--- title: Quantum Connectivity Measure (QCM) url: https://www.emergentmind.com/topics/quantum-connectivity-measure-qcm type: topic --- # Quantum Connectivity Measure (QCM) Searching arXiv for recent and foundational papers on “Quantum Connectivity Measure” and related formulations. Quantum Connectivity Measure (QCM) does not denote a single universally standardized quantity across the literature. Instead, it refers to a family of quantum connectivity constructs that quantify connectedness, effective link quality, or multipartite connectivity structure in distinct settings: quantum networks, quantum graphs, graph states, entanglement-percolation models, and entanglement classification. In the most explicit network-level usage, QCM is introduced as a measure of the average connection quality between pairs of network nodes, filtered by a task threshold [2603.29601]. In other lines of work, closely related quantities are defined through algebraic connectivity in quantum key distribution networks [2004.11374], through final-state entanglement multiplied by network integrity in entanglement percolation [2502.17191], through operator-algebraic connectedness and separator rank in quantum graphs [1911.00446; 2505.22519], through correlator signatures in graph states [2308.07690], and through the connectivity order of entangled states and measurement devices [1407.5920]. The term therefore designates a research theme rather than a single invariant, with each formulation tailored to the operational notion of connectivity relevant to the underlying model. ## 1. Network-level QCM as functional quantum connectivity The clearest explicit definition of QCM appears in the theory of quantum networks, where classical path existence is distinguished from the ability to establish entanglement of sufficient quality to execute a target quantum information processing task [2603.29601]. In that framework, QCM for a node subset \(\mathcal{N}\subseteq V\) is defined by \[ \mathcal{Q}_\mathcal{N}(\{\mathcal{S}_{ij}\}_{i,j \in \mathcal{N}) := \frac{1}{\mathcal{N}_P} \sum_{i,j \in \mathcal{N} \mathcal{S}_{ij}~\Theta[\mathcal{S}_{ij}-\epsilon]. \] Here \(\mathcal{N}_P=|\mathcal{N}|(|\mathcal{N}|-1)/2\) is the number of node pairs, \(\mathcal{S}_{ij}\) is the effective connection strength between nodes \(i\) and \(j\), \(\Theta[\mathcal{S}_{ij}-\epsilon]\) is the Heaviside step function, and \(\epsilon\in[0,1]\) is the task threshold. The supplementary presentation gives the same definition as \[ \mathcal{Q}_\mathcal{N}(\{\mathcal{S}_{ij}\}) = \frac{2}{|\mathcal{N}|(|\mathcal{N}|-1)} \sum_{i,j \in \mathcal{N} \mathcal{S}_{ij}\,\Theta(\mathcal{S}_{ij}-\epsilon). \] The effective connection strength is protocol-dependent. For a pair \(i,j\), if \(\mathcal{P}_{ij}\) denotes the optimal path and \(\mathcal{D}\) the protocol map from edge parameters to end-to-end connection strength, then \[ \mathcal{S}_{ij}(\{\mu_{k}\}_{k\in\mathcal{P}_{ij}) := \mathcal{D}(\{\mu_{k}\}_{k\in\mathcal{P}_{ij})}. \] For pure bipartite states with entanglement swapping, the map is multiplicative in concurrence: \[ \mathcal{S}_{ij}=\prod_{k\in\mathcal{P}_{ij}} c_k, \qquad \mathcal{D}(\{c_{k}\}_{k=1}^{\ell}) = \prod_{k=1}^{\ell} c_k. \] This formulation makes QCM a measure of average effective entanglement connection strength, but only counting node pairs whose end-to-end connection exceeds the threshold required by the task [2603.29601]. It is therefore neither a purely topological observable nor merely a mean entanglement statistic. It combines path structure, edge-resource quality, and task feasibility in one quantity. The same work introduces two naturally derived companions. The quantum-connected fraction (QCF) is \[ \mathcal{F}_\mathcal{N} = \frac{1}{\mathcal{N}_P} \sum_{i,j \in \mathcal{N} } \Theta[\mathcal{S}_{ij}-\epsilon], \] or equivalently in the supplementary normalization, \[ \mathcal{F}_\mathcal{N} = \frac{2}{|\mathcal{N}|(|\mathcal{N}|-1)} \sum_{i,j \in \mathcal{N} } \Theta(\mathcal{S}_{ij}-\epsilon). \] QCF counts the fraction of pairs above threshold, whereas QCM weights those pairs by their connection strength. The quantum clustering coefficient (QCC) is defined locally by applying QCM to the neighbor set \(\mathcal{N}^{(i)}\): \[ \mathcal{C}_Q(i) := \mathcal{Q}_{\mathcal{N}^{(i)}}. \] The distinction from classical connectivity metrics is central. A network can be topologically connected yet functionally disconnected for quantum tasks if entanglement quality is too low. In the homogeneous fully connected case, \[ \overline{\mathcal{Q}^{(G)}} = \begin{cases} \overline{c}, & \text{for } \overline{c} > \epsilon \\ 0, & \text{for } \overline{c} \leq \epsilon, \end{cases} \qquad \overline{\mathcal{F}^{(G)}} = \begin{cases} 1, & \text{for } \overline{c} > \epsilon \\ 0, & \text{for } \overline{c} \leq \epsilon. \end{cases} \] Thus even a complete graph can be functionally disconnected when average edge concurrence is below threshold [2603.29601]. This directly addresses a common misconception: topological completeness does not imply quantum-functional connectivity. ## 2. Alternative network metrics described as quantum connectivity measures A different network-level construction appears in preliminary work on quantum key distribution networks, where a quantum-network connectivity measure is obtained by embedding quantum physical parameters into a weighted graph Laplacian and then using weighted algebraic connectivity [2004.11374]. The core metric is \[ \lambda_{1}^{t} = \min_{y} y^{T} L\big(V(R^{t}), E(C^{t})\big) y \quad \text{subject to } y^{T}y = 1,\; y^{T}\mathbf{1}=0. \] This is the weighted algebraic connectivity of a graph whose edge weights reflect channel capacity or link quality and whose vertices encode node properties through \(R^t\). The physical layer is expressed through the quantities \(f_{ref}\), \(\mu\), \(t_{link}\), \(\eta\), \(p_{opt}\), \(p_{dark}\), and \(n\). The paper defines \[ R_{raw} = f_{ref}\mu t_{link}\eta, \qquad R_{sift} = \frac{1}{2}f_{ref}\mu t_{link}\eta, \] with error model \[ R_{err} = R_{opt} + R_{det}, \qquad R_{opt} = R_{sift}p_{opt}, \qquad R_{det} = \frac{1}{4} f_{ref} p_{dark} n, \] leading to \[ QBER = p_{opt} + \frac{1}{2}\frac{p_{dark}n}{\mu t_{link}\eta}, \qquad R = 1 - 2h(QBER). \] The link-quality update is smoothed through \[ t_{link} = W_{t+1} = \alpha \, w_t + (1-\alpha)W_t, \] with \[ w_t = \frac{\left\{ R_s < C \right\}_t} {\left\{ R_s < C \right\}_t + \left\{ R_s \ge C \right\}_t}. \] Here the connectivity metric indicates transmission throughput in a grid quantum network and is used to study how tuning physical parameters affects resilience under node or edge removal [2004.11374]. The use of the Fiedler value places this QCM variant closer to weighted spectral graph theory than to end-to-end entanglement-quality averaging. A plausible implication is that the literature contains at least two distinct network-level QCM paradigms: thresholded pairwise connection-quality averaging [2603.29601] and weighted algebraic connectivity derived from quantum-physical link models [2004.11374]. A third network quantity close to a QCM is introduced for entanglement percolation in qubit-based planar quantum networks [2502.17191]. There, the key connectivity metric is \[ K(A, B) := E\left(\ket{\lambda_{AB}}\right)\cdot\mathcal{I}(A,B), \] where \[ E(\ket{\lambda}) = 2(1-\lambda), \qquad \mathcal{I}(A,B) := \frac{d(A,B)}{\mathcal{N}(A,B)}. \] The distance is defined operationally by the minimum number of swapping operations: \[ d = n_s + 1. \] This metric combines final-state entanglement and network integrity after the percolation process. It is explicitly used to classify percolation strategies and identify distinct regimes separated by threshold values of the initial Schmidt parameter \(\lambda\) [2502.17191]. Relative to the QCM of [2603.29601], this quantity is pair-specific rather than network-averaged and incorporates resource destruction directly through \(\mathcal{I}(A,B)\). ## 3. Quantum graphs: connectivity witnesses, separator rank, and Laplacians In quantum graph theory, QCM is not usually defined as a named scalar, but the literature provides several mathematically precise connectivity diagnostics that serve the same role [1911.00446; 2505.22519]. In the operator-system model, a quantum graph on \(M_n\) is a linear subspace \(\mathcal{S}\subseteq M_n\) that is closed under adjoints and contains the identity: \[ \mathcal{S} = \mathcal{S}^\dagger,\qquad I_n\in \mathcal{S}. \] Connectedness is defined by \[ \mathcal{S}\subseteq M_n \text{ is connected if there exists } m\in\mathbb{N} \text{ such that } \mathcal{S}^m=M_n. \] The central equivalent criterion is that \(\mathcal{S}\) is disconnected if and only if there exists a nontrivial projection \(P\in M_n\) such that \[ P\mathcal{S}(I_n-P)=\{0\}. \] This is the main connectivity witness in the operator-system setting [1911.00446]. The same paper defines a separator: a projection \(P\in M_n\) is a separator of \(\mathcal{S}\) if \((I_n-P)\mathcal{S}(I_n-P)\) is either disconnected or \(1\)-dimensional. A quantum graph is \(k\)-connected if every separator has rank at least \(k\). For a classical graph \(G\) and its quantum graph \(\mathcal{S}_G\), \[ G \text{ is \(k\)-connected } \iff \mathcal{S}_G \text{ is \(k\)-connected}. \] The paper explicitly identifies the minimum rank of a separator, or equivalently the largest \(k\) such that \(\mathcal{S}\) is \(k\)-connected, as the natural quantum analogue of classical vertex connectivity [1911.00446]. This suggests a QCM in the sense of a separator-rank invariant. The quantum adjacency matrix framework extends these ideas to general quantum graphs and to the non-tracial setting [2505.22519]. There, a quantum graph is \(G=(M,\psi,A)\), with \(A:\mathcal{M}\to\mathcal{M}\) a quantum Schur idempotent: \[ m(A\otimes A)m^* = A. \] In the tracial setting, an undirected quantum graph is connected if the quantum adjacency matrix \(A\) does not commute with any non-trivial projection \(p\in M\): \[ L_p A = A L_p \implies p=0 \text{ or } \mathds{1}. \] In the non-tracial setting, connectivity is reformulated using the KMS inner product and the KMS implementation \(\widetilde A\), for example through the condition that no non-trivial projection \(p\in M\) satisfies \[ \lambda(p)A = A\lambda(p), \] equivalently \[ \rho(p)A = A\rho(p). \] The same work establishes the equivalence \[ A \text{ is irreducible} \iff G \text{ is connected}, \] and characterizes connectivity through the graph Laplacian \(\Delta\): an undirected quantum graph is connected if and only if \(0\) is a simple eigenvalue of \(\Delta\) [2505.22519]. These results indicate several possible QCM-like choices: irreducibility, nullity of the Laplacian, or support growth of powers \(A^k\). The literature therefore supports a structural rather than uniquely numerical interpretation of QCM in quantum graph theory. ## 4. State-based and correlator-based measures of connectivity For graph states, connectivity can be probed operationally through Pauli correlators rather than through a global scalar [2308.07690]. The state preparation is \[ |G(\varphi)\rangle = \prod_{(a,b)\in E} U_{ab}(\varphi)\,|\Psi\rangle, \qquad |\Psi\rangle = |+\rangle^{V}, \qquad |+\rangle=\frac{1}{\sqrt{2}}(|0\rangle+|1\rangle), \] with genuine graph states corresponding to \(\varphi=\pi\). The neighborhood notation is \[ N(\nu)=\{\text{first neighbors of }\nu\},\qquad n_\nu=|N(\nu)|. \] The central result is that specific Pauli correlators are nonzero only when the graph has specific connectivity relations. Exact formulas include \[ \langle G|\sigma_x^\nu\sigma_x^\mu|G\rangle = \begin{cases} 1 & \text{if } N(\nu)=N(\mu),\\ 0 & \text{else,} \end{cases} \] so the \(xx\) correlator detects twin vertices; \[ \langle G|\sigma_x^\nu\sigma_z^\mu|G\rangle = \begin{cases} 1 & \text{if } N(\nu)=\{\mu\},\\ 0 & \text{else,} \end{cases} \] so the \(xz\) correlator detects a leaf vertex; and \[ \langle G|\sigma_y^\nu\sigma_y^\mu|G\rangle = \begin{cases} 1 & \text{if } N(\nu)\cup\{\nu\}=N(\mu)\cup\{\mu\},\\ 0 & \text{else,} \end{cases} \] so the \(yy\) correlator detects adjacent twins [2308.07690]. The same work gives a direct adjacency test after measuring all other qubits in the \(z\)-basis: \[ \bra{G}\left(\prod_{\mu\neq a,b}P_{z\pm}^\mu\right)\sigma_y^a\sigma_y^b \left(\prod_{\mu\neq a,b}P_{z\pm}^\mu\right)\ket{G} = \begin{cases} \pm 1 & \text{if } b\in N(a)\; (\Leftrightarrow a\in N(b)),\\ 0 & \text{else.} \end{cases} \] This is an operational connectivity witness rather than an aggregate scalar. The paper emphasizes that for genuine graph states these correlators take only the values \[ -1,\;0,\;1, \] which simplifies data analysis [2308.07690]. A different state-based classification appears in the theory of connectivity structures of quantum entanglement, where the “connectivity order” \(\Omega\) of a quantum state or a measurement device plays the role of a QCM [1407.5920]. For a finite set of subsystem indices \[ I=\{1,\dots,k\}, \] and a pure state \[ \psi\in H=\bigotimes_{i\in I} H_i, \] the paper constructs connectivity structures from the behavior of the state under determining measurements on complementary subsystems. Several families of subsets are introduced, including \[ GI(\psi),\; BIP(\psi),\; MT(\psi),\; IP(\psi),\; ML(\psi),\; NCS(\psi), \] with corresponding generated integral connectivity structures \[ \kappa_{GI}(\psi)=[GI(\psi)]_1,\qquad \kappa_{BIP}(\psi)=[BIP(\psi)]_1, \] \[ \kappa_{MT}(\psi)=[MT(\psi)]_1,\qquad \kappa_{IP}(\psi)=[IP(\psi)]_1, \] \[ \kappa_{ML}(\psi)=[ML(\psi)]_1,\qquad \kappa_{NCS}(\psi)=[NCS(\psi)]_1. \] The connectivity order \(\Omega(X)\) is the height of the acyclic directed graph formed by irreducible connected subsets ordered by inclusion, and the overall connectivity order of a pure state is \[ \Omega(\psi)=\max\big(\Omega_c(\psi),\Omega_f(|\psi\rangle\langle\psi|)\big). \] This yields, for example, \[ \Omega(EPR)=1,\qquad \Omega(GHZ)=1,\qquad \Omega(O_2)=2 \] for the specific states discussed in the paper [1407.5920]. In this setting QCM classifies entanglement by hierarchical connectivity complexity rather than by link quality or operator irreducibility. ## 5. Algorithmic and spectroscopic procedures that implicitly support a QCM Some papers do not define a scalar called QCM but supply measurement protocols whose outputs encode connectivity and could serve as its operational basis. One example is a constant-measurement quantum algorithm for graph connectedness [2411.15015]. There, the input undirected graph \(\mathcal{G}=(N,E)\) is mapped to qubits and non-unitary two-qubit \(Z\) spiders from ZX calculus. Starting from \(|0_n\rangle\), the algorithm prepares a post-encoding state of the form \[ |0_n\rangle \to \bigotimes_k |\mathrm{GHZ}\rangle_k, \] where each factor corresponds to one connected component. Connectivity is thereby represented by the presence or absence of GHZ-like entanglement across qubits. The algorithm uses the fusion rule of ZX calculus so that all spiders belonging to the same connected component contract into a single larger spider. Two measurements are sufficient to distinguish the connected case from the disconnected case with success probability \[ p(\text{success})=1-\frac14=\frac34. \] For a disconnected graph with \(k\) components, the probability of failure after \(M\) measurements is \[ p(\text{disconnected},\text{failure})=\frac{1}{k^M}, \] and for a connected graph, \[ p(\text{connected},\text{failure})=\frac{1}{2^M}. \] The paper does not define a scalar QCM, but it explicitly identifies connectivity-related observables such as whether all qubits yield the same bitstring under repeated measurement, the XOR discrepancy \[ a\veebar b= \begin{cases} 1 & a\ne b\\ 0 & a=b, \end{cases} \] and the success/failure probabilities as functions of component number \(k\) and measurement count \(M\) [2411.15015]. This suggests a measurement-based QCM grounded in multipartite GHZ structure. Another example is a local probe for connectivity in quantum complex networks [1708.09625]. The network is a system of identical quantum harmonic oscillators with Hamiltonian \[ H_E=\frac{p^T p}{2}+\frac{q^T(\omega_0^2 I + gL)q}{2}, \] where \(L\) is the graph Laplacian. A local oscillator probe \[ H_S=\frac{p_S^2+\omega_S^2 q_S^2}{2} \] is coupled to one node \(j\) through \[ H_I = -k q_S q_j. \] By sweeping \(\omega_S\) and monitoring the probe’s mean excitation number \[ n(t) = \langle a_S^\dagger(t)a_S(t)\rangle, \] one detects resonances with network normal-mode frequencies \(\Omega_i\), which satisfy \[ \lambda_i = \frac{\Omega_i^2-\omega_0^2}{g}. \] The Laplacian spectrum then constrains the degree sequence through \[ \mathrm{Tr}L = \sum_{i=1}^N d_i = \sum_{i=1}^N \lambda_i, \] \[ \mathrm{Tr}L^2 - \mathrm{Tr}L = \sum_{i=1}^N d_i^2 = \sum_{i=1}^N(\lambda_i^2-\lambda_i), \] and \[ 1+\sum_{i=1}^{m