Routing plays a pivotal role in quantum communication as it directly impacts the efficiency, reliability, and scalability of quantum networks. While several studies in the literature have explored routing algorithms leveraging quantum memories, current quantum memory technologies are unable to simultaneously achieve high fidelity, extended storage durations, wide bandwidths, multimode capacity, and high efficiency. To address this limitation, our study focuses on fidelity-guaranteed entanglement routing within a memoryless network architecture, employing both distributed and centralized routing approaches. Utilizing our proposed routing algorithm, MEFID, we achieved a throughput of 72 qubits per second under a fidelity threshold of 0.8 and within three iterative rounds. By integrating a purification process to ensure that the final fidelity consistently exceeds the specified threshold, our algorithm facilitates the development of robust and high-performance quantum networks.

In this paper, we consider the problem of concurrent multipath routing and end-to-end entanglement distribution for online resource allocation in quantum networks. We propose a heuristic algorithm to solve this problem, considering quantum memory, decoherence time, entanglement distribution probability, and fidelity. A time-slotted quantum network operation model is considered based on the cut-off decoherence time of quantum memories. The proposed heuristic is designed for a quantum network with noisy intermediate scale quantum (NISQ) constraints, including fixed quantum memory decoherence time, and probabilistic entanglement generation and swapping. It considers integer linear programming (ILP)-based and heuristic approaches to select multiple paths and resource allocation in the network. Simulations are performed to evaluate the performance of ILP and heuristic based approaches in terms of requests using multipath approach, blocking ratio, and computation time in a small sized quantum network with few requests. Next, the performance of heuristic based approaches are evaluated for a large problem size. The obtained results ensure that performance of ILP and heuristic based approaches are comparable, and multipath routing outperforms single path routing in terms of blocking ratio.

Due to cost and service level agreement, upgrading network infrastructure is always challenging for network operators, particularly for backbone optical networks (ONs). Nowadays, to support several bandwidth-hungry applications, network operators adopt the migration process from fixed-grid nodes to flex-grid in backbone ONs. However, without proper planning, it may lead to inefficient resource utilization. This paper proposes optimization models for resource allocation in mixed-grid ONs, which initially identifies the set of eligible nodes for upgradation from fixed-grid to flex-grid capabilities for a given set of offline traffic and then uses them in the network to maximize the resource utilization. We formulate an integer linear programming (ILP) problem, named Opti-node, to determine the set of nodes selected for fixed-grid to flex-grid conversion in mixed-grid ONs. We then introduce an ILP, named Opti-resource, to enhance the resource utilization based on the determined set of nodes converted from fixed-grid to flex-grid using Opti-node. We prove that the decision problem of resource allocation in mixed-grid ONs is NP-Complete. Heuristic approaches based on the ILP-based approach and simulated annealing are introduced for a large problem size when Opti-resource is not tractable. We evaluate the proposed resource allocation model extensively, considering different networks, and observe that the simulated annealing-based approach performs better in terms of blocking ratio and spectrum utilization than the ILP-based heuristic approach, but it requires additional computation time.

Optical regenerators are beneficial in resource utilization as they provide additional functionalities, such as modulation format (MF) and spectrum conversion, besides signal regeneration. In spectrally-spatially elastic optical networks (SS-EONs), regenerators can perform core and mode switching, which further improves the spectrum utilization. For the first time, this paper proposes a regenerator-aware routing, spectrum, core, and mode allocation model while avoiding inter-mode and inter-core crosstalks during resource allocation to enhance the spectrum utilization in SS-EONs. The proposed model performs core/mode switching operations at the regeneration sites along with spectrum and MF conversions. Apart from the regeneration sites, the proposed model maintains the spectrum continuity, spectrum contiguity, core continuity, and mode continuity constraints in the remaining intermediate nodes. We model the regenerator-aware resource allocation as an integer linear programming to minimize the highest utilized spectrum slot index under the condition that a limited number of regenerators with their placement are given in the network. We introduce a heuristic when the optimization problem is not tractable. Numerical results indicate that the proposed model improves resource utilization compared to a benchmark model that does not consider core and mode switching.

To accommodate the exponential growth of the inter-network services, infrastructure as a service (IaaS) allows the different parties to share the physical infrastructure of the optical network resources using network virtualization. The virtual optical network embedding (VONE) enables efficient resource virtualization to map several virtual optical network (VON) requests over a substrate optical network. On the other hand, the spectrally-spatially elastic optical network (SS-EONs) is becoming a promising solution for the increasing volume of the demanded traffic due to its higher fiber capacity. For the first time, this paper proposes a routing, spectrum, core, and mode allocation (RSCMA) model for VONE over SS-EONs while avoiding inter-core and inter-mode crosstalks. The proposed model allocates spectrum for VON requests over the substrate SS-EONs while maintaining the spectrum continuity, spectrum contiguity, core continuity, and mode continuity constraints. We model the RSCMA model for the VONE problem over SS-EONs as an integer linear programming (ILP) to maximize the number of VON requests placed in the substrate network. We introduce a heuristic when the optimization problem is not tractable for large networks. Numerical results indicate that the performances of the ILP and heuristic approaches are comparable and the computation time of the heuristic approach is much smaller than that of the ILP approach.

Nowadays, quantum-key-distribution (QKD) gains popularity for providing high-security in high-capacity spectrally-spatially elastic optical network (SS-EONs). Considering QKD in SS-EONs requires addition resources for classical and quantum channels and introduces addition noises. The key challenge during resource allocation in QKD-enabled SS-EONs is to deal with the background noises generated by classical and quantum channels along with inter-core and inter-mode XT while reducing the highest utilized spectrum slot index. For the first time, this paper proposes a routing, spectrum, core, and mode allocation (RSCMA) model in QKD-enabled SS-EONs while avoiding intercore and inter-mode XT and the noises generated by the classical and quantum channels. The proposed model allocates spectrum for the data and classical channels for each request maintaining the spectrum continuity and contiguity constraints. It selects a dedicated mode for the quantum channels along the routing path for each request. We model the QKD-enabled resource allocation as an integer linear programming (ILP) to minimize the highest utilized spectrum slot index. We introduce a heuristic when the optimization problem is not tractable for large networks. Numerical results indicate that the values of the highest utilized spectrum slot index using the ILP and heuristic approaches are comparable and the computation time of the heuristic approach is much smaller than that of the ILP approach.

In this paper, we have proposed two new Routing, spectrum and core allocation (RSCA) schemes for offline survivable space division multiplexing elastic optical network (SDM-EON) supporting multi-class (mc) traffic in temporal domain. Multipath based protection scheme is applied to ensure survivability against single link failure, thus allowing traffic for each connection to be split amongst those paths. Advanced reservation (AR), being a part of multi-class traffic can withstand some delay from its time of arrival. Two schemes are designed based on the service of AR connections: all the split traffic in the multiple paths for a connection at the source node are set to start (i) at the same time strictly (mc-RSCA-S) and (ii) in a relaxed way by allowing different permissible times (mc-RSCA-R). Mixed integer linear programming (MILP) models are formulated for both the schemes. RSCA problem being an NP-hard one, those MILPs are not tractable and two heuristics with polynomial time complexity are developed. Two heuristics are based on a greedy approach and a simulated annealing (SA) based approach. Simulation studies of MILPs and the respective heuristics confirm that near optimal solutions are obtained by the proposed heuristics in identical experimental setup. It is observed that with an increase in the number of cores as well as in the percentage of AR connections in the incoming request set, spectrum utilization in the network and fragmentation of cores get reduced more by SA based approach compared to greedy approach.