article, please use Azizi, A.,
Talbi, M., Talbi, M. (2019), “OnHilbert class field tower for some
quartic number fields”, Arab Journal of Mathematical Sciences, Vol.
27 No. 1, pp. 20-25. The original publication date for this paper
was 22/05/2019.
The current issue and full text archive of this journal is
available on Emerald Insight at:
https://www.emerald.com/insight/1319-5166.htm
Received 15 March 2019 Revised 19 May 2019 Accepted 19 May
2019
Arab Journal of Mathematical Sciences Vol. 27 No. 1, 2021 pp. 20-25
Emerald Publishing Limited e-ISSN: 2588-9214 p-ISSN: 1319-5166 DOI
10.1016/j.ajmsc.2019.05.005
Definition 1 (Taussky’s Conditions). Let F=k be a cyclic unramified
extension and j ¼ jF=k ðrespectively NF=kÞ denote the conorm
(respectively the norm) of F=k.
F=k is of type ðAÞ if and only if kerðjÞ ∩ NF=kðCFÞ
> 1,
F=k is of type ðBÞ if and only if kerðjÞ ∩ NF=kðCFÞ
¼ 1.
Note that kerðjÞ is the set of all the class ideals of k which
capitulate in F.
Theorem 1. Let k be a number field such that Ck;2 is isomorphic to
=23=2;F1;F2;F3
the three unramified quadratic extensions of k within kð1Þ and G be
the Galois group of
kð2Þ=k, then
(1) G is abelian if and only if the four classes of Ck;2 capitulate
in each extension Fi=k ;
(2) G ’ Q3 if and only if the three extensions Fi=k are of type ðAÞ
and in each extension Fi=k only two classes of Ck;2
capitulate;
(3) G ’ Qm avec m > 3 if and only if uniquely one extension Fi=k
is of type ðAÞ and in each extension Fi=k two classes of Ck;2
capitulate;
(4) G ’ Sm if and only if the three extensions Fi=k are of type ðBÞ
and in each extension Fi=k two classes of Ck;2 capitulate;
(5) G ’ Dm if and only if the four classes of Ck;2 capitulate
uniquely in one extension Fi=k .
Proof. See [8]. ,
2. Units of some number fieldss In the remainder of this paper, let
‘ be a prime number congruent to 1 modulo 8, ε the
fundamental unit of ð ffiffi ‘
p Þ and L ¼ ð ffiffiffiffiffiffiffiffiffi ε
ffiffi ‘
pp Þ.
The extension L= is real cyclic of degree 4, of Galois group H ¼
hσi and quadratic
subfieldð ffiffi ‘
p Þ. Since L is of conductor ‘, then L is a subfield of the ‘th
cyclotomic fieldðζ‘Þ and there is a character χ0 of Galððζ‘Þ=Þ ’
ð=‘Þ*, where its kernel is Galððζ‘Þ=LÞ, we will call the character
of L.
Let χ ¼ χ0 þ χ0−1, then χ is a rational character of ðζ‘Þ and L is
fixed by the Common
kernel of χ0 and χ−1. Let EL be the group of units of L, Eχ ¼ fω ∈
ELjω1þσ2 ¼ ±1g the group of χ-relatives units of L (according to
definition of H. W. Leopoldt in [10]), jELj (respectively jEχ j)
the group of the absolute values of EL, (respectively of Eχ), jELj
¼ jELj⊕ jEχ j and εχ a generator of Eχ, then:
Theorem 2. Let L ¼ ð ffiffiffiffiffiffiffiffiffi ε
ffiffi ‘
pp Þwhere ‘ be a prime number congruent to 1modulo 8 and ε
the fundamental unit ofð ffiffi ‘
p Þ , then there exists ξ in EL , such as ξ 2 ¼ ±εε1−σχ and fξ; ξσ
; ξσ2g
is a fundamental system of units of L
Proof. See [4]. ,
(1) ξ1þσ ¼ ±εχ;
(2) ξ1þσ2 ¼ ±ε;
21
(4) Nð ffiffi ‘
p Þ=ðεÞ ¼ NL=ð ffiffi ‘
p ÞðεχÞ ¼ NL=ðξÞ ¼ −1.
Lemma 2. With the same notation of Theorem 2 , fξ; ξσ ; ξσ2g is
also a fundamental system of units of F ¼ Lð
ffiffiffiffiffiffi
−n p Þ where n is a positive integer prime to ‘ and
squarefree.
Proof. If Lð ffiffiffiffiffiffi −n
p Þ ≠ Lð ffiffiffiffiffiffi −1
p Þ, then, according to [1, Proposition 3], to show that fξ; ξσ ;
ξσ2g is a fundamental system of units of F it suffices to show that
nμ is not a square in L, for
μ ¼ ξ01 j1 ξ02
j2ξ03 where fξ01; ξ02; ξ03g ¼ fξ; ξσ ; ξσ2g and j1; j2 ∈ f0; 1g.
Indeed, if μ ¼ ξ, then if nξ ¼ x2 in L, by calculating the norm in
L=k, we find that
ξ1þσ2 ¼ ±ε is a square in k, which is impossible. If μ ¼ ξσ, then
if nξσ ¼ ±n εχ
ξ ¼ x2 in L, by calculating the norm in L=k, we find that
±ε1þσ
ε ¼ ±1 ε is a square in k, which is absurd.
If μ ¼ ξσ 2
, then if nξσ 2 ¼ ±n ε
ξ ¼ x2 in L, by calculating the norm in L=k, we find that
±ε2 ε ¼ ±ε is a square in k, which is not the case.
If μ ¼ ξ1þσ ¼ ±εχ, then if ±nεχ ¼ x2 in L, by calculating the norm
in L=k, we find that
ε1þσ2 χ ¼ ε1þσ ¼ −1 is a square in k, which is absurd.
If μ ¼ ξ1þσ2 ¼ ±ε, then ±nε cannot be a square in L.
If μ ¼ ξσþσ2, then if nξσþσ2 ¼ x2 in L, by calculating the norm in
L=k, we find that
ξ1þσþσ2þσ3 ¼ −1 is a square in k, which is not the case.
If μ ¼ ξ1þσþσ2, then if nξ1þσþσ2 ¼ x2 in L, by calculating the norm
in L=k, we find that
ξ1þσþσ2þσ3ξ1þσ2 ¼ ±ε is a square in k, which is impossible.
If F ¼ Lð ffiffiffiffiffiffi −1
p Þ, then we have m ¼ 2 is the largest integer such that ζm ∈F, so,
from
[1, Proposition 2], to show that fξ; ξσ ; ξσ2g is a fundamental
system of units of F it suffices to
show that 2μ is not a square in L, for μ ¼ ξ01 j1 ξ02
j2ξ03 where fξ01; ξ02; ξ03g ¼ fξ; ξσ; ξσ2g and j1; j2 ∈ f0; 1g.
Using the same reasoning as above we find the result, which
completes the proof of the lemma. ,
3. Hilbert 2-class field tower of ð
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi −2ε
ffiffi ‘
ffiffi ‘
pp Þ where ‘ is a prime number such that ‘ ≡ 1ðmod 8Þ and ð2‘Þ4 ¼
−1,
according to [2], Ck;2 is isomorphic to =2 3 =2, thus kð1Þ=k has
three intermediate
subfields, F1; F2; F3 and letG be the Galois group of kð2Þ=k. Using
[6, Theorem 4, p. 48–49],
it is easy to show that the genus field of k is kð*Þ ¼ F3 ¼ kð
ffiffiffiffiffiffi −2
p Þ, this is the maximal extension of k of the form kMwhich is
unramified for all prime ideals of k, finite and infinite, and such
that M is abelian over .
As ð2‘Þ ¼ 1, then there exist two prime ideals
B1 ¼ 2; aþ b
ffiffi ‘
p
2
ffiffi ‘
p
with a and b are integers, but Bhð‘Þ 1 and Bhð‘Þ
2 are principal ideals of ð ffiffi ‘
p Þwhere hð‘Þ is the class number of ð ffiffi
‘ p Þ, so we can choose c and d positive integers such as
Bhð‘Þ 1 ¼ ðαÞ and Bhð‘Þ
2 ¼ ðαÞ in ffiffi
‘ p
p 2
p 2
are positive. Let H1 and H2 be prime ideals of kaboveB1 and B2
respectively and H ¼ H1H2.
Proposition 1. Let k ¼ ð ffiffiffiffiffiffiffiffiffiffiffiffiffiffi
−2ε
ffiffi ‘
pp Þ where ‘ is a prime number such that ‘ ≡ 1ðmod 8Þ
and ð2‘Þ4 ¼ −1 . Then the class ½H is of order 2 in k . Moreover H
capitulates in kð ffiffiffiffiffiffi −2
p Þ. Proof. The class ofH is of order 2, indeed, we haveH2 ¼ ð2Þ in
k. Suppose thatH ¼ ðλÞ for some λ in k, which is equivalent to ðλ2Þ
¼ ð2Þ in k, therefore, there exists a unit ηof k such that
2η ¼ λ2, although there exist x and y in ð ffiffi ‘
p Þ such that λ ¼ xþ y ffiffiffiffiffiffiffiffiffiffiffiffiffiffi
−2ε
ffiffi ‘
p y2 þ 2xy
ffiffi ‘
pq :
Since ε is a fundamental unit of k and i ¼ ffiffiffiffiffiffi
−1
p ∉ k, then 2η ∈ ð ffiffi
‘ p Þ, therefore x or y is
equal to 0. If y ¼ 0, then 2η ¼ x2, so 2 ¼ x02 or 2ε ¼ x002 in ð
ffiffi ‘
p Þ, which gives thatffiffiffi 2
p ∈ ð ffiffi
−1 p
∈ ð ffiffi ‘
p Þ in the second case, which is impossible. Similarly, if x ¼ 0 we
find that ±‘ is a square in , consequently ½H is of order 2.
To show thatH is capitulated in F3 ¼ kð ffiffiffiffiffiffi −2
p Þ, it suffices to remark that ffiffiffiffiffiffi −2
p is in F3 and
ð ffiffiffiffiffiffi −2
p 2Þ ¼ ð2Þ in F3, thus H capitulates in F3 ,.
Theorem 3. Let k ¼ ð ffiffiffiffiffiffiffiffiffiffiffiffiffiffi
−2ε
ffiffi ‘
pp Þwhere ‘ is a prime number such that ‘ ≡ 1ðmod 8Þ and
ð2‘Þ4 ¼ −1 and F3 ¼ kð ffiffiffiffiffiffi −2
p Þ. Then CF3;2, the 2-part of the class group of F3, is cyclic
of
order h2ðF3Þ ¼ 4.
Proof. Let α, α, B1, B2,H1,H2 andH ¼ H1H2 defined as above and let
L ¼ ð ffiffiffiffiffiffiffiffiffi ε
ffiffi ‘
we have kL ¼ F3,Nk=ð ffiffi ‘
p ÞðH1Þ ¼ B1 andB1 is unramified in L=ð ffiffi ‘
p Þ. So, to prove thatH1
is inert in F3=k it suffices to show that B1 is inert in L=ð ffiffi
‘
p Þ (Translation Theorem), and for
this, we calculate the following norm residue symbol ðα;ε ffiffi
‘
p B1
ffiffi ‘
p
B1
¼
2
soB1 is inert in L=ð ffiffi ‘
p Þ, which givesH1 remains inert in F3=kand in the samewaywe have
shown thatH2 remains inert in F3=kand sinceH ¼ H1H2 capitulates in
F3, then by the Artin reciprocity law, we find that F3=K is of type
ðAÞ, therefore, according to [8], we find that CF3 ;2
is cyclic.
Since F3=ð ffiffi ‘
p Þ is a normal biquadratic extension with Galois group of type ð2;
2Þ and where k;L; k0 ¼ ð ffiffi
‘ p
; ffiffiffiffiffiffi −2
p Þ are these intermediate subfields, then, by [9], we have
h2ðF3Þ ¼ 22−1−2−0qðF3
. ð ffiffi
:
But h2ð‘Þ ¼ 1; h2ðkÞ ¼ 4; h2ðLÞ ¼ 1 (see [12]), and by [7,11] we
have h2ðk0Þ ¼ 2, so
h2ðF3Þ ¼ 4qðF3=ð ffiffi ‘
p ÞÞand we have ε is a fundamental unit of kand since 2ε is not a
square in ð ffiffi
‘ p Þ, then, by [1, Proposition 3], fεg is also a fundamental
system of units of k0 and
according to Lemma 2, L and F3 have the same fundamental system of
units which is
fξ; ξσ; ξσ2g, thus qðF3=ð ffiffi ‘
p ÞÞ, the unit index of F3=ð ffiffi ‘
p Þ, is equal to 1, which gives that h2ðF3Þ ¼ 4. ,
Hilbert class field tower for number fields
23
ffiffi ‘
pp Þwhere ‘ is a prime number such that ‘ ≡ 1 ðmod 8Þ and
ð2‘Þ4 ¼ −1 and let G be the Galois group of kð2Þ=k, then G is of
order 8.
Indeed, we have kð1Þ=F3 is an unramified extension and the 2-class
group of F3 is cyclic,
thus F3 and kð1Þ have the same Hilbert 2-class field, namely kð2Þ,
so # G ¼ 2h2ðF3Þ ¼ 8.
Theorem 4. Let k ¼ ð ffiffiffiffiffiffiffiffiffiffiffiffiffiffi
−2ε
ffiffi ‘
pp Þwhere ‘ is a prime number such that ‘ ≡ 1ðmod 8Þ and
ð2‘Þ4 ¼ −1 , F1; F2; F3 the quadratic subfields of kð1Þ=k . Then,
in each extension Fi=k ,
i∈ f1; 2; 3g , there exist exactly two classes of Ck;2 which
capitulates and G is the quaternion group of order 8.
Proof. According to Lemma 2, we have fξ; ξσ ; ξσ2g is a fundamental
system of units of F3. Since
NF3=kðξÞ ¼ NF3=k
ξσ
2 ¼ ±ε and NF3=kðξσÞ ¼ ±ε−1;
then NF3=kðEF3Þ ¼ Ek, consequently, by [5], we have only two
classes of Ck;2 which capitulates in F3, namely ½H and its square
and since the extension F3=k is of type ðAÞ, then, by Theorem 1, we
find that G is a quaternion group, and by Remark 2, the group G is
of order 8, therefore G ’ Q3. ,
Remark 3. As G ’ Q3, then, by [8],
CF1 ;2 ’ CF2 ;2 ’ CF3 ;2 ’ 4:
Example 1. Let k ¼ ð
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
−2ð17þ 4
p Þ q
Þ, we have 17 ≡ 1ðmod 8Þ and ð 217Þ4 ¼ −1, then
the Galois group of kð2Þ=k is the quaternion group of order 8,
there exist exactly two classes of
Ck;2 which capitulates in F3 ¼ kð ffiffiffiffiffiffi −2
p Þ, as well for F1 and for F2 and the Hilbert 2-class field
tower of k stops at kð2Þ.
References
[1] A. Azizi, Unites de certains corps de nombres imaginaires et
abeliens Sur , Ann. Sci. Math. Quebec 23 (2) (1999) 15–21.
[2] E. Brown, C.J. Parry, The 2-class group of certain biquadratic
number fields II, Pac. J. Math. 78 (1) (1978) 11–26, Pacific
Journal of Mathematics.
[3] D. Gorenstein, Finite Groups, Harper and Row, New York,
1968.
[4] M.N. Gras, Table numerique du nombre de classes et des unites
des extensions cycliques reelles de degre 4 de ,, Publ. Math. Fac.
Sci. Besançon, Theor. Nombres 2 (1977-78) 1–79.
[5] F.P. Heider, B. Schmithals, Zur kapitulation der idealklassen
in unverzweigten primzyklischen erweiterungen, J. Reine Angew.
Math. 366 (1982) 1–25.
[6] M. Ishida, The Genus Fields of Algebraic Number Fields, in:
Lecture Notes in Mathematics, vol. 555, Springer-Verlag, London.,
1976.
[7] P. Kaplan, Sur le 2-groupe des classes d’ideaux des corps
quadratiques, J. Reine Angew. Math. 283/284 (1976) 313–363.
[8] H. Kisilevsky, Number fields with class number congruent to 4
modulo 8 and Hilbert’s theorem 94, J. Number Theory 8 (1976)
271–279.
[9] F. Lemmermeyer, Kuroda’s class number formula, Acta Arith. 66
(3) (1994) 245–260.
AJMS 27,1
[10] H.W. Leopoldt, Uber einheitengruppe und klassenzahl reeller
abeslscher Zahlk€orper, Abh. Deutsche Akad. Wiss. Berlin, Math. 2
(1953) 1–48.
[11] T.M. McCall, C.J. Parry, R.R. Ranalli, Imaginary bicyclic
biquadratic fields with cyclic 2-class group, J. Number Theory 53
(1) (1995) 88–99.
[12] L.C. Washington, Introduction To Cyclotomic Fields, (Book 83),
second ed., in: Graduate Texts in Mathematics, Springer–Verlag, New
York, 1997.
Corresponding author Mohammed Talbi can be contacted at:
[email protected]
For instructions on how to order reprints of this article, please
visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm Or contact us
for further details:
[email protected]
Hilbert class field tower for number fields
25
Quarto trim size: 174mm x 240mm
A Method of approximation for a zero of the sum of maximally
monotone mappings in Hilbert spaces
Getahun Bekele Wega and Habtu Zegeye Department of Mathematics and
Statistical Sciences,
Botswana International University of Science and Technology,
Palapye, Botswana
Abstract Our purpose of this study is to construct an algorithm for
finding a zero of the sum of twomaximally monotone mappings in
Hilbert spaces and discus its convergence. The assumption that one
of the mappings is α-inverse strongly monotone is dispensed with.
In addition, we give some applications to the minimization problem.
Our method of proof is of independent interest. Finally, a
numerical example which supports our main result is presented. Our
theorems improve and unifymost of the results that have been proved
for this important class of nonlinear mappings.
Keywords Firmly nonexpansive, Hilbert spaces, Maximally monotone
mapping, Strong convergence,
Zero points
Paper type Original Article
1. Introduction Let H be a real Hilbert space with inner product
h:; :i and induced norm k:k. Let A : H → 2H
be a nonlinear mapping. The domain, range, zero, and graph of A are
respectively the sets DomðAÞ ¼ fx ∈ H : Ax ≠ 0=g; RðAÞ ¼ fAx : x ∈
DomðAÞg; ZeroðAÞ ¼ fx ∈ H : 0 ∈ Axg and GphðAÞ ¼ fðx; yÞ ∈ H 3H : y
∈ Axg. A mappingA : H → 2H is called monotone if for any x; y ∈ H
and u ∈ Ax; v ∈ Aywe have
AJMS 27,1
JEL Classification — 47H05, 47J25, 49M27, 90C25 ©Getahun BekeleWega
and Habtu Zegeye. Published in theArab Journal of Mathematical
Sciences.
Published by Emerald Publishing Limited. This article is published
under the Creative Commons Attribution (CCBY4.0) license. Anyonemay
reproduce, distribute, translate and create derivativeworks of this
article (for both commercial and non-commercial purposes), subject
to full attribution to the original publication and authors. The
full terms of this license may be seen at http://creativecommons.
org/licences/by/4.0/legalcode
The authors thank the anonymous referees for useful suggestions
which improved the contents of this paper.
Funding: The first author gratefully acknowledges the funding
received from Simons Foundation based at Botswana International
University of Science and Technology (BIUST), Palapye,
Botswana.
Declaration of Competing Interest: None. The publisher wishes to
inform readers that the article “AMethod of approximation for a
zero of the
sum of maximally monotone mappings in Hilbert spaces” was
originally published by the previous publisher of the Arab Journal
of Mathematical Sciences and the pagination of this article has
been subsequently changed. There has been no change to the content
of the article. This change was necessary for the journal to
transition from the previous publisher to the new one. The
publisher sincerely apologises for any inconvenience caused. To
access and cite this article, please useWega, G. B., Zegeye, H.
(2019), “AMethod of approximation for a zero of the sum of
maximally monotone mappings in Hilbert spaces”, Arab Journal of
Mathematical Sciences, Vol. 27 No. 1, pp. 26-40. The original
publication date for this paper was 22/05/2019.
The current issue and full text archive of this journal is
available on Emerald Insight at:
https://www.emerald.com/insight/1319-5166.htm
Received 4 April 2019 Revised 16 May 2019 Accepted 17 May
2019
Arab Journal of Mathematical Sciences Vol. 27 No. 1, 2021 pp. 26-40
Emerald Publishing Limited e-ISSN: 2588-9214 p-ISSN: 1319-5166 DOI
10.1016/j.ajmsc.2019.05.004
hu v; x yi≥ 0: (1.1)
A monotone mapping A : H → 2H is called maximally monotone if
GphðAÞ is not properly contained in the graph of any other monotone
operator. The resolvent of A is given by JλA ¼ ðI þ λAÞ−1, where I
is the identity mapping on H and λ > 0.
Let A; B : H → 2H be maximally monotone mappings. Consider the
problem of finding z∈H such that
0 ∈ Azþ Bz: (1.2)
We denote the solution set of (1.2) by ðAþ BÞ−1ð0Þ. This problem
includes, as special cases, convex programming, variational
inequalities, split feasibility problem and minimization problem.
For solving problem (1.2), we remark that several authors have
studied different iterative schemes (see, for example,
[3,4,10,11,14,21] and the references therein).
In 1979, Passty [11] introduced Forward–backward splitting method
which defines a sequence fxng by
xnþ1 ¼ ðI þ rnBÞ−1ðI rnAÞxn; n ≥ 1; (1.3)
where frng is a sequence of positive numbers, A and B are maximal
monotone mappings with DomðAÞ ⊂ DomðBÞ, and A is single valued.
This method is essentially a generalization of the classical
gradient method for constrained convex optimization and monotone
variational inequalities, and inherit restrictions similar to those
methods such as A is single valued. In general, this method
provides weak convergence even with the restriction that A is
single-valued. It fails to provide weak convergence results to the
zero of the sum of the general maximal monotone mappings.
In 1979, Lions and Mercier [7] introduced Peaceman–Rachford
splitting method whose iteration method is given by
yn ¼ 2ðI þ λBÞ−1 IÞxn;
zn ¼ 2ðI þ λAÞ−1 IÞyn;
xnþ1 ¼ ð1 ρnÞxn þ ρn zn; n ≥ 1;
(1.4)
where λ > 0 is a fixed scalar, and fρng⊂ ð0; 1 is a sequence of
relaxation parameters. They proved weak convergence of this
sequence to the solution of problem (1.2) under certain
conditions.
In 2008, Eckstein and Svaiter [5] constructed new approach
splitting algorithms which starts by reformulating (1.2) as the
problem of locating a point in a certain extended solution set
SeðA; BÞ ⊂ H 3H and proved weak convergence results provided that H
has finite dimension or Aþ B is maximal monotone mapping. The
extended solution set for the problem (1.2), which is the subset of
H 3H is defined by:
SeðA; BÞ ¼ fðz; wÞ∈ H 3 H : w ∈ BðzÞ;w ∈ AðzÞg: (1.5)
We treat H 3 H as a Hilbert space by endowing it with the canonical
inner product
hðx; x0 Þ; ðy; y0 Þi ¼ hx; yi þ hx0 ; y
0 i: More recently, in order to solve problem (1.2), Svaiter [17]
studied the following Algorithm: for any λ > 0; x0; b0 ∈H,
Step 1. Compute; yn; an such that an ∈ AðynÞ; λan þ yn ¼ xn−1 −
λbn−1:
Step 2. Compute; xn; bn such that bn ∈ BðxnÞ; λbn þ xn ¼ yn þ
λbn−1; n ≥ 1:
Approximation in Hilbert
spaces
27
They proved that fðxn; bnÞg and fðyn; − anÞg converge weakly to a
point ðz; wÞ in SeðA; BÞ, where z ∈ ðAþ BÞ−1ð0Þ. We remark that the
convergence is still weak convergence.
With regard to a strong convergence, several authors have studied
different iterative schemes (see for example, [12,18–20,22] and the
references therein) for a zero of the sum of monotone m