Nowhere-zero integral flows on a bidirected graph

It is proved that every bidirected graph which can be provided with a nowhere-zero integral flow can also be provided with a nowhere-zero integral flow with absolute values less than 216. The connection between these flows and the local tensions on a graph which is 2-cell imbedded in a closed 2-manifold is explained. These local tensions will be studied in a subsequent paper.     

Nowhere-zero 15-flow in 3-edge-connected bidirected graphs

It was conjectured by Bouchet that every bidirected graph which admits a nowhere-zero κ flow will admit a nowhere-zero 6-flow. He proved that the conjecture is true when 6 is replaced by 216. Zyka improved the result with 6 replaced by 30. Xu and Zhang showed that the conjecture is true for 6-edge-connected graphs. And for 4-edge-connected graphs, Raspaud and Zhu proved it is true with 6 replaced by 4. In this paper, we show that Bouchet＇s conjecture is true with 6 replaced by 15 for 3-edge-connected graphs.     

Nowhere-zero flows in low genus graphs

A nowhere-zero k-flow is an assignment of edge directions and integer weights in the range 1,…, k ? 1 to the edges of an undirected graph such that at every vertex the flow in is equal to the flow out. Tutte has conjectured that every bridgeless graph has a nowhere-zero 5-flow. We show that a counterexample to this conjecture, minimal in the class of graphs embedded in a surface of fixed genus, has no face-boundary of length <7. Moreover, in order to prove or disprove Tutte's conjecture for graphs of fixed genus γ, one has to check graphs of order at most 28(γ ? 1) in the orientable case and 14(γ ? 2) in the nonorientable case. So, in particular, it follows immediately that every bridgeless graph of orientable genus ?1 or nonorientable genus ?2 has a nowhere-zero 5-flow. Using a computer, we checked that all graphs of orientable genus ?2 or nonorientable genus ?4 have a nowhere-zero 5-flow.     

Nowhere-zero eigenvectors of graphs

A vector is called nowhere-zero if it has no zero entry. In this article we search for graphs with nowhere-zero eigenvectors. We prove that distance-regular graphs and vertex-transitive graphs have nowhere-zero eigenvectors for all of their eigenvalues and edge-transitive graphs have nowhere-zero eigenvectors for all non-zero eigenvalues. Among other results, it is shown that a graph with three distinct eigenvalues has a nowhere-zero eigenvector for its smallest eigenvalue.     

Flows in 3-edge-connected bidirected graphs

It was conjectured by A. Bouchet that every bidirected graph which admits a nowhere-zero k-flow admits a nowhere-zero 6-flow. He proved that the conjecture is true when 6 is replaced by 216. O. Zyka improved the result with 6 replaced by 30. R. Xu and C. Q. Zhang showed that the conjecture is true for 6-edge-connected graph, which is further improved by A. Raspaud and X. Zhu for 4-edge-connected graphs. The main result of this paper improves Zyka’s theorem by showing the existence of a nowhere-zero 25-flow for all 3-edge-connected graphs.     

Free multiflows in bidirected and skew-symmetric graphs

A graph (digraph) G=(V,E) with a set T⊆V of terminals is called inner Eulerian if each nonterminal node v has even degree (resp. the numbers of edges entering and leaving v are equal). Cherkassky and Lovász, independently, showed that the maximum number of pairwise edge-disjoint T-paths in an inner Eulerian graph G is equal to , where λ(s) is the minimum number of edges whose removal disconnects s and T-{s}. A similar relation for inner Eulerian digraphs was established by Lomonosov. Considering undirected and directed networks with “inner Eulerian” edge capacities, Ibaraki, Karzanov, and Nagamochi showed that the problem of finding a maximum integer multiflow (where partial flows connect arbitrary pairs of distinct terminals) is reduced to O(logT) maximum flow computations and to a number of flow decompositions.In this paper we extend the above max-min relation to inner Eulerian bidirected graphs and inner Eulerian skew-symmetric graphs and develop an algorithm of complexity for the corresponding capacitated cases. In particular, this improves the best known bound for digraphs. Our algorithm uses a fast procedure for decomposing a flow with O(1) sources and sinks in a digraph into the sum of one-source-one-sink flows.     

Nomadic decompositions of bidirected complete graphs

We use to denote the bidirected complete graph on n vertices. A nomadic Hamiltonian decomposition of is a Hamiltonian decomposition, with the additional property that “nomads” walk along the Hamiltonian cycles (moving one vertex per time step) without colliding. A nomadic near-Hamiltonian decomposition is defined similarly, except that the cycles in the decomposition have length n-1, rather than length n. Bondy asked whether these decompositions of exist for all n. We show that admits a nomadic near-Hamiltonian decomposition when .     

Nowhere-zero 4-flow in almost Petersen-minor free graphs

Tutte [W.T. Tutte, On the algebraic theory of graph colorings, J. Combin. Theory 1 (1966) 15-20] conjectured that every bridgeless Petersen-minor free graph admits a nowhere-zero 4-flow. Let be the graph obtained from the Petersen graph by contracting μ edges from a perfect matching. In this paper we prove that every bridgeless -minor free graph admits a nowhere-zero 4-flow.     

Nowhere-zero 3-flows of highly connected graphs

Let G be a k-edge-connected graph of order n. If k4log₂ n then G has a nowhere-zero 3-flow.     

On degree sequences of undirected,directed, and bidirected graphs

Bidirected graphs generalize directed and undirected graphs in that edges are oriented locally at every node. The natural notion of the degree of a node that takes into account (local) orientations is that of net-degree. In this paper, we extend the following four topics from (un)directed graphs to bidirected graphs:

• –Erdős–Gallai-type results: characterization of net-degree sequences,
• –Havel–Hakimi-type results: complete sets of degree-preserving operations,
• –Extremal degree sequences: characterization of uniquely realizable sequences, and
• –Enumerative aspects: counting formulas for net-degree sequences.

To underline the similarities and differences to their (un)directed counterparts, we briefly survey the undirected setting and we give a thorough account for digraphs with an emphasis on the discrete geometry of degree sequences. In particular, we determine the tight and uniquely realizable degree sequences for directed graphs.     

On the representability of totally unimodular matrices on bidirected graphs

We present a polynomial time algorithm to construct a bidirected graph for any totally unimodular matrix B by finding node-edge incidence matrices Q and S such that QB=S. Seymour’s famous decomposition theorem for regular matroids states that any totally unimodular (TU) matrix can be constructed through a series of composition operations called k-sums starting from network matrices and their transposes and two compact representation matrices B₁,B₂ of a certain ten element matroid. Given that B₁,B₂ are binet matrices we examine the k-sums of network and binet matrices. It is shown that thek-sum of a network and a binet matrix is a binet matrix, but binet matrices are not closed under this operation for k=2,3. A new class of matrices is introduced, the so-called tour matrices, which generalise network, binet and totally unimodular matrices. For any such matrix there exists a bidirected graph such that the columns represent a collection of closed tours in the graph. It is shown that tour matrices are closed under k-sums, as well as under pivoting and other elementary operations on their rows and columns. Given the constructive proofs of the above results regarding the k-sum operation and existing recognition algorithms for network and binet matrices, an algorithm is presented which constructs a bidirected graph for any TU matrix.     

Nowhere-zero 3-flows in Cayley graphs on generalized dihedral group and generalized quaternion group

Tutte conjectured that every 4-edge-connected graph admits a nowhere-zero 3-flow. In this paper, we show that this conjecture is true for Cayley graph on generalized dihedral groups and generalized quaternion groups, which generalizes the result of F. Yang and X. Li [Inform. Process. Lett., 2011, 111: 416–419]. We also generalizes an early result of M. Nánásiová and M. ?koviera [J. Algebraic Combin., 2009, 30: 103–110].     

Multicommodity flows in graphs

Suppose that G is a graph, and (s_i,t_i) (1≤i≤k) are pairs of vertices; and that each edge has a integer-valued capacity (≥0), and that q_i≥0 (1≤i≤k) are integer-valued demands. When is there a flow for each i, between s_i and t_i and of value q_i, such that the total flow through each edge does not exceed its capacity? Ford and Fulkerson solved this when k=1, and Hu when k=2. We solve it for general values of k, when G is planar and can be drawn so that s₁,…, s_l, t₁, …, t_l,…,t_l are all on the boundary of a face and s_l+1, …,S_k, t_l+1,…,t_k are all on the boundary of the infinite face or when t₁=?=t_l and G is planar and can be drawn so that s_l+1,…,s_k, t₁,…,t_k are all on the boundary of the infinite face. This extends a theorem of Okamura and Seymour.     

Walks and regular integral graphs

We establish a useful correspondence between the closed walks in regular graphs and the walks in infinite regular trees, which, after counting the walks of a given length between vertices at a given distance in an infinite regular tree, provides a lower bound on the number of closed walks in regular graphs. This lower bound is then applied to reduce the number of the feasible spectra of the 4-regular bipartite integral graphs by more than a half.Next, we give the details of the exhaustive computer search on all 4-regular bipartite graphs with up to 24 vertices, which yields a total of 47 integral graphs.     

Multicommodity flows in cycle graphs

This paper considers the multicommodity flow problem and the integer multicommodity flow problem on cycle graphs. We present two linear time algorithms for solving each of the two problems.     

Bounded-excess flows in cubic graphs

An (r, α)-bounded-excess flow ((r, α)-flow) in an orientation of a graph G = (V, E) is an assignment f : E → [1, r−1], such that for every vertex x ∈ V, $|{\sum }_{e\in E^{+}\left(x\right)}f\left(e\right)-{\sum }_{e\in E^{-}\left(x\right)}f\left(e\right)|\le \alpha$. E⁺(x), respectively E⁻(x), is the set of edges directed from, respectively toward x. Bounded-excess flows suggest a generalization of Circular nowhere-zero flows (cnzf), which can be regarded as (r, 0)-flows. We define (r, α) as Stronger or equivalent to (s, β), if the existence of an (r, α)-flow in a cubic graph always implies the existence of an (s, β)-flow in the same graph. We then study the structure of the bounded-excess flow strength poset. Among other results, we define the Trace of a point in the r − α plane by $tr\left(r,\alpha \right)=\frac{r-2\alpha }{1-\alpha }$ and prove that among points with the same trace the stronger is the one with the smaller α (and larger r). For example, if a cubic graph admits a k-nzf (trace k with α = 0), then it admits an $(r,\frac{k-r}{k-2})$-flow for every r, 2 ≤ r ≤ k. A significant part of the article is devoted to proving the main result: Every cubic graph admits a $(3\frac{1}{2},\frac{1}{2})$-flow, and there exists a graph which does not admit any stronger bounded-excess flow. Notice that $tr(3\frac{1}{2},\frac{1}{2})=5$ so it can be considered a step in the direction of the 5-flow Conjecture. Our result is the best possible for all cubic graphs while the seemingly stronger 5-flow Conjecture relates only to bridgeless graphs. We also show that if the circular-flow number of a cubic graph is strictly less than 5, then it admits a $(3\frac{1}{3},\frac{1}{3})$-flow (trace 4). We conjecture such a flow to exist in every cubic graph with a perfect matching, other than the Petersen graph. This conjecture is a stronger version of the Ban-Linial Conjecture [1]. Our work here strongly relies on the notion of Orientable k-weak bisections, a certain type of k-weak bisections. k-Weak bisections are defined and studied by L. Esperet, G. Mazzuoccolo, and M. Tarsi [4].     

Constructably Laplacian integral graphs

A graph is Laplacian integral if the spectrum of its Laplacian matrix consists entirely of integers. We consider the class of constructably Laplacian integral graphs - those graphs that be constructed from an empty graph by adding a sequence of edges in such a way that each time a new edge is added, the resulting graph is Laplacian integral. We characterize the constructably Laplacian integral graphs in terms of certain forbidden vertex-induced subgraphs, and consider the number of nonisomorphic Laplacian integral graphs that can be constructed by adding a suitable edge to a constructably Laplacian integral graph. We also discuss the eigenvalues of constructably Laplacian integral graphs, and identify families of isospectral nonisomorphic graphs within the class.