A Survey on Client Throughput Prediction Algorithms in Wired and Wireless Networks

Network communication has become a part of everyday life, and the interconnection among devices and people will increase even more in the future. Nevertheless, prediction of Quality of Service parameters, particularly throughput, is quite a challenging task. In this survey, we provide an extensive insight into the literature on Transmission Control Protocol throughput prediction. The goal is to provide an overview of the used techniques and to elaborate on open aspects and white spots in this area. We assessed more than 35 approaches spanning from equation-based over various time smoothing to modern learning and location smoothing methods. In addition, different error functions for the evaluation of the approaches as well as publicly available recording tools and datasets are discussed. To conclude, we point out open challenges especially looking in the area of moving mobile network clients. The use of throughput prediction not only enables a more efficient use of the available bandwidth, the techniques shown in this work also result in more robust and stable communication.

Effective TCP Flow Management Based on Hierarchical Feedback Learning in Complex Data Center Network

Many studies focusing on improving Transmission Control Protocol (TCP) flow control realize a more effective use of bandwidth in data center networks. They are excellent ways to more effectively use the bandwidth between clients and back-end servers. However, these schemes cannot achieve the total optimization of bandwidth use for data center networks as they do not take into account the path design of TCP flows against a hierarchical complex structure of data center networks. To address this issue, this paper proposes a TCP flow management scheme specified a hierarchical complex data center network for effective bandwidth use. The proposed scheme dynamically controls the paths of TCP flows by reinforcement learning based on a hierarchical feedback model, which obtains an optimal TCP flow establishment policy even if both the network topology and link states are more complicated. In evaluation, the proposed scheme achieved more effective bandwidth use and reduced the probability of TCP incast up to 30% than the conventional TCP flow management schemes: Variant Load Balancing (VLB), Equal Cost Multi Path (ECMP), and Intelligent Forwarding Strategy Based on Reinforcement Learning (IFS-RL) in the complex data center network.

Design and simulation of video monitoring structure over TCP/IP system using MATLAB

Video monitoring systems are undergoing an evolution from conventional analog to digital clarification to provide better rate and security over internet protocols. In addition, analog surveillance becomes insufficient to face enormous demand of security of system contains more than hundreds of camera often deployed in hotels environments far away from room control. This paper presents the design and simulation of a video monitoring scheme in excess of a transmission control protocol/internet protocol (TCP/IP) system using MATLAB. Sophisticated cameras could record directly high-definition digital videos based on digital technology which simply communicate the control room relaying on ordinary internet protocol infrastructure networks. This technology provides a flexible network interface over a wide variety of heterogeneous technology networks. Though, the acceptance of IP designed for video monitoring pretense severe difficulties in terms of power processing, system dependability, required bandwidth, and security of networks. The advantage of IP based video monitoring system has been investigated over conventional analog systems and the challenges of the method are described. The open research issues are still requiring a final solution to permits complete abandon against conventional technology of analog methods. In conclusion, the method to tackle the purpose of video monitoring in actual operation is proposed and verified properly by means of model simulation.

Distributed reflection denial of service attack: A critical review

As the world becomes increasingly connected and the number of users grows exponentially and “things” go online, the prospect of cyberspace becoming a significant target for cybercriminals is a reality. Any host or device that is exposed on the internet is a prime target for cyberattacks. A denial-of-service (DoS) attack is accountable for the majority of these cyberattacks. Although various solutions have been proposed by researchers to mitigate this issue, cybercriminals always adapt their attack approach to circumvent countermeasures. One of the modified DoS attacks is known as distributed reflection denial-of-service attack (DRDoS). This type of attack is considered to be a more severe variant of the DoS attack and can be conducted in transmission control protocol (TCP) and user datagram protocol (UDP). However, this attack is not effective in the TCP protocol due to the three-way handshake approach that prevents this type of attack from passing through the network layer to the upper layers in the network stack. On the other hand, UDP is a connectionless protocol, so most of these DRDoS attacks pass through UDP. This study aims to examine and identify the differences between TCP-based and UDP-based DRDoS attacks.

Machine-Learning-Enabled DDoS Attacks Detection in P4 Programmable Networks

Distributed Denial of Service (DDoS) attacks represent a major concern in modern Software Defined Networking (SDN), as SDN controllers are sensitive points of failures in the whole SDN architecture. Recently, research on DDoS attacks detection in SDN has focused on investigation of how to leverage data plane programmability, enabled by P4 language, to detect attacks directly in network switches, with marginal involvement of SDN controllers. In order to effectively address cybersecurity management in SDN architectures, we investigate the potential of Artificial Intelligence and Machine Learning (ML) algorithms to perform automated DDoS Attacks Detection (DAD), specifically focusing on Transmission Control Protocol SYN flood attacks. We compare two different DAD architectures, called Standalone and Correlated DAD, where traffic features collection and attack detection are performed locally at network switches or in a single entity (e.g., in SDN controller), respectively. We combine the capability of ML and P4-enabled data planes to implement real-time DAD. Illustrative numerical results show that, for all tested ML algorithms, accuracy, precision, recall and F1-score are above 98% in most cases, and classification time is in the order of few hundreds of $$\upmu \text {s}$$ μ s in the worst case. Considering real-time DAD implementation, significant latency reduction is obtained when features are extracted at the data plane by using P4 language. Graphic Abstract

Detection of Anonymising Proxies using Machine Learning

Network Proxies and Virtual Private Networks (VPN) are tools that are used every day to facilitate various business functions. However, they have gained popularity amongst unintended userbases as tools that can be used to hide mask identities while using websites and web-services. Anonymising Proxies and/or VPNs act as an intermediary between a user and a web server with a Proxy and/or VPN IP address taking the place of the user’s IP address that is forwarded to the web server. This paper presents computational models based on intelligent machine learning techniques to address the limitations currently experienced by unauthorised user detection systems. A model to detect usage of anonymising proxies was developed using a Multi-layered perceptron neural network that was trained using data found in the Transmission Control Protocol (TCP) header of captured network packets

Distributed Database System Optimization for Improved Service Delivery in Mobile and Cloud BigData Applications

Many issues associated with managing centralized database include data isolation, redundancy, inconsistency, and atomicity of updates, among others; however, distributed database implementation over high-performance compute nodes maximizes information value across the networks. Also, analysis of bigdata generated/consumed over the mobile Internet, Internet of Things (IoT) and cloud computations necessitates low-latency reads and updates over cloud clusters. Conventionally, services in distributed systems demand optimized transactions. This paper examines transaction generation over distributed storage pool using suggested reference architectures of fragmentation using hybrid semi-join operations to offer mobility transparency as an additional ingredient of integrity transparency offer of DDBMS. Distributed storage pool is simulated using configured WLAN to activate multiple file transfers concurrently, engaging mobile nodes and large file sizes. Major functionality desired in the storage pool is improvised by storage virtualization whereby a global schema query optimizer effects transaction management to characterized latency-driven throughputs achieved by joint optimization of network and storage virtualization. Measurements and evaluations gave the best overall performance of low-latency reads and updates using the provisioned mobile-transmission control protocol (M-TCP). Appreciable improvement in service delivery is offered using distributed storage pool (DSP) facilitated with hybridized RAID construction and copy mechanisms. Improved response-time and speed-up transmissions evidently showed low-latency read and update transactions, depicting improved service delivery. Evaluating the DDBMS model simulated in the DSP architecture, all complexity (overheads) associated with conventional shared systems were minimized.

BA-MPCUBIC: Bottleneck-Aware Multipath CUBIC for Multipath-TCP

The Congestion Control Algorithm (CCA) in the Multipath Transmission Control Protocol (MPTCP) is fundamental to controlling the flow of data through multiple subflows (SF) simultaneously. The MPTCP CCA has two design goals: first, always ensure better throughput than single path TCP (SPTCP) flows, and second, collectively, MPTCP SFs going through a shared bottleneck (SB) should occupy bandwidth fairly, i.e., close to the bandwidth occupied by an SPTCP flow. Although several MPTCP CCAs exist, they primarily focus on specific scenarios and could not satisfy the design goals in diverse and dynamic scenarios. Recently, CUBIC has become a widely used CCA for SPTCP for its better compatibility with high-speed internet. CUBIC’s effective implementation in the MPTCP is expected to provide improved throughput and fairer behavior, thus satisfying the design goals. However, although the current multipath CUBIC (MPCUBIC) implementation ensures better fairness, it fails to ensure better throughput. We believe the application of same rule for SFs going through an SB and non-shared bottleneck (NSB) makes it difficult for MPCUBIC to adapt to diverse and dynamically changing network scenarios, thus resulting in poor throughput. Therefore, we present an improved version of MPCUBIC, namely bottleneck-aware MPCUBIC (BA-MPCUBIC), to resolve the throughput issue. First, we deploy an innovative bottleneck detection method that successfully differentiates between an SB and NSB based on round-trip-time, enhanced congestion notification, and packet loss. Then, we implement SPTCP CUBIC and MPCUBIC as the CCAs for SFs going through NSBs and SBs, respectively. Extensive emulation experiments demonstrate that the BA-MPCUBIC successfully detects SBs and NSBs with the highest detection accuracy and the lowest detection time compared with other approaches. Moreover, BA-MPCUBIC successfully satisfies the MPTCP design goals in the considered diverse and dynamic scenarios by ensuring both better throughput and fairness.