Neural Routing Research Articles

GROM: A generalized routing optimization method with graph neural network and deep reinforcement learning

Routing optimization, as a significant part of Traffic Engineering (TE), plays an important role in balancing network traffic and improving quality of service. With the application of Machine Learning (ML) in various fields, many neural network-based routing optimization solutions have been proposed. However, most existing ML-based methods need to retrain the model when confronted with a network unseen during training, which incurs significant time overhead and response delay. To improve the generalization ability of the routing model, in this paper, we innovatively propose a routing optimization method GROM which combines Deep Reinforcement Learning (DRL) and Graph Neural Networks (GNN), to directly generate routing policies under different and unseen network topologies without retraining. Specifically, for handling different network topologies, we transform the traffic-splitting ratio into element-level output of GNN model. To make the DRL agent easier to converge and well generalize to unseen topologies, we discretize the huge continuous traffic-splitting action space. Extensive simulation results on five real-world network topologies demonstrate that GROM can rapidly generate routing policies under different network topologies and has superior generalization ability.

Neural Network Multipath Routing in Software Defined Networks Based on Genetic Algorithm

Currently, a wide demand for the implementation and use of various cloud solutions is a modern trend and the driving force behind the development of network technologies. The growth of cloud application services delivered through data centers with varying network traffic needs demonstrates the limitations of traditional routing and load balancing methods. The combination of the advantages of software defined networks (SDN) technology and artificial intelligence (AI) methods ensures efficient management and operation of computer network resources. The paper proposes an approach to neural network multipath routing in SDN based on a genetic algorithm. The architecture and model of an artificial neural network has been developed to solve the problem of multipath routing in the SDN, which is able to predict the shortest paths based on the metrics of communication links. To optimize the hyperparameters of the neural network model, it is proposed to use a modified genetic algorithm. A visual software system SDNLoadBalancer has been developed and an experimental SDN topology has been designed, which makes it possible to study in detail the processes of neural network multipath routing in SDN based on the proposed approach. The obtained results show that the proposed neural network model has the ability to predict routes with high accuracy in real time, which makes it possible to implement various load balancing schemes in order to increase performance of SDN.

Lifetime Extension Approach Based on the Levenberg–Marquardt Neural Network and Power Routing of DC–DC Converters

The power conversion system based on the modular connection has widespread applications in various power electronic systems. To accurately estimate the state of health without recognizing the systematic mathematical model and to extend the lifetime, this paper proposes a lifetime extension approach based on the Levenberg-Marquardt back propagation neural network (LM-BPNN) and power routing of interleaved DC-DC boost conversion systems. The LM-BPNN model is constructed based on the voltage, current, and temperature data generated by the system. On the basis of the trained LM-BPNN, the real-time cumulated damage estimation of each power cell in the conversion system can be accomplished. Applying the power routing concept, the DC-DC boost conversion system allocates different power to the cells according to the cumulated damage of each cell, thereby delaying the failure of cells with higher cumulated damage. Numerical simulation results show that the proposed lifetime extension approach can extend the overall system lifetime. Furthermore, an experimental setup of the interleaved DC-DC boost conversion is constructed to verify the proposed approach, which is of great significance for predictive maintenance in the industrial system.

G-Routing: Graph Neural Networks-Based Flexible Online Routing

G-Routing: Graph Neural Networks-Based Flexible Online Routing

Retracted: Neural Network Optimal Routing Algorithm Based on Genetic Ant Colony in IPv6 Environment.

[This retracts the article DOI: 10.1155/2021/3115704.].

Towards Neural Routing with Verified Bounds on Performance

When data-driven algorithms, especially the ones based on deep neural networks (DNNs), replace classical ones, their superior performance often comes with difficulty in their analysis. On the way to compensate for this drawback, formal verification techniques, which can provide reliable guarantees on program behavior, were developed for DNNs. These techniques, however, usually consider DNNs alone, excluding real-world environments in which they operate, and the applicability of techniques that do account for such environments is often limited. In this work, we consider the problem of formally verifying a neural controller for the routing problem in a conveyor network. Unlike in known problem statements, our DNNs are executed in a distributed context, and the performance of the routing algorithm, which we measure as the mean delivery time, depends on multiple executions of these DNNs. Under several assumptions, we reduce the problem to a number of DNN output reachability problems, which can be solved with existing tools. Our experiments indicate that sound-and-complete formal verification in such cases is feasible, although it is notably slower than the gradient-based search of adversarial examples.The paper is structured as follows. Section 1 introduces basic concepts. Then, Section 2 introduces the routing problem and DQN-Routing, the DNN-based algorithm that solves it. Section 3 proposes the contribution of this paper: a novel sound and complete approach to formally check an upper bound on the mean delivery time of DNN-based routing. This approach is experimentally evaluated in Section 4. The paper is concluded with some discussion of the results and outline of possible future work.

Correction to: Deep Neural Network Routing with Dynamic Space Division for 3D UAV FANETs

Correction to: Deep Neural Network Routing with Dynamic Space Division for 3D UAV FANETs

Loading Cost-Aware Model Caching And Request Routing In Edge-enabled Wireless Sensor Networks

Abstract Existing works on caching in multi-access edge computing focus on service caching and request routing. However, loading cost and execution time influenced by resource sharing have not been well exploited. To fill this gap, we investigate the joint optimization problem over deep neural network (DNN) model caching and DNN request routing with edge collaboration in edge-enabled wireless sensor networks. A problem is formulated, with the objective of maximizing throughput, under constraints of budget, accuracy and latency etc. The proof of NP-hardness for the formulated problem is provided. To solve the problem, an approximation algorithm based on randomized rounding is presented. In addition, the approximation ratio for the presented algorithm is proved to be $1/(1-\sqrt{4\ln S/\xi^\dagger})$, where $S$ is the number of edge servers and $\xi^\dagger$ is the objective value from linear relaxation. Extensive experiments demonstrate that the system throughput for the presented algorithm can be improved by 58.8% on average, compared with that of the baseline algorithm.

Neural Network-Based Routing Energy-Saving Algorithm for Wireless Sensor Networks.

With the evolvement, standards have changed, mobile Internet technology has also been upgraded, and it has also driven the development of smart objects mobile. With the continuous development of smart objects mobile, the bottleneck of small node size and low battery energy storage has not been solved in the end, which makes the research of wireless sensor network energy-saving technology become the focus, and the improvement of routing technology is an effective way to improve energy-saving technology. From the data transmission energy consumption of smart objects mobile, the routing algorithm of smart objects mobile is discussed and analyzed and the classical representative LEACH is the object of in-depth research. Routing algorithms can easily and reliably process network data and make the network work well and are widely used in highly secure military systems and smaller commercial networks. Aiming at these deficiencies, a corresponding improved algorithm is proposed, and it is tested through simulation and specific experiments to verify the correctness and the system's reliability. The SMPSO-BP algorithm converges when the number of iterations is about 600, which is earlier than the LEACH algorithm and the improved LEACH algorithm, so the SMPSO-BP algorithm is due to the other two algorithms. In the wireless sensor network routing energy consumption experiment, in addition, the SMPSO-BP algorithm uses less energy than the other two methods. Therefore, the energy-saving algorithm under the neural network data fusion mechanism is still feasible.

A Multineuron-Based Routing Algorithm of Tile-Based 2-D Mesh

Tile-based architecture is broadly utilized in the structuring of the system-on-chip by different vendors. Nevertheless, the performance of the system-on-chip (SoC) is incredibly influenced by the performance of the network hidden on a chip named network-on-chip (NoC). Routing of network-on-chip network chips plays a crucial role in the overall performance of the NoC. In this paper, we have proposed a routing algorithm that utilizes the neural network to perform the routing. This routing algorithm updates the route dependent on the port execution of the switch. From the outcome, the execution of the directing has worked successfully and has the option to deal with the enormous burden viably. The result obtained shows that the performance metric for the uniform traffic is slightly better in comparison to XY routing at the higher loads of 80%. In the case of neighbor traffic, bit complement traffic, and tornado traffic, these values are higher on 80% of the load. The reason for better handling of the loads is due to the parallelization due to the pipeline created by the neural network routing decision.

Deep Neural Network Routing with Dynamic Space Division for 3D UAV FANETs

Deep Neural Network Routing with Dynamic Space Division for 3D UAV FANETs

GraphNET: Graph Neural Networks for routing optimization in Software Defined Networks

In this paper, a graph neural net-based routing algorithm is designed which leverages global information from controller of a software-defined network to predict optimal path with minimum average delay between source and destination nodes in software-defined networks. Graph nets are used because of their generalization capability which allows the routing algorithm to scale across varying topologies, traffic schemes and changing conditions. A deep reinforcement learning framework is developed to train the Graph Neural Networks using prioritized experience replay from the experiences learnt by the controllers. The algorithm is tested on various small and large topologies in terms of packets successfully routed and average packet delay time. Experiments are performed to check robustness of routing algorithms to changes in network structure and effects of varying hyperparameters. The proposed algorithm shows impressive results when compared to q-routing and shortest path routing algorithm in terms of above experiments and is robust to varying graphical structure of the network.

Neural Network Optimal Routing Algorithm Based on Genetic Ant Colony in IPv6 Environment.

The traditional IPv6 routing algorithm has problems such as network congestion, excessive energy consumption of nodes, and shortening the life cycle of the network. In response to this phenomenon, we proposed a routing optimization algorithm based on genetic ant colony in IPv6 environment. The algorithm analyzes and studies the genetic algorithm and the ant colony algorithm systematically. We use neural network to build the initial model and combine the constraints of QoS routing. We effectively integrate the genetic algorithm and ant colony algorithm that maximize their respective advantages and apply them to the IPv6 network. At the same time, in order to avoid the accumulation of a lot of pheromones by the ant colony algorithm in the later stage of the network, we have introduced an anticongestion reward and punishment mechanism. By comparing the search path with the optimal path, rewards and punishments are based on whether the network path is smooth or not. Finally, it is judged whether the result meets the condition, and the optimal solution obtained is passed to the BP neural network for training; otherwise, iterative iterations are required until the optimal solution is satisfied. The experimental results show that the algorithm can effectively adapt to the IPv6 routing requirements and can effectively solve the user's needs for network service quality, network performance, and other aspects.

Characterization of Health Data using Neural Network and Routing in Health Monitoring

Characterization of Health Data using Neural Network and Routing in Health Monitoring

Highly Reliable Control Computer Networks with Neural Network Adaptive Routing Based on Prospective Models of the Elbrus Family

Highly Reliable Control Computer Networks with Neural Network Adaptive Routing Based on Prospective Models of the Elbrus Family

NNIRSS: neural network-based intelligent routing scheme for SDN

With the increasing diversification of network applications, SDN tends to be inefficient to satisfy the diversified application demands. Meanwhile, the continuous update of OpenFlow and flow table expansion causes the efficiency of routing and forwarding ability decreased as well as the storage space of ternary content addressable memory (TCAM) occupied by flow tables increased. In this paper, we present NNIRSS, a novel neural network (NN)-based intelligent routing scheme for SDN, which leverages the centralized controller to achieve transmission patterns of data flow by utilizing NN and replaces flow table with well-trained NN in the form of NN packet. The route of data flow can be predicted based on its application type to meet the quality of service requirements of network applications. Furthermore, we devise a radial basis function neural network-based intelligent routing mechanism. With combining APC-III and K-means algorithm, we propose APC-K-means algorithm to determine radial basis function centers. Finally, the simulation results demonstrate that our proposed NNIRSS is feasible and effective. It can reduce storage space of TCAM and routing time overhead as well as improve routing efficiency.

A Fuzzy Knowledge Based Fault Tolerance Mechanism for Wireless Sensor Networks

Wireless Sensor Networks (WSNs) are the focus of considerable research for different applications. This paper proposes a Fuzzy Knowledge based Artificial Neural Network Routing (ANNR) fault tolerance mechanism for WSNs. The proposed method uses an exponential Bi-directional Associative Memory (eBAM) for the encoding and decoding of data packets and application of Intelligent Sleeping Mechanism (ISM) to conserve energy. A combination of fuzzy rules is used to identify the faulty nodes in the network. The Cluster Head (CH) acts as the data aggregator in the network. It applies the fuzzy knowledge based Node Appraisal Technique (NAT) in order to identify the faulty nodes in the network. The performance of the proposed ANNR is compared with that of Low-Energy Adaptive Clustering Hierarchy (LEACH), Dual Homed Routing (DHR) and Informer Homed Routing (IHR) through simulation.

An Advanced 2.5-D Heterogeneous Integration Packaging for High-Density Neural Sensing Microsystem

In the traditional neural sensing microstructure, the limited metal line pitch and the metal layer numbers restrict the neural signal routing ability from electrodes to circuit chips. Miniature packaging and excessive noise interference bottlenecks are some of the challenges faced by the electrodes and circuit chips integration with traditional wire bonding. This paper proposes a 2.5-D heterogeneous integration neural sensing microsystem based on the silicon substrate to overcome these issues. With standard semiconductor and 3-D integration processes, high-channel-density (256 channels at 25 mm2) neural sensing microsystem is achieved. Through silicon via provides the shortest vertical interconnection and dramatically minimizes the packaging. Furthermore, the interposer can carry multiple chips to enhance the function of the biosensor. Electrical characteristics and reliability examinations reveal its high quality and great performance as compared to traditional approaches. This novel highly integrated neural sensing microsystem is expected to contribute to the biomedical field for exploring and solving unknown biological mysteries.

A Neural Networks-Based Hybrid Routing Protocol for Wireless Mesh Networks

The networking infrastructure of wireless mesh networks (WMNs) is decentralized and relatively simple, but they can display reliable functioning performance while having good redundancy. WMNs provide Internet access for fixed and mobile wireless devices. Both in urban and rural areas they provide users with high-bandwidth networks over a specific coverage area. The main problems affecting these networks are changes in network topology and link quality. In order to provide regular functioning, the routing protocol has the main influence in WMN implementations. In this paper we suggest a new routing protocol for WMN, based on good results of a proactive and reactive routing protocol, and for that reason it can be classified as a hybrid routing protocol. The proposed solution should avoid flooding and creating the new routing metric. We suggest the use of artificial logic—i.e., neural networks (NNs). This protocol is based on mobile agent technologies controlled by a Hopfield neural network. In addition to this, our new routing metric is based on multicriteria optimization in order to minimize delay and blocking probability (rejected packets or their retransmission). The routing protocol observes real network parameters and real network environments. As a result of artificial logic intelligence, the proposed routing protocol should maximize usage of network resources and optimize network performance.

Transient chaotic neural network-based disjoint multipath routing for mobile ad-hoc networks

Due to mobility of wireless hosts, routing in mobile ad-hoc networks (MANETs) is a challenging task. Multipath routing is employed to provide reliable communication, load balancing, and improving quality of service of MANETs. Multiple paths are selected to be node-disjoint or link-disjoint to improve transmission reliability. However, selecting an optimal disjoint multipath set is an NP-complete problem. Neural networks are powerful tools for a wide variety of combinatorial optimization problems. In this study, a transient chaotic neural network (TCNN) is presented as multipath routing algorithm in MANETs. Each node in the network can be equipped with a neural network, and all the network nodes can be trained and used to obtain optimal or sub-optimal high reliable disjoint paths. This algorithm can find both node-disjoint and link-disjoint paths with no extra overhead. The simulation results show that the proposed method can find the high reliable disjoint path set in MANETs. In this paper, the performance of the proposed algorithm is compared to the shortest path algorithm, disjoint path set selection protocol algorithm, and Hopfield neural network (HNN)-based model. Experimental results show that the disjoint path set reliability of the proposed algorithm is up to 4.5 times more than the shortest path reliability. Also, the proposed algorithm has better performance in both reliability and the number of paths and shows up to 56% improvement in path set reliability and up to 20% improvement in the number of paths in the path set. The proposed TCNN-based algorithm also selects more reliable paths as compared to HNN-based algorithm in less number of iterations.