Heavy Network Load Research Articles

An Effective Intrusion Detection System for Edge Computing Using ConvNeXt and ResNet152V2

The proliferation of edge computing, driven by network applications and wireless devices, increases the vulnerability of confidential information to security risks. In this environment, existing intrusion detection algorithms fail to satisfy the requirements of prompt responses, heavy network load management, inadequate extraction of features, and imprecise model classification. In this work, the imbalanced data problem in the input dataset is mitigated using the Data Augmentation Generative Adversarial Network (DAGAN). Next, an efficient ConvNeXt-based feature extraction method is created to retrieve the key characteristics from the dataset for every class. Last, multi-attack intrusion detection is achieved through the deployment of an optimized deep learning classifier based on ResNet152V2. Furthermore, simulation experiments are carried out on the ToN-IoT and BoT-IoT datasets, and the outcomes demonstrate that our suggested model performs better than the existing models, with accuracy levels of 99.20% and 99.31%, respectively. These findings show that this approach is successful in building and refining large-scale IDS in the edge computing framework.

Event‐based reduced‐order H∞$H_{\infty }$ estimation for switched complex networks based on T‐S fuzzy model

AbstractThis paper investigates the design of a reduced‐order filter for a specific class of switched complex network systems with time‐varying delays. To handle the nonlinear components of the complex network, a T‐S fuzzy model is employed to convert them into a set of linear components. To tackle the issue of heavy network load, a memory‐based adaptive trigger mechanism is proposed. In this study, a reduced‐order state estimator is designed using the T‐S fuzzy model. To demonstrate the exponential stability of the error system, a suitable Lyapunov function is constructed based on the Lyapunov stability principle. The parameter matrix of the reduced‐order estimator is determined using the linear matrix inequality and convex linearization method. Additionally, a sufficient condition for the exponential stability of the error system at the suppression level is provided. Finally, the feasibility of the proposed scheme is validated through a simulation experiment.

Improved Grant-Free Access for URLLC via Multi-Tier-Driven Computing: Network-Load Learning, Prediction, and Resource Allocation

Grant-Free (GF) access has been recognized as a promising candidate for Ultra-Reliable and Low-Latency Communications (URLLC). However, even with GF access, URLLC still may not effectively gain high reliability and millimeter-level latency, simultaneously. This is because the network load is typically time-varying and not known to the base station (BS), and thus, the resource allocated for GF access cannot well adapt to variations of the network load, resulting in low resource utilization efficiency under light network load and leading to severe collisions under heavy network load. To tackle this problem, we propose a multi-tier-driven computing framework and the associated algorithms for URLLC to support users with different QoS requirements. Especially, we concentrate on <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$K $ </tex-math></inline-formula> - repetition GF access in light of its simplicity and well-balanced performance for practical systems. In particular, our framework consists of three tiers of computation, namely network-load learning, network-load prediction, and adaptive resource allocation. In the first tier, the BS can learn the network-load information from the states (success, collision, and idle) of random-access resources in terms of resource blocks (RB) and time slots. In the second tier, the network-load variation is effectively predicted based on estimation results from the first tier. Finally, in the third tier, by deriving and weighing the failure probabilities of different groups of users, their QoS requirements, and the predicted network loads, the BS is able to dynamically allocate sufficient resources accommodating the varying network loads. Simulation results show that our proposed approach can estimate the network load more accurately compared with the baseline schemes. Moreover, with the assistance of network-load prediction, our adaptive resource allocation offers an effective way to enhance the QoS for different URLLC services, simultaneously.

Two stage downlink scheduling for balancing QoS in multihop IAB networks

The 3GPP has envisioned Integrated Access and Backhaul (IAB) as a key enabler to support the flexible and ultra-dense deployment of 5G cells with significantly reduced deployment costs. However, IAB introduces new research challenges, especially when studying multihop topology. This paper presents a QoS-based downlink scheduler designed explicitly for IAB networks. The scheduler is devised after considering multihop relaying topology, QoS requirements, and backhaul constraints. We investigate its performance using system-level simulations under diverse IAB topologies. The performance results show that the scheduler is capable of fulfilling QoS requirements for different types of services, even at heavy network load. The scheduler also maintains excellent fairness among QoS flows belonging to the same service type.

Cache-Enabled Multicast Content Pushing With Structured Deep Learning

The cache-enabled multicast content pushing, which multicasts the content items to multiple users and caches them until requested, is a promising technique to alleviate the heavy network load by enhancing the traffic offloading. This, in turn, has called for the optimization of content pushing strategy while considering both the transmission and caching resources, which jointly result in the complicated coupling among pushing decisions and lead to high computational complexity. Unlike most existing approaches which simplify the pushing problem via bypassing the complicated coupling, in this paper, we propose a multicast content pushing strategy to maximize the offloaded traffic with the cost on content caching based on structured deep learning. Specifically, we design the convolution stage to extract the spatio-temporal correlations of one content item between different pushing decisions, and construct the fully-connected stage to capture the spatial coupling among the decisions of pushing different content items to different user devices. Moreover, to address the absence of the ground truth on multicast content pushing, we relax the transmission constraint to derive a performance upper bound for guiding the training direction. This relaxed problem is solved based on dynamic programming in a bottom-up manner. Compared to the state-of-the-art baselines including both the traditional model-based and the general neural network-based strategies, the proposed pushing strategy achieves significant performance gain in both the random-generated dataset and the real LastFM dataset. In addition, it is also shown that the proposed strategy is robust to the uncertainty of user request information.

Opt-ACM: An Optimized load balancing based Admission Control Mechanism for Software Defined Hybrid Wireless based IoT (SDHW-IoT) network

The wide network applications of Internet of Things (IoT) urge to integrate the different wireless networking technologies. However, the future generation technologies like Software Defined Networking (SDN) can be useful to alleviate the associated challenges such as security, installation issues, coverage, etc. The growing number of IoT users produce massive volume of traffic causing heavy network load conditions, resulting in the state of network congestion. The network congestion may cause severe performance degradation issues such as frequent packet drops, longer delays, low throughput, etc. Thus, a promising solution is needed to reduce and/or prevent the network congestion. This paper presents an Optimized load balancing based Admission Control Mechanism (Opt-ACM) for effective network flow management resulting in the reduced network congestion. In addition to this, the paper highlights the challenges of the existing solutions, and discusses a Software Defined Hybrid Wireless based IoT (SDHW-IoT) network architecture consisting of Software Defined Wireless Sensor Network (SDWSN) and Software Defined Wireless Mesh Network (SDWMN). To validate the efficiency of Opt-ACM, a Mixed-Integer Linear Programming (MILP) based optimization problem is formulated and tested using a well-known mathematical optimization solver called Gurobi. Additionally, Opt-ACM is also emulated in Mininet-Wifi with varying network scenarios against some traditional (OLSR and OSPF) and state-of-the-art (FACOR and EASDN) approaches. Opt-ACM achieves an overall efficiency of 9.47% and 12.32% over other approaches for Packet Delivery Ratio (PDR) and Packet Loss Ratio (PLR) respectively. Similarly, an average improved efficiency of 26.77% and 33.10% is achieved with respect to the Average Delay (AD) and Average Jitter (AJ) metrics respectively.

Routing Algorithm for Elastic Optical Network Based on Machine Learning assisted Traffic Prediction

Elastic Optical Network (EON) is the promising technology for future network which may suffer congestion or fragment problems when huge amount of burst traffic arrives. This paper proposes an improved routing strategy for EON based on Artificial Intelligent (AI) assisted traffic prediction. Theoretical analysis and simulation results show that compared with traditional algorithms, the improved algorithm can effectively reduce the blocking rate of the network especially under heavy network load.

Global load-balancing ONU assignment for a software-defined reconfigurable PON

Rapidly growing and fluctuating traffic demands lead to unbalanced traffic scenarios in optical access networks, which have an impact on network efficiency. Thus, the flexible software-defined reconfigurable passive optical network (PON) architecture has been proposed to handle this challenge. To improve the efficiency of a reconfigurable PON facing unbalanced traffic demands, this paper proposes a global load-balancing (GLB) optical network unit (ONU) assignment method, which aims to achieve load balancing among the optical line terminals (OLTs) in the OLT pool as well as among the aggregation switch ports that connect the OLTs by optimizing the affiliations between the OLTs and the ONUs. For on-line optimization, a greedy GLB algorithm is introduced that regroups the ONUs and generates load-balanced target ONU-OLT affiliations in accordance with dynamic traffic demands. For off-line optimization, bee colony optimization (BCO) is invoked to search for better ONU regrouping solutions for GLB. Simulation results show that the proposed GLB algorithm can significantly improve the throughput and delay performance of an optical access network faced with unbalanced traffic demands, especially when the network load is heavy. Moreover, the proposed BCO technique for GLB can obviously improve the worst delay performance and reduce the migrated traffic of the GLB solution, especially under network conditions with high peer-to-peer traffic proportions and heavy network loads. Thus, the proposed global load-balancing ONU assignment method is effective for helping a software-defined reconfigurable PON to cope with unbalanced traffic demands.

Q-Learning-Based Spectrum Access for Content Delivery in Mobile Networks

By the wide development of wireless communication technology, mobile networks are becoming the promising paradigm for mobile users to access the Internet to download contents. However, transmitting contents to multiple mobile users causes excessive interference and heavy network load. Moreover, the service demands for rich contents over mobile networks have been ever-increasing, which leads serious traffic loads to the networks. To improve mobile user’s quality of experience (QoE), in this paper, a Q-learning based spectrum access scheme in mobile networks is firstly proposed for mobile users to access the optimal spectrum and maximize the transmission rate. Secondly, based on the optimal spectrum decision, we present a content delivery scheme for edge nodes, device-to-device (D2D) pairs and mobile users. After that, we model the content delivery process among edge nodes, D2D pairs and mobile users as a non-cooperative Stackelberg game to improve the efficiency of content delivery. Then, the content delivery strategy algorithm is proposed to derive the Stackelberg equilibrium. Each player can obtain the maximum utility from the game based on the optimal strategy. Finally, we conduct extensive simulations to validate the performance of the proposed scheme. Simulation results demonstrate that the proposed scheme can jointly maximize the throughput of mobile networks and improve the QoE of mobile users compared with the conventional schemes.

DCN-MAC: A Dynamic Channel Negotiation MAC Mechanism for Underwater Acoustic Sensor Networks.

In the design of media access control (MAC) mechanism in underwater acoustic sensor networks (UASNs), due to the propagation characteristic of low-speed underwater acoustic signals, it is necessary to solve the spatial–temporal uncertainty problem. In order to avoid the multi-user access conflict in underwater networks, reduce the energy cost and improve the throughput and fairness across the network, a dynamic channel negotiation MAC mechanism based on spatial–temporal mapping of receiving queue (DCN-MAC) was proposed. DCN-MAC uses a duty cycle mechanism and implements a network management based on dynamic single node wake-up. The awakening node collects the request to send (RTS) and network status information in the network to solve the problem of space-temporal uncertainty and the highly dynamic needs of network access nodes and access services. The simulation results show that in different network scenarios, especially in those featuring high density and heavy network load, compared with the traditional underwater acoustic network MAC protocols, this protocol can effectively improve the network throughput and reduce the packet loss probability caused by multi-node conflict.

Decentralized distribution-sampled classification models with application to brain imaging

BackgroundIn this age of big data, certain models require very large data stores in order to be informative and accurate. In many cases however, the data are stored in separate locations requiring data transfer between local sites which can cause various practical hurdles, such as privacy concerns or heavy network load. This is especially true for medical imaging data, which can be constrained due to the health insurance portability and accountability act (HIPAA) which provides security protocols for medical data. Medical imaging datasets can also contain many thousands or millions of features, requiring heavy network load. New methodOur research expands upon current decentralized classification research by implementing a new singleshot method for both neural networks and support vector machines. Our approach is to estimate the statistical distribution of the data at each local site and pass this information to the other local sites where each site resamples from the individual distributions and trains a model on both locally available data and the resampled data. The model for each local site produces its own accuracy value which are then averaged together to produce the global average accuracy. ResultsWe show applications of our approach to handwritten digit classification as well as to multi-subject classification of brain imaging data collected from patients with schizophrenia and healthy controls. Overall, the results showed comparable classification accuracy to the centralized model with lower network load than multishot methods. Comparison with existing methodsMany decentralized classifiers are multishot, requiring heavy network traffic. Our model attempts to alleviate this load while preserving prediction accuracy. ConclusionsWe show that our proposed approach performs comparably to a centralized approach while minimizing network traffic compared to multishot methods.

SPP 2.0: Simple Packet Protocol for ultra low cost wired networks

This article presents the Simple Packet Protocol version 2.0 (SPP 2.0). The protocol core is technology-independent and it is designed to provide communication for low data rate wired bus networks. The protocol is characterized by its low hardware requirements which allows for implementing it on embedded systems with limited hardware resources and achieving a very low price per node. An important feature of the SPP 2.0 is that its design addresses the multi-master asynchronous communication with flexible addressing model. Thanks to an effective collision avoidance and detection mechanism, the protocol is reliable under heavy network load conditions even without any additional feedback from a receiver. The presented SPP 2.0 is designed to provide wired communication for embedded systems connected with economical and maintenance-free wired network. SPP 2.0 systems can be easily integrated into larger networks, including IoT applications. The possible use cases for the protocol include: wired sensor and actuator networks, building automation systems, and modular embedded hardware platform prototyping.

CCLBR: Congestion Control-Based Load Balanced Routing in Unstructured P2P Systems

Given the growing popularity of the peer-to-peer (P2P) network systems in the recent years, efficient query routing under highly dynamic environments is still lacking in several P2P network systems. In response to this challenge, this paper proposes a new churn-resilient system to find alternative routing paths for the purpose of balancing the query loads under higher network churns and heavy workloads, ultimately to improve the search efficiency. Two novel methods are devised to balance the network query loads among both inter- and intragroup level peers. First, a resource grouping and a rewiring method is proposed to spontaneously organize and cluster the peers having same resources together. This strategy facilitates the peers to evolve the network into a cluster-like topology and balances the query loads among the intergroup peers. Second, a collaborative Q -learning method is proposed to balance the query loads among the intragroup peers in order to intelligently avoid queries being forwarded to the congested peers in the network. Experiments conducted under dynamic network scenarios demonstrate that our proposed method achieves better search performances with a more balanced network load than the existing methods, and further exhibits higher robustness and adaptability under higher network churns and heavy network loads.

Low Polling Time TDM ONoC With Direction-Based Wavelength Assignment

Optical network-on-chip (ONoC) is a promising technology for future on-chip many-core communications. ONoC can achieve high bandwidth density and low communication latency under low network cost, which leads to an energy-efficient network. Time-division multiplexing (TDM)-based ONoC aims to solve the congestion problem faced by many optical circuit switching (OCS)-based mesh ONoCs. But TDM-based ONoC still suffers fromhigh network polling time, which causes high average network latency, especially under low network load. In this paper, we propose a novel TDMONoC using torus topology and direction-based wavelength assignment, named TTWA. TTWA ONoC combines TDM communication strategy and torus network topology to solve the congestion problem under the heavy network load faced by OCS-based ONoCs.With the utilization of direction-based wavelength assignment and torus topology, we further propose a high-efficiency communication strategy to improve the adjacent inter-cluster parallel communications and decrease the total TDM slot number, which contributes to a smaller network polling time and lower average network latency. A novel optical router is simultaneously designed to meet the communication strategy in TTWA. The analyses and simulation results indicate that TTWA ONoCimproves the polling time compared with other TDM-based ONoCs and achieves better network performance compared with the equivalent OCS-based ONoCs under various synthetic traffic patterns and realistic scientific applications. After the calculation,TTWA ONoC reduces the network insertion loss of OCS-based mesh ONoC 46.92% and that of OCS-based torus ONoC 31.75% and reduces the required laser power of OCS-based mesh ONoC 62.90% and that of OCSbased torus ONoC 40.66% under the worst-case situation.

An extended AODV routing protocol for secure MANETs based on node trust values

In MANETs, various QoS problems exist such as high error rates, handling node mobility, inefficient routing, power conservation, and limited processing capabilities of network devices. To obtain secure ad hoc routing, misbehaving or malicious nodes should be detected and isolated. Otherwise, the overall network performance will be decreased. The participating nodes should establish a network with an acceptable level of trust relationships among themselves. This problem is solved with the adoption of a trust management framework and by extending the conventional routing protocol. This paper proposes a new routing protocol (TRUST_AODV) that extends the AODV routing protocol by incorporating a trust algorithm that detects misbehaving nodes. For each node, the proposed algorithm computes the weighted trust by taking into account the packet delivery ratio, energy consumption rate and buffer length. TRUST_AODV secures the MANET against potential packet drop attacks and denial-of-service attacks. It also improves the network performance under heavy network load conditions. The simulation results demonstrate that it supports secure routing and optimised performance.

A high performance reliable NoC router

Aggressive scaling of CMOS process technology allows the fabrication of highly integrated chips, and enables the design of multiprocessors system-on-chip connected by the network-on-chip (NoC). However, it brings about widespread reliability challenges.Aiming to tackle the permanent faults on the router components, we propose a high performance, high reliability and low cost router design based on a generic 2-stage router. Five fault tolerant strategies are employed in our reliable router. We exploit an ECC detection and virtual channel closing strategy for input buffer faults, a double routing strategy for the routing computation (RC) faults, a default winner strategy for the virtual allocation (VA) faults, a runtime arbiter selection strategy for the switch allocation (SA) faults and a double bypass bus strategy for the crossbar faults. Different from previous reliable routers, our design leverages the feature of pipeline optimization and routing algorithm to maintain the performance in fault tolerance especially under heavy network loads. Besides, our proposed router provides higher reliability with lower hardware consumption than previous reliable router designs.

Energy-Efficient Boarder Node Medium Access Control Protocol for Wireless Sensor Networks

This paper introduces the design, implementation, and performance analysis of the scalable and mobility-aware hybrid protocol named boarder node medium access control (BN-MAC) for wireless sensor networks (WSNs), which leverages the characteristics of scheduled and contention-based MAC protocols. Like contention-based MAC protocols, BN-MAC achieves high channel utilization, network adaptability under heavy traffic and mobility, and low latency and overhead. Like schedule-based MAC protocols, BN-MAC reduces idle listening time, emissions, and collision handling at low cost at one-hop neighbor nodes and achieves high channel utilization under heavy network loads. BN-MAC is particularly designed for region-wise WSNs. Each region is controlled by a boarder node (BN), which is of paramount importance. The BN coordinates with the remaining nodes within and beyond the region. Unlike other hybrid MAC protocols, BN-MAC incorporates three promising models that further reduce the energy consumption, idle listening time, overhearing, and congestion to improve the throughput and reduce the latency. One of the models used with BN-MAC is automatic active and sleep (AAS), which reduces the ideal listening time. When nodes finish their monitoring process, AAS lets them automatically go into the sleep state to avoid the idle listening state. Another model used in BN-MAC is the intelligent decision-making (IDM) model, which helps the nodes sense the nature of the environment. Based on the nature of the environment, the nodes decide whether to use the active or passive mode. This decision power of the nodes further reduces energy consumption because the nodes turn off the radio of the transceiver in the passive mode. The third model is the least-distance smart neighboring search (LDSNS), which determines the shortest efficient path to the one-hop neighbor and also provides cross-layering support to handle the mobility of the nodes. The BN-MAC also incorporates a semi-synchronous feature with a low duty cycle, which is advantageous for reducing the latency and energy consumption for several WSN application areas to improve the throughput. BN-MAC uses a unique window slot size to enhance the contention resolution issue for improved throughput. BN-MAC also prefers to communicate within a one-hop destination using Anycast, which maintains load balancing to maintain network reliability. BN-MAC is introduced with the goal of supporting four major application areas: monitoring and behavioral areas, controlling natural disasters, human-centric applications, and tracking mobility and static home automation devices from remote places. These application areas require a congestion-free mobility-supported MAC protocol to guarantee reliable data delivery. BN-MAC was evaluated using network simulator-2 (ns2) and compared with other hybrid MAC protocols, such as Zebra medium access control (Z-MAC), advertisement-based MAC (A-MAC), Speck-MAC, adaptive duty cycle SMAC (ADC-SMAC), and low-power real-time medium access control (LPR-MAC). The simulation results indicate that BN-MAC is a robust and energy-efficient protocol that outperforms other hybrid MAC protocols in the context of quality of service (QoS) parameters, such as energy consumption, latency, throughput, channel access time, successful delivery rate, coverage efficiency, and average duty cycle.

Clock Compensation Strategy in Train Ethernet

How to degrade the delay time of gateway queuing with heavy network load in train communication networks is the key to guarantee the stability of the train. This article models the queue scheduling problem modeling into reinforcement learning process, puts forward an adaptive weights learning polling scheduling algorithm with the purpose of dynamic scheduling for vehicle gateway node queue. With the compare of algorithm in this paper and weighted round-robin scheduling algorithm in adequate and inadequate bandwidth resources situations, we have approved the superiority of the algorithm in this paper.

Alamut: a high‐performance network intrusion detection system in support of virtualized environments

ABSTRACTOne of the benefits of virtualization technology is the provision of secure and isolated computing environments on a single physical machine. However, the use of virtual machines for this purpose often degrades the overall system performance that is due to emulation costs, for example, packet filtering on every virtual machine. To allow virtual machines to be favorably used as before for the provision of secure environments but with comparably less performance degradation, we propose a new architecture called Alamut in this paper for restructuring any typical network intrusion detection system (NIDS) to run in a Xen‐based virtual execution environment. In the proposed architecture, primitive mechanisms for implementing the security concerns of typical NIDSs such as signature matching are placed at the kernel level of driver domain (dom0), whereas security policies and management modules are kept in user space of that domain. Separation of mechanisms from policies allows network packets to be verified at the kernel level first hand more efficiently without requiring costly context switches to push them to user space for validation. In addition, system administrators can easily define new policies at user level and determine on which virtual machines these policies should be enforced. A proof‐of‐concept implementation of Alamut has been prototyped on the Xen hypervisor using Bro open‐source NIDS. Experimental results show approximately 3.5‐fold increase in the overall system performance when our prototype is run compared with when Bro is run. Results also show 19% improvement in network throughput. The comparison of Alamut with Snort with the same set of signatures and attacks shows that our prototyped NIDS has lower processor utilization and has captured more packets in heavy network loads. Copyright © 2013 John Wiley &amp; Sons, Ltd.

DATCP: deadline-aware TCP for the commoditized data centers

Many flows in data centers have deadlines and missing deadlines would hurt application performance such as affecting response quality in Web applications or delaying computing jobs in MapReduce-like systems. However, transmission control protocol (TCP) which is widely used in data centers now cannot provide deadline-aware transmission service. Service differentiation only distinguishes flows with different priority but is unable to guarantee completion time. In this paper, we propose a new protocol named deadline-aware TCP (DATCP) to provide deadline-aware transmission service for the commoditized data centers, which can be used as a flexible method for flow scheduling. DATCP combines flow urgency and importance to calculate precedence. Flow urgency is dynamically adjusted according to the gap between desired rate and actual throughput. Setting importance can avoid starving the important but no-urgent flows. Furthermore, a flow quenching method is presented which allows as many high precedence flows as possible to meet deadlines under heavy network load. By extensive simulations, the performance of DATCP was evaluated. Simulation results show that DATCP can make flows meet deadlines effectively.