Network Optimization Problem Research Articles

An improved artificial electric field algorithm and its application in neural network optimization

Artificial electric field (AEF) algorithm is a newly developed meta-heuristic optimization technique inspired by the Coulomb's law of electrostatic force. As so far, AEF algorithm has been successfully applied in some scientific research and engineering fields, but it still has the defects of premature convergence and poor search ability in handling complex optimization problems. To address such issues, a novel Coulomb's constant generation scheme is explored to calculate the electrostatic force, and then an improved AEF (IAEF) algorithm is developed to strengthen the global exploration capability. Subsequently, 18 popular test functions with different dimensions are selected as benchmarks for performance evaluation. Furthermore, two nonlinear problems of neural network optimization are employed to further investigate the effectiveness and superiority of IAEF algorithm. Experimental results demonstrate that the proposed IAEF is efficient and effective optimization method in comparison with the conventional AEF and several famous evolutionary algorithms.

A Novel Image Recognition Method Based on DenseNet and DPRN

Image recognition is one of the important branches of computer vision, which has important theoretical and practical significance. For the insufficient use of features, the single type of convolution kernel and the incomplete network optimization problems in densely connected networks (DenseNet), a novel image recognition method based on DenseNet and deep pyramidal residual networks (DPRN) is proposed in this paper. In the proposed method, a new residual unit based on DPRN is designed, and the idea of a pyramid residual unit is introduced, which makes the input greater than the output. Then, a module based on dilated convolution is designed for parallel feature extraction. Finally, the designed module is fused with DenseNet in order to construct the image recognition model. This model not only overcomes some of the existing problems in DenseNet, but also has the same general applicability as DensenNet. The CIFAR10 and CIFAR100 are selected to prove the effectiveness of the proposed method. The experiment results show that the proposed method can effectively reuse features and has obtained accuracy rates of 83.98 and 51.19%, respectively. It is an effective method for dealing with images in different fields.

Echo chambers in the Ising model and implications on the mean magnetization

The echo-chamber effect is a common term in opinion dynamic modeling to describe how a person’s opinion might be artificially enhanced as it is reflected back at her through social interactions. Here, we study the existence of this effect in statistical mechanics models, which are commonly used to study opinion dynamics. We show that the Ising model does not exhibit echo-chambers, but this result is a consequence of a special symmetry. We then distinguish between three types of models: (i) those with a strong echo-chamber symmetry, that have no echo-chambers at all; (ii) those with a weak echo-chamber symmetry that can exhibit echo-chambers but only if there are external fields in the system, and (iii) models without echo-chamber symmetry that generically have echo-chambers. We use these results to construct an efficient algorithm to efficiently and precisely calculate magnetization in arbitrary tree networks. Finally, we apply this algorithm to study two systems: phase transitions in the random field Ising model on a Bethe lattice and the influence optimization problem in social networks.

From Resource Auction to Service Auction: An Auction Paradigm Shift in Wireless Networks

In 5G and beyond, the newly emerging services, such as edge computing/intelligence services, may demand the provision of heterogeneous communications, computing, and storage (CCS) resources on and across network entities multihop apart. In such cases, traditional resource-oriented auction schemes, where buyers place bids on resources, may not be effective in providing end-to-end (E2E) quality-of-service (QoS) guarantees. To overcome these limitations, in this article, we coin the concept of E2E service auction where the auction commodities are E2E services rather than certain resource. Under this framework, buyers simply bid for services with E2E QoS requirements without having to know the inner working (which resources are behind). To guarantee E2E QoS for winning bids while ensuring essential economic properties, E2E service auction requires addressing the joint problem of network optimization and auction design with both economical and QoS constraints. To substantiate the mechanism design, we illustrate how to devise E2E service auctions for edge computing systems under various scenarios. We also identify the research opportunities on E2E service auction mechanism design for other critical use cases, including edge intelligence.

A Sustainable Intermodal Location-Routing Optimization Approach: A Case Study of the Bohai Rim Region

The optimal intermodal nodes and routes are two of the most challenging issues for intermodal participants. We present a two-phase approach that includes the fuzzy c-means clustering method (FCM) and a multi-objective optimization model to solve intermodal location-routing issues. A weighted sum technique and a genetic algorithm (GA) are designed to address this model. The two-phase approach is beneficial in meeting different market demand preferences of intermodal participants. It also has applications in solving the sustainable intermodal location-routing problems, further solving the network optimization problem in large-scale scenarios. A typical intermodal transport network in the Bohai Rim region is used to verify the effectiveness of this approach. The results provide references for the participants in the Bohai Rim region to choose the optimal intermodal nodes and routes. The findings also offer theoretical insights for optimizing intermodal networks in other regions of China, with goals of improving sustainable transport efficiencies.

Velocity-Adaptive Access Scheme for MEC-Assisted Platooning Networks: Access Fairness Via Data Freshness

Platooning strategy is an important part of autonomous driving technology. Due to the limited resource of autonomous vehicles in platoons, mobile edge computing (MEC) is usually used to assist vehicles in platoons to obtain useful information, increasing its safety. Specifically, vehicles usually adopt the IEEE 802.11 distributed coordination function (DCF) mechanism to transmit large amount of data to the base station (BS) through vehicle-to-infrastructure (V2I) communications, where the useful information can be extracted by the edge server connected to the BS and then sent back to the vehicles to make correct decisions in time. However, vehicles may be moving on different lanes with different velocities, which incurs the unfair access due to the characteristics of platoons, i.e., vehicles on different lanes transmit different amount of data to the BS when they pass through the coverage of the BS, which also results in the different amount of useful information received by various vehicles. Moreover, age of information (AoI) is an important performance metric to measure the freshness of the data. Large average age of data implies not receiving the useful information in time. It is necessary to design an access scheme to jointly optimize the fairness and data freshness. In this paper, we formulate a joint optimization problem in the MEC-assisted V2I networks and present a multi-objective optimization scheme to solve the problem through adjusting the minimum contention window under the IEEE 802.11 DCF mode according to the velocities of vehicles. The effectiveness of the scheme has been demonstrated by simulation.

A Hierarchical Strategy for Multi-Objective Optimization of Distribution Network Considering DGs and V2G-Enabled EVs Integration

As an important part of smart grid construction, the distribution network (DN) optimization problem has always attracted great attention, especially under the background that large-scale penetration of distributed generators (DGs) and electric vehicles (EVs) into building cluster poses both opportunities and challenges to the energy management. This research presents a hierarchical optimization strategy, for improving the safe and economical operation of DN considering the DGs and EVs integration. In Stage 1, the MPPT control model of DGs is designed to obtain the best energy conversion efficiency. In Stage 2, load models of EVs and battery energy storage system (BESS) under coordinate charging/discharging stimulated by a time-of-use incentive mechanism are established respectively, to achieve a load curve with a minimized peak-to-valley difference (PVD). In Stage 3, aiming for the best compromise between the active power loss and node voltage excursion, daily optimal scheduling of the static Var compensator (SVC) capacitors is dynamically worked out according to the varying power demand, as the solution for the defined multi-objective optimization problem. For enhancing the convergence speed, an advanced genetic algorithm with elite preservation strategy is employed. The proposed hierarchical strategy is demonstrated on an IEEE 33-node DN test case, and the simulation results show that first, the MPPT control ensures the maximum power outputs of DGs; next, power supply pressure could be relieved by the load shifting effects of the coordinated vehicle-to-grid (V2G) service and BESS configuration, reflected in the decreased load peak from 4,370.1 to 3,424.99 kW, and the optimized PVD from 1763.8 to 703.8 kW; meanwhile, via applicable power planning of the SVC components, optimized power loss and voltage quality can both be achieved, proving the feasibility of the optimization strategy, which promotes the economic and reliable operation of the DN system.

Optimization of Water Distribution Networks Using Genetic Algorithm Based SOP–WDN Program

Water distribution networks are vital hydraulic infrastructures, essential for providing consumers with sufficient water of appropriate quality. The cost of construction, operation, and maintenance of such networks is extremely large. The problem of optimization of a water distribution network is governed by the type of water distribution network and the size of pipelines placed in the distribution network. This problem of optimal diameter allocation of pipes in a distribution network has been heavily researched over the past few decades. This study describes the development of an algorithm, ‘Smart Optimization Program for Water Distribution Networks’ (SOP–WDN), which applies genetic algorithm to the problem of the least-cost design of water distribution networks. SOP–WDN demonstrates the application of an evolutionary optimization technique, i.e., genetic algorithm, linked with a hydraulic simulation solver EPANET, for the optimal design of water distribution networks. The developed algorithm was applied to three benchmark water distribution network optimization problems and produced consistently good results. SOP–WDN can be utilized as a tool for guiding engineers during the design and rehabilitation of water distribution pipelines.

When Optimization Meets Machine Learning: The Case of IRS-Assisted Wireless Networks

Performance optimization of wireless networks is typically complicated because of high computational complexity and dynamic channel conditions. Considering a specific case, the recent introduction of intelligent reflecting surface (IRS) can reshape the wireless channels by controlling the scattering elements' phase shifts, namely, passive beamforming. However, due to the large size of scattering elements, the IRS's beamforming optimization becomes intractable. In this article, we focus on machine learning (ML) approaches for complex optimization problems in wireless networks. ML approaches can provide flexibility and robustness against uncertain and dynamic systems. However, practical challenges still remain due to slow convergence in offline training or online learning. This motivated us to design a novel optimization-driven ML framework that exploits the efficiency of model-based optimization and the robustness of model-free ML approaches. Splitting the control variables into two parts allows one part to be updated by the outer loop ML approach while the other part is solved by the inner loop optimization. The case study in IRS-assisted wireless networks confirms that the optimization-driven ML framework can improve learning efficiency and the reward performance significantly compared to conventional model-free ML approaches.

A novel method to solve the optimization problem of uncertain network system based on uncertainty theory

<abstract> <p>When the network optimization problem is discussed, in the actual situation, it is necessary to consider the uncertain factors in the network. This paper employs the theories of uncertainty, uncertain programming and network optimization to solve the uncertain network optimization problem. First, based on uncertainty theory and uncertainty graph, we redefine the concept of an uncertain network system, and propose a unified identification method for an uncertain network system based on a conditional uncertain measure matrix. Second, we establish the network optimization model for the shortest path problem based on a conditional uncertain measure matrix. Third, according to the measure simulation technology, a hybrid intelligent algorithm is designed to solve the model. Finally, the correctness and feasibility of the approach is illustrated by a numerical example of an underground logistics system.</p> </abstract>

An adaptive scheduling routing for multimedia services based on software-defined networks

We address the routing optimization problem in Software-Defined Networks (SDN) while satisfying multiple QoS constraints of multimedia services. In this paper, we 1) formally define the problem of routing optimization in SDN under multiple QoS constraints, and 2) suggest a novel architecture based on SDN for providing QoS services in the network. We propose an adaptive scheduling routing optimization approach in SDN that improves the classical Dijkstra routing algorithm. We report the results of experiments using the Mininet simulation framework that confirm the advantages of our proposed QoS routing optimization approach as compared to the state-of-the-art solutions.

Design and Research of M2M Message Transfer Mechanism of Looms for Information Transmission

Machine-to-machine (M2M) communication is a key implementation technology for future Industrial Internet of Things (IIoT) applications. The requirements for flexibility, efficiency, and crossplatform compatibility in intermodule communication between the connected machines raise challenges for the M2M messaging mechanism. For the loom message transmission mechanism, it is necessary to have certain flexibility and meet certain M2M communication requirements to understand each other. In this paper, we analyze and study the data model, data exchange language, MQTT protocol and logo resolution technology of looms to address the current problems of low data transmission efficiency, poor flexibility and lack of flexibility in the network.By establishing the data model of looms and classifying them by topic, selecting JSON format as the transmission format of data, and using MQTT protocol to realize data transmission between loom networks, an MQTT-based messaging mechanism is established.The data involved in the communication of the loom equipment are grouped according to the communication requirements, and only the loom data required by the application are subscribed. At the same time, the hardware and software parts of the communication module are designed with ESP32 as the core, so that it can complete the functions of loom network communication, data analysis, format conversion and so on. In addition, the real-time requirements of the loom network communication and the network requirements for future intelligent manufacturing must be met. This paper uses the BP neural network to intelligently optimize the loom network, and the existing problems related to performance indices are analyzed. The original loom network is optimized by integrating the RSSI, QoS requirements, transmission priority, real-time communication and data transmission volume of nodes. In the meantime, the particle swarm optimization algorithm is introduced to optimize the BP neural network to avoid the problem that it easily can fall into the local optimal solution and effectively improving the network optimization efficiency. Finally, the feasibility of this method is proven by using the experimental platform, which can solve the problems of low network transmission efficiency and difficult network optimization of looms and provide an effective technical example of loom information transmission and network optimization for industry.

Cluster-Based Radio Resource Management in Dynamic Vehicular Networks

With the rapid growth of traffic demand for vehicular data transmissions, the limited cellular band becomes a constraint to meet the needs of all cellular-vehicle-to-everything (C-V2X) users. In this paper, we investigate an efficient clustering-based spectrum resource management scheme for dynamic heterogeneous vehicular networks. Due to the frequent changes of channel state information (CSI) caused by high mobility of vehicles, the traditional radio resource management scheme based on complete CSI is no longer applicable. To solve this problem, we consider different quality of service (QoS) requirements for vehicular communications and propose an optimization problem for C-V2X networks. The proposed dynamic vehicular optimization strategy is carried out on the vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) links, which can construct a more realistic two-way lane in actual communication environments. We also propose a low-complexity vehicle matching algorithm based on large-scale fading to improve the stability of communication links. The desirable performance is confirmed by computer simulations.

Mean-Field Game Theory Based Optimal Caching Control in Mobile Edge Computing

Mobile edge computing (MEC) can use wireless access network (RAN) to provide users with nearby information technology (IT) services and cloud computing functions, which creates a high-performance and low latency service environment. By caching the popular content at small base station (SBS) can reduce the heavy backhaul load and the content retransmission. However, the time-varying and dynamic of the content requests may lead to the base station to cache the useless contents. In this paper, we study a distributed caching optimization problem in edge networks (ENs) with the spatio-temporal requirements. In the considered ENs, the cache control is described as a stochastic differential game (SDG) in which each SBS defines a caching strategy to reduce the total cost in terms of the service delay and backhaul link load. To reduce the computational complexity, the original optimization problem is transformed into a mean field game (MFG). We propose a distributed caching iterative control algorithm that decouples the information interaction between the general SBS and others through the mean field distribution. In addition, we obtain the optimal edge caching control strategy, while the existence and uniqueness of the mean field equilibrium (MFE) can also be guaranteed. Simulation results demonstrate that our proposed caching control algorithm can average reduce 27.12% storage cost and achieve better performance than other existing schemes.

The Neutrosophic Traveling Salesman problem with Neutrosophic EdgeWeight: Formulation and A Genetic Algorithm

The traveling salesman problem (TSP) is an important and well known classical combinatorial network optimization problem in operation research, where the TSP finds a shortest possible route through a set of n nodes such that each and every node are visited exactly one time except for the starting node. In this problem, the arc lengths are generally considered to represent the traveling time or travelling cost rather than geographical distance. It is not possible to predict the exact arc length because the traveling time or traveling cost fluctuated with payload, weather, traffic conditions and so on. neutrosophic set theory provides a new tool to handle the uncertainties in TSP. In this paper, we concentrate on TSP on a network in which neutrosophic set, Instead of real number is assigned to edge as edge weight. We propose a mathematical model for a TSP with neutrosophic arc lengths. We present the utility of neutrosophic sets as arc length for TSP. An algorithmic method based on Genetic Algorithm (GA) is proposed for solving this problem. We have designed a new heuristic crossover and heuristic mutation our proposed GA. We have used a numerical example to illustrate the effectiveness of our proposed algorithm.

A Reinforcement Learning Variable Neighborhood Search for the Robust Dynamic Bayesian Network Optimization Problem under the Supply Chain Ripple Effect

Due to the impact of the global COVID-19, numerous industries have suffered from the disruption propagating along the supply chain, i.e. the ripple effect. To reduce adverse impact of the ripple effect, supply chain (SC) risk management under it is becoming an increasingly hot topic in both practice and research. In our former research, a robust dynamic bayesian network (DBN) approach has been developed for disruption risk assessment, whereas the solution methods adopted before (commercial solvers and simulated annealling algorithm) are not efficient enough, especially for large-size instances. For this reason, a new reinforcement learning variable neighborhood search (QVNS) is developed for solving the robust DBN optimization model, where the Q-learning algorithm is implemented to select the most efficient neighborhood structure in different stages of the search process. We conduct computational experiments on randomly generated instances, which indicates that Q-learning algorithm can improve significantly the performance of the VNS on large-size instances of the robust DBN optimization problem.

Agent-Supervisor Coordination for Decentralized Event-Triggered Optimization

This paper proposes decentralized resource-aware coordination schemes for solving network optimization problems defined by objective functions which combine locally evaluable costs with network-wide coupling components. These methods are well suited for a group of supervised agents trying to solve an optimization problem under mild coordination requirements. Each agent has information on its local cost and coordinates with the network supervisor for information about the coupling term of the cost. The proposed approach is feedback-based and asynchronous by design, guarantees anytime feasibility, and ensures the asymptotic convergence of the network state to the desired optimizer. Numerical simulations on a power system example illustrate our results.

MATHEMATICAL MODELS OF DECISION SUPPORT IN THE PROBLEMS OF LOGISTICS NETWORKS OPTIMIZATION

The subject of research in the article is the process of decision support in the problems of logistics networks optimization. The goal of the work is to develop a set of mathematical models of logistics network optimization problems to increase the efficiency of decision support systems by coordinating the interaction between automatic and interactive procedures of computer-aided design systems. The following tasks are solved in the article: review and analysis of the current state of the problem of decision support in the problems of logistics networks optimization; decomposition of the problem of decision support for the optimization of logistics networks; development of a mathematical model of the general problem of network optimization in terms of economy, efficiency, reliability and survivability; development of a set of technological mathematical models for the correct reduction of many effective options for building logistics networks for the final choice, taking into account difficult to formalize factors, knowledge and experience of the decision maker (DM). The following methods are used: systems theory, utility theory, optimization and operations research. Results. Analysis of the current state of the problem of logistics networks optimization has established the existence of the problem of correct reduction of a subset of effective options for their construction for ranking, taking into account difficult to formalize factors, as well as knowledge and experience of DM. The decomposition of the problem into tasks is performed: definition of the principles of network construction; network structure selection; determination of the topology of network elements; choice of network operation technology; determination of parameters of elements and communications (means of cargo delivery); multi criteria evaluation and selection of the best option for building a network. A mathematical model of the general problem of network optimization in terms of economy, efficiency, reliability and survivability is proposed. To coordinate the interaction between automatic and interactive network optimization procedures, it is proposed to use a combined method of ranking options, which allows you to identify and correctly reduce the subset of effective options for ranking DM. To implement the method, mathematical models of problems of the procedure of ranking options in the technologies of project decision support have been developed, which allow to combine the advantages of the technologies of the ordinalistic and cardinalistic approaches. Conclusions. The developed set of mathematical models expands the methodological bases of automation of processes of support of multi criteria decisions on optimization of logistic networks, allows to carry out correct reduction of set of effective options of their construction for the final choice taking into account factors, knowledge and experience of DM. The practical use of the proposed models and procedures will reduce the time and capacity complexity of decision support technologies, and through the use of the proposed selection procedures - to improve their quality across a variety of functional and cost indicators.

On the Implications of Routing Models on Network Optimization

In network optimization problems, from traffic engineering to network monitoring, the routing model is typically considered as something given and frozen. This paper is motivated by the fundamental question how the ability to <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">change</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">optimize</i> the routing model itself influences the efficiency at which communication networks can be operated. To this end, we identify two main dimensions of a routing model: <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">consistency</i> (of a single route) and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">coherence</i> (of sets of routes). We present analytical results on the impact of the routing model on the achievable route diversity as well as on the runtime of solving optimization problems underlying different case studies. We also uncover that it can sometimes be beneficial to <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">artificially</i> restrict the routing model, to significantly reduce the computational complexity without negatively affecting the route diversity much.

Deep Learning for Wireless Networking: The Next Frontier

With the growth of mobile technology in the last decade, wireless networks have become an integral part of our everyday lives. To meet the increasingly stringent application requirements, more and more network resources and features are becoming available, which requires innovative system designs such that the configuration and management of the networks can be performed automatically and autonomously. Due to its superior capability of discovering insightful knowledge in a data-driven manner, the emerging deep learning (DL) technology has shown great potential to fulfil this goal. This article systematically reviews recent efforts in leveraging DL for addressing wireless network optimization problems, presenting a fundamental understanding of where and how the supremacy of DL based approaches comes versus the conventional modeling based approaches. The basic research challenges and some promising research directions for fully exploiting the potential of DL in wireless network optimization are also discussed. The effectiveness of DL is illustrated with an innovative case study of integrating DL with multi-hop wireless network flow optimization.