
ISSN: 2644-1381
Ramy Shaheen*
Received: February 03, 2020; Published: February 27, 2020
*Corresponding author: Ramy Shaheen, Department of Mathematics Faculty of Science, Tishreen University, Lattakia, Syria
DOI: 10.32474/CTBB.2020.02.000137
A mixed graph GM (V, E, A) is a graph containing unoriented edges (set E) as well as oriented edges (set A), referred to as arcs. In this paper we calculate the domination number of the Cartesian product of a path Pm with directed path Pn (mixed-grid graph pm × pn) for some values of m and arbitrary n.
Keywords: graph, directed graph, Cartesian product, path, directed path, mixed graph, mixed-grid graph, dominating set, domination number.
All graphs and digraphs are assumed to be loop less and without duplicate edges or arcs. A mixed graph GM (V, E, A) is a graph containing unoriented edges (set E) as well as oriented edges (set A), referred to as arcs. This notion was first introduced in [1]. Let G = (V1,E) be a graph and D = (V2, A) be a digraph. The Cartesian product G × D is the mixed-graph with vertex set V (G × D) = V1(G) ×V2 (D) and edge (arc) set is ((u1, v1), (u2,v2 )∈ E(G × D) if and only if either v1 = v2 and (u1,u2 )∈ E(G) or u1 = u2 and( v1,v2 )∈ A(D). A subset S of the vertex set V (G × D) is a dominating set of (G × D) if for each vertex V ∈G × D there exists a vertex u ∈ s such that (u, v) is an edge (arc) of G × D . The domination number of G × D , γ G × D , is the cardinality of the smallest dominating set of G × D . Let Pm be a path with vertex set V (Pm) = {1, 2,..., n} , and edge set E(Pm) = {(i, i +1) :1 ≤ i ≤ n −1} and let Pn be a directed path with vertex set V (Pn ) = {1,2,...,n}, and arc set A(Pn ) = {(i, i +1) :1 ≤ i ≤ n −1}. Then for Cartesian product Pm and Pn is mixed-grid graph Pm × Pn with V (Pm × Pn ) = {(i, j) :1 ≤ i ≤ m,1 ≤ j ≤ n}, such that there is an arc from (i, j) to ( p, q)if and only fi = p and q − j = 1and is an edge from (i, j) to ( p, q)if and only if j = q and p − i = 1. The i’th row of V (Pm × Pn ) is Ri = {is a directed pathPn(i, j): j=1,2,...,n} The j’th column Kj = {(i, j):is a path Pm(i, j):i=1,2,...,m} . If S is a dominating set for Pm × Pn, then we denote Wj = S ∩ K j.
Let Sj= |Wj| , where the sequence (s1, s2, .....,sn ) is called a dominating sequence corresponding to S. For 1 ≤ i ≤ n , the vertices of j-column are dominated by vertices of j-column or (j-1) column. The vertices of the first column are dominated only by [2] vertices of K1. Also, for 1 ≤ i ≤m, the vertices of i-row are dominated by vertices of i-row or (i −1) − row. The vertices of the first row are dominated only by vertices of R1 or R2. Thus, the following is true γ (Pm× Pn) ≤ γ ( Pm× Pn ) ≤ γ ( Pn × Pn ). For finding domination number of grid graphs Pm × Pn, Jacobson and Kinch in [2], were calculated the domination number of cartesian product of undirected paths Pm and Pn for m = 1, 2, 3, 4 . The cases m = 5, 6 were calculated by Chang and Clark [3]. Also, Chang et al. [4], established the upper bounds of cartesian product of undirected paths Pm and Pn for 5 ≤ m ≤ 10 and arbitrary n. In [5], Gravier and Mollard given un upper and lower bounds of general cartesian product of two undirected paths. Goncalves et al. [6] proved Chang’s conjecture saying that for every 16 ≤ n ≤ m,γ (Pm × Pn ) = [(n +1)(m + 2) / 5] − 4. For domination number of directed grid graphs, Liu et al. [7], they studied the domination number of Pm × Pn for m = 2, 3, 4, 5, 6 and arbitrary n. Also, in [8] the author studied the domination number of Pm × Pn for arbitraries m and n.
In this section we calculate the domination number of the Cartesian product of a path Pm and a directed path Pn for some values of m and arbitrary n.
Since for each vertex (i, j)∈V (Pm × Pn ) has two undirected degrees in V(Kj) one outdegree in V(Kj+1) + and one indegree from V(Kj-1), then can it dominates at must four vertices of Pm × Pn with itself. Thus, implies that γ (Pm × Pn ) ≥ mn / 4.
Let S be a dominating set of Pm × Pn. Since the vertices of the first column are dominated only by vertices of K1. Also, for 1 ≤ i ≤ n , the vertices of j-column are dominated by vertices of j-column or (j-1)-column. Then the following are holds:
1. s1 ≥ [m / 3].
2. sj/ + 3sj+1≥m for allj=1,....,n
3 s j−1 + 4sj + sj+1 ≥ 2mfor all j = 2,....,n
There is a minimum dominating set S for Pm × Pn with dominating sequence ( s1,s2,....,sn) such that, for all j =1, 2,..., n, , is [( 2) / 4] [ / 2] j m+ ≤ s ≤ m .
Proof: Let S be a minimum dominating set for Pm × Pn with dominating sequence (s1,s2,....,sn ). Assume that for some j, sj is large. Then we modify S by moving some vertices from column j to column j +1, such that the resulting set is still dominating set for Pm × Pn (because each vertex in s ∩ Kjis dominates only vertices from Kj and Kj+1 ). For 1 ≤ i ≤ m and 1 ≤ i ≤ n , let W = s∩{(i, j), (i +1, j), (i + 2, j), (i + 3, j)}. If W ≥ 3 , then we have three cases.
Case: If {(i, j), (2, j)} ⊆ S or {(m −1, j(m, j)} ⊆ S . Then we can moving (1, j) to (1, j +1) or (m, j) to (m, j +1) . Furthermore, S is still dominating set of Pm × Pn.
Case: |W| = 4 , then we put
S1 = S −W ∪ i j i + j + i + j + i + j
Case: W = 3 then we have two sub cases
a. SubCase: W = 3 and W = S ∩{(i, j),(i +1, j),(i + 2, j),(i + 3, j)}={(i, j), (i+1, j), (i+2, j)} or W = {(i+1, j), (i+2, j), (i+3, j)} . Two cases are similar by symmetry, for the first we put S1=(S −W) ∪{(i, j), (i + 2, j), (i +1, j +1)} and for the second we put S1=(S −W) ∪{(i, j), (i + 3, j), (i + 2, j +1)}
b. Subcase: |W| = 3 and W = S ∩{(i, j),(i +1, j),(i + 2, j),(i + 3, j)}= {(i, j), (i+1, j), (i+3, j)} or W = {(i, j), (i + 2, j), (i+3, j)}. Also, two cases are similar by symmetry. Then we change S, respectively as follows: S1=(S −W) ∪{(i, j),(i+3, j)(i+1,j+1)} , S1=(S −W) ∪{(i, j), (i+3, j),(i+2, j+1)} , see (Figure 1) for cases 2, 3. We repeat this process if necessary eventually leads to a dominating set with required properties. Also, we get S1 is a dominating set for Pm × Pn with |S| = |S1|. Thus, we can assume that every four consecutive vertices of the j’th column include at most two vertices of S. This implies that Sj ≤ [m / 2], for all 1 ≤ j ≤ n .To prove the lower bound, we suppose that Kj ∩ D is be a maximum, i.e Sj = [m / 2] Then there at must m −[m / 2] vertices in Kj+1 are not dominated (Figure 2). By maximality of Sj, does not have three successive vertices do not belong to S (i.e., does not exist three successive vertices Kj+1 from does not dominated by vertices from S ∩ Kj ). Thus, implies that Sj+1 ≥ (m −[m / 2]) / 2 = [(m + 2) / 4] By
c. Lemma: always we have a minimum dominating set S with dominating sequence 1 2 ( s1,s2,...., sn ), such that [(m + 2) / 4] ≤ Sj ≤ [m / 2], for all j = 1, 2,..., n
Proposition: Clearly,γ(P1 × Pn)= γ(Pn)= [n/2]
Theorem γ(P2 × Pn ) = n
Proof. Let S be a set defined as follows
We have
|S| = n , also S is a dominating set of P2 × Pn. By Observation 2, we
have Sj + 3Sj+1 ≥ m. Which implies that, if
Sj = 0 is Sj+1 = 2
and if Sj = 1 is Sj+1 ≥ 1. Then we get γ(P2 × Pn ) = Σj=1nsj ≥n
Thusγ (γ(P2 × Pn) = n
Theorem: γ(P3 × Pn ) = n
Proof: We define S as follows: S = {(2, j) :1 ≤ j ≤ n}
Certainly, S is a dominating set of P3 × Pn with |S| = n
Also, we have γ (P3 × Pn ) ≥ γ (P2 × Pn ) = n Thus
γ(P4 × Pn )=n
Proposition: γ(P4 × Pn≤[3n/2] )
Proof. Here, we define S as follows:
s={(1, 4j- 3), (4, 4j-3 3) : 1 ≤j ≤ [n/ 2]} ∪{(3, 4j- 2) :1≤j≤ [n/2] }
Certainly S is a dominating set P4 × Pn (see Figure 3 for γ(P4 × P10 ), . Also, we have
|S| = 2[n /4]+ 2[(n −1) / 4] + [(n − 2) / 4] + [(n − 3) / 4] = [3n / 2]
γ (P4 × Pn) ≤ [3n / 2].
Proposition:
a. There only two cases of (Sj,Sj+1) = (1,1) such that Kj∩S = {(1,j )} and Kj+1 ∩ S = {(3,j+1)} or Kj ∩ S = {(4, j)} and Kj+1 ∩ S = {(2, j +1)} Furthermore, two cases are similar by symmetry. We consider the first case.
b. The case (Sj, Sj+1 Sj+2) = (1,1,1) is not possible.
c. The case (S1, S2, S3) = (2,1,1) is not possible.
d. If K2 ∩ S = {(1, 2) then S1 = 3
e. If K3 ∩ S = {(1, 3)} then S1 + S2 = 4 .
Proof: Immediately by drawing.
Proposition: If (Sj , Sj+1) = (1,1) , then (Sj−2, Sj−3) = (2, 2) or (Sj−2r , Sj−2r+1) = (1, 2) {where 1 ≤ r and 1 ≤ j − 2r . Furthermore, the second case repeated until exist two columns before Kj , Kj+1, contains four vertices from S.
Proof: From Proposition 4.5.1 (1), we have Kj ∩ S = {(1, j)} and Kj+ ∩ S = {(3, j +1)}. This implies that Kj−1 ∩ S = {(3, j −1), (4, j −1)} and then {(1, j − 2)} ⊆ Kj−2 ∩ S . If |Kj−2 ∩S| = 2 then Sj−1 + Sj−2 = 4 otherwise | Kj−2 ∩ S| = 1 and Kj-3 ∩ = {(3,j- 3), (4,j-4)} {(1, 4)} 4 Kj-4∩ S={(1,j-4)} and repetition this case until (S1, S2) = (1, 2) or (S2, S3) = (1, 2) {where j is odd or even, respectively}.
Theorem: γ(P4 × Pn=[3n/2]
Proof. By Lemma 2.1, 1 ≤ Sj ≤ 2 for all j = 1,..., n .We have two cases:
Case 1. Sj + Sj+1 ≥ 3 for j = 1, …, n-1. Then we get
Case 2. Sj + Sj+1 = 2 for some 1 ≤ j ≤ n −1 . Thus, we have (Sj + Sj+1) = (1,1). By Proposition 2.3(1), we consider the first case Kj ∩ S = {(1,j)} and Kj+1 ∩ S = {(3,j+ 1)} . There are two subcases
Sub Case Sj−2r + Sj−2r+1 = 4 {where 4 ≤ j ≤ −2r ≤ j − 2 }. Then we can assume that each two columns including at least three
vertices from S and then
Sub Case (Sj−2, Sj−1) = (Sj−4, Sj−3) = ... = (Sj−2r + Sj−2r+1) = (1, 2) .
Then we study the cases:
a. If j − 2r = 1 i.e., S1= 1 and this impossible.
b. If j − 2r = 2 . Then K2 ∩ S = {(1, 3)} and {(3,1), (4, 2)} ⊆ K2 ∩ S
. But, needs (1,1)∈ S or (2,1)∈ S for dominate (1, 1). Furthermore S1 + S2 = 4 .
c. If j − 2r = 3. Then K3 ∩S = {(1,3)} and {(3, 2), (4, 2)} ⊆ K2 ∩ S . Then needs (1,1)∈ S and (3,1)∈ S or (4,1)∈ S , also S1 + S2 = 4 .
Now, we conclude that for all cases, when exist the case
(Sj , Sj+1) = (1,1) there is the case Sj−2r + Sj−2r+1 ≥ 4 before
repeated the case (Sq , Sq+1) = (1,1) . Thus, we can assume that
each four columns including at least 6 vertices from S. Finally, get γ (P4 × Pn)= Finally, Proposition 2.2 together
with the last result gets
γ (P4 × Pn) ≤ [3n/2]
γ (P5 × Pn) ≤ [3n / 2]+1.
Proof: Let S defined as follows:
S{(1,1), (3,1), (5,1)}∪ {(3, 2j ) :1 ≤j ≤ [n/2]} ∪{(1, 2j+ 1):(5, 2j+ 1) :1 [(n- 1) / 2]}
We can check that S is a dominating set of P5 × Pn (see Figure 3 for γ (P5 × P12 )), with
S = 3 + [n / 2] + 2[(n −1) / 2] = [3n / 2] +1We need to the following fact:
Fact 1.
1. There is one possible for (Sj , Sj+1, Sj+2 ) = (1, 2,1) . Furthermore, if (Sj , Sj+1, Sj+2 ) = (1, 2,1) then Kj ∩ S = {(3, j)}, Kj+1 ∩ S = {(1, j+1),(5 + j +1)} and Kj ∩ S = {(3, j + 2)} .
2. (S1, S2, S3) = (2, 2,1) in not possible.
3. (S1, S2, S3)= (2,1, 2) in not possible. Furthermore, S1+ S2 ≥ 4
and S1+ S2 + S3 ≥ 6
Proof. 1). By Observation 2.2, we have (Sj , Sj+1) = (1,1) is not possible. But, (Sj , Sj+1) = (1, 2) or (2, 1) are possible, also (Sj , Sj+1, Sj+2 ) = (1, 2,1) is possible. Thus for Sj+2 = 1 we must have {(1, 1), (2, 1)} Kj+1 ∩ S = j + j + , {(4, j +1), (5, j +1)} or {(1, j +1), (5, j +1)}. For the first and second needs Sj = 2 , we rejected. Then for the case {(1, 1), (5, 1)} Kj+1 ∩ S = j + j + , we get Sj = 1 and Sj+2= 1 , furthermore Kj ∩ S = {(3, j)}and Kj ∩ S = {(3, j + 2)} .
2). From (1), if S3 = 1 then we have
K2 ∩ S = {(1, 2), (2, 2)},{(4, 2), (5, 2)}or {(1, 2), (5, 2)} . For all
these cases S1 = 2 is not possible.
3) Immediately from prove (1).
Now, From Lemma 2.1, we have 1 ≤ Sj ≤ 3 . Also, Observation
2.2, gets (3, 1, 1) is not possible. By Fact 1, we include that
S1 + S2 + S3 ≥ 6 , furthermore if Sj = 2 then for
r ≥ 2. Thus
By recent results with (1), we get the required.
Theorem γ(P6 × Pn=n
Proof. Let S defined as follows: S = S = {(2,1), (5, j) :1≤ j ≤ n} . Definitely S is a dominating set of P6 × Pn with γ (P6 × Pn ) ≤ |S| = 2n.
By Observation 2.2, S1 ≥ 2 and (Sj , Sj+1) = (1, 2) is not
possible. This implies that γ(P6 × Pn)= Thus we get
Theorem γ(P7 × Pn)=2n+2
Proof. For P7 × Pn we define S as follows:
S = {(1,1),(3,1),(5,1),(7,1),}∪{(3,2 j),(7,2 j) :1≤ j ≤ |n / 2| }∪{(1,2 j +1),(5,2 j +1) :1≤ j ≤ |(n −1) / 2| }
We have S is a dominating set of P7 × Pn (see (Figure 4), for γ(P7 × P15 ), with |s| = 4 + 2|n / 2| + 2| (n −1) / 2| = 2n + 2 We need to the following fact.
a. Fact 2.
b. s1 ≥ 3
c. s1 ≥ 3
d. (s1,s2 ) = (4,1) and (s1,s2,s3)= (3, 2, 2)
e. s1+s2 ≥ 6 or s2+s3 ≥ 6
Proof. 1). It is clear from Observation 2.2.
2) Immediately by drawing.
3) From (2), it’s clear.
4) By Observation 2.2, we have sj 3sj+1 ≥ m. This implies that
(sj, Sj+1) = (3,1) is not possible. Then we get the required for (4).
Now, by Fact 2, We conclude that
Thus, by (2) together with the last result gets (P7 × Pn ) = 2n +2.
Bio chemistry
University of Texas Medical Branch, USADepartment of Criminal Justice
Liberty University, USADepartment of Psychiatry
University of Kentucky, USADepartment of Medicine
Gally International Biomedical Research & Consulting LLC, USADepartment of Urbanisation and Agricultural
Montreal university, USAOral & Maxillofacial Pathology
New York University, USAGastroenterology and Hepatology
University of Alabama, UKDepartment of Medicine
Universities of Bradford, UKOncology
Circulogene Theranostics, EnglandRadiation Chemistry
National University of Mexico, USAAnalytical Chemistry
Wentworth Institute of Technology, USAMinimally Invasive Surgery
Mercer University school of Medicine, USAPediatric Dentistry
University of Athens , GreeceThe annual scholar awards from Lupine Publishers honor a selected number Read More...