High Rate Packet Research Articles

A Long Short-Term Memory-Based Model for Kinesthetic Data Reduction

This paper proposes a novel mathematical model for teleoperation over communication networks. For teleoperation over a communication network, a high packet rate can result in inefficient data transmission and cross-traffic problems, leading to extra delay and jitter. This paper proposes an LSTM-based mathematical model which focuses on kinesthetic data reduction without loss of transparency during the transmission process through joint training combined with haptic data and perceptual deadband. Since the LSTM network can deal with a time series of haptic data, we further test the system performance through practically collected data. We investigate the packet rate and perceptual transparency of the proposed mathematical model by comparing with the conventional deadband. Additionally, we compare the proposed mathematical model with the perceptual deadband-based codecs. Simulation results show that the proposed solution further reduces the packet rate when dealing with haptic data without noticeable distortion. Also, comparing with the current JND perceptual threshold, the proposed mathematical model helps improve the practicality of the bilateral teleoperation system without losing transparency.

Universal and Accurate Sketch for Estimating Heavy Hitters and Moments in Data Streams

In computer networks, traffic measurement is a module in a network probe to measure flow-level statistics from an IP packet stream, which are the basis for network performance monitoring and malicious activity detection. This module extracts the flow IDs from incoming IP packets, classifies packets into flows, and counts the number of packets (or bytes) for each flow. It is a great challenge to measure the per-flow statistics for a high-speed network device, using only the size-limited SRAM on its line cards. Therefore, many algorithms using sublinear memory have been proposed, such as CountMin and CountSketch. However, most of previous algorithms are designed for specific measurement tasks. To obtain multiple types of statistics, people have to deploy multiple sketches, which demands more resources of a network device. It is useful to design a universal sketch that can track not only the top- <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> largest individual flows (called heavy hitters) but also the overall traffic distribution statistics (called moments). Prior work named UnivMon successfully tackled this ambitious quest. However, it incurs large and variable per-packet processing overhead, which may result in a significant throughput bottleneck in high-rate packet stream, given that each packet requires 33 hashes and 32 memory accesses on average and many times of that in the worst case. To address this performance issue, we fundamentally redesign the solution architecture from hierarchical sampling to new progressive sampling and from CountSketch to new GenericCM, which ensure that per-packet overhead is a small constant (5 hashes and 8 memory accesses in the worst case), making it more suitable for online operations, especially for hardware pipeline implementation. This new design also makes effort to reduce memory footprint or equivalently improve measurement accuracy under the same memory. Our experiments show that our solution reduces measurement error by roughly 98.1% for second-order moment and by 91.5% for entropy, when given the same 0.2MB memory as UnivMon.

Reinforcement Learning-Aided Edge Intelligence Framework for Delay-Sensitive Industrial Applications.

With the advancement in next-generation communication technologies, the so-called Tactile Internet is getting more attention due to its smart applications, such as haptic-enabled teleoperation systems. The stringent requirements such as delay, jitter, and packet loss of these delay-sensitive and loss-intolerant applications make it more challenging to ensure the Quality of Service (QoS) and Quality of Experience (QoE). In this regard, different haptic codec and control schemes were proposed for QoS and QoE provisioning in the Tactile Internet. However, they maximize the QoE while degrading the system’s stability under varying delays and high packet rates. In this paper, we present a reinforcement learning-based Intelligent Tactile Edge (ITE) framework to ensure both transparency and stability of teleoperation systems with high packet rates and variable time delay communication networks. The proposed ITE first estimates the network challenges, including communication delay, jitter, and packet loss, and then utilizes a Q-learning algorithm to select the optimal haptic codec scheme to reduce network load. The proposed framework aims to explore the optimal relationship between QoS and QoE parameters and make the tradeoff between stability and transparency during teleoperations. The simulation result indicates that the proposed strategy chooses the optimal scheme under different network impairments corresponding to the congestion level in the communication network while improving the QoS and maximizing the QoE. The end-to-end performance of throughput (1.5 Mbps) and average RTT (70 ms) during haptic communication is achieved with a learning rate and discounted factor value of 0.5 and 0.8, respectively. The results indicate that the communication system can successfully achieve the QoS and QoE requirements by employing the proposed ITE framework.

A Survey of Low Rate DDoS Detection Techniques Based on Machine Learning in Software-Defined Networks

Software-defined networking (SDN) is a new networking paradigm that provides centralized control, programmability, and a global view of topology in the controller. SDN is becoming more popular due to its high audibility, which also raises security and privacy concerns. SDN must be outfitted with the best security scheme to counter the evolving security attacks. A Distributed Denial-of-Service (DDoS) attack is a network attack that floods network links with illegitimate data using high-rate packet transmission. Illegitimate data traffic can overload network links, causing legitimate data to be dropped and network services to be unavailable. Low-rate Distributed Denial-of-Service (LDDoS) is a recent evolution of DDoS attack that has been emerged as one of the most serious vulnerabilities for the Internet, cloud computing platforms, the Internet of Things (IoT), and large data centers. Moreover, LDDoS attacks are more challenging to detect because this attack sends a large amount of illegitimate data that are disguised as legitimate traffic. Thus, traditional security mechanisms such as symmetric/asymmetric detection schemes that have been proposed to protect SDN from DDoS attacks may not be suitable or inefficient for detecting LDDoS attacks. Therefore, more research studies are needed in this domain. There are several survey papers addressing the detection mechanisms of DDoS attacks in SDN, but these studies have focused mainly on high-rate DDoS attacks. Alternatively, in this paper, we present an extensive survey of different detection mechanisms proposed to protect the SDN from LDDoS attacks using machine learning approaches. Our survey describes vulnerability issues in all layers of the SDN architecture that LDDoS attacks can exploit. Current challenges and future directions are also discussed. The survey can be used by researchers to explore and develop innovative and efficient techniques to enhance SDN’s protection against LDDoS attacks.

A New QoS Adaptive Multi-path Routing for Video Streaming in Urban VANETs Integrating Ant Colony Optimization Algorithm and Fuzzy Logic

The multi-path transmission is an appropriate transmission method for high data rate packets like video streaming. To provide video streaming with high quality, the video packets are divided into different frames for transmitting through various paths. Nevertheless, regarding the results of numerous inherent features of vehicular ad hoc networks (VANETs), designing a stable and efficient routing protocol is difficult for various applications of VANETs. In particular, the dynamic nature of topology and intermittent connectivity make maintaining the quality of service (QoS) task very difficult. To provide QoS to entertainment applications and traffic safety in VANET, we offer a routing protocol based on the adaptive junction with QoS support regarding packet delivery ratio (PDR), connectivity probability, and delay. To establish the best QoS routes, we considered the equivalent routing problem as the optimization problem and then proposed an algorithm based on ant colony optimization for solving it. Moreover, a fuzzy logic-based algorithm was employed to select the best next-hop vehicle by incorporating multiple metrics associated with the vehicle’s position, link quality, and vehicle mobility. The transmission control protocol (TCP) and user datagram protocol (UDP) are the two most known transference layer protocols. In this study, we provide an adaptation of TCP and UDP protocols for video streaming in VANET. To indicate the behavior of our suggested method, the network simulator NS-2 is used for the simulation. The simulation findings show the efficiency of our proposed approach in comparison with the existing research works in terms of the average packet delivery ratio, end-to-end delay, overhead, and peak signal to noise ratio.

Software Packet-Level Network Analytics at Cloud Scale

As networks grow in speed, scale, and complexity, operating them reliably requires continuous monitoring and increasingly sophisticated analytics. Because of these requirements, the platforms that support analytics in cloud-scale networks face demands for both higher throughput (to keep up with high packet rates) and increased generality and programmability (to cover a wider range of applications). Recent proposals have worked toward these goals by offloading analytics application logic to line-rate programmable data plane hardware, as scaling existing software analytics platforms is prohibitively expensive. The rigid design and constrained resources of data plane devices, however, fundamentally limit the types of analysis and the number of tasks that can run concurrently. In this article, we demonstrate that generality need not be sacrificed for high performance. Rather than offloading entire analytics applications to hardware, the core idea of our work is to offload only critical preprocessing tasks that are shared among applications (e.g., load balancing) to a line-rate hardware frontend while optimizing the core analytics software to exploit properties of network analytics workloads. Based on this design, we present Jetstream, a hybrid platform for network analytics that can run custom software-based analytics pipelines at throughputs of up to 250 million packets per second on a 16-core commodity server. Jetstream makes sophisticated, network-wide packet analytics feasible without compromising on generality or performance.

Uplink NOMA in Body Area Networks With Simple Node Pairing Strategies

In body area networks (BANs), a node placed near the hub has a significant signal strength advantage. This issue, known as near-far problem, indicates that a near node may serve as a relay for a far node. Currently, the established orthogonal multiple access (OMA) requires the allocation of a specific resource block to a single node. In contrast, the emerging non-orthogonal multiple access (NOMA) allows simultaneous transmissions and uses successive interference cancellation (SIC) to recover the signals. For the first time in the literature, we introduce NOMA in BANs, to allow simultaneous transmissions from two nodes at a time, with distinguishable power levels. The crucial part of this technique is the dynamic pairing of the nodes by the hub. First, we review the fundamentals of uplink NOMA and we propose an access protocol in compliance with IEEE 802.15.6-2012, which combines random OMA with scheduled NOMA. Next, we propose three node pairing strategies, namely a random, a correlative and a conditional one, to meet the requirements of several applications. These strategies are evaluated with packet level simulations in OMNeT++. Our results show that the proposed scheme outperforms the basic OMA scheme at high packet rates and provides a feasible direction for novel relay-based applications.

Experimental evaluation of haptic data communication with prediction

The transmission of haptic data is relatively challenging in multimedia communication. In this research study, the methods are presented for exploiting the properties of human haptic perception for data reduction of haptic data transmission. Packet-switched communication of haptic data is characterized by high packet rates on the communication channel. The quality of the internet-based haptic tele-control/tele-presence systems is highly dependent on the quality of the communication channel between the operator and the remote site, and on the delay jitter in the data exchange. The proposed research work is evaluated experimentally using a Geomagic Touch (previously PHANTOM Sensable Omni) haptic device with a sphere as a virtual model. Four experiments were conducted to evaluate the proposed research study. In the first experiment the JND Weber’s law is applied on sent force values only, while in the second experiment, the force calculation algorithm has been modified to include human movement velocity. The third experiment discusses the use of JND on the sent velocity values. The evaluation of the human’s perception shows that the proposed modification to the basic dead-band approach highly reduces the number of sent packets with minimal disturbance in haptic feeling. Further enhancements using prediction techniques have also been introduced in the fourth experimental evaluation. The linear predictions are added to the above proposed reduction methods. Combining the dead-band approach with a fast, configurable and accurate prediction algorithm enables a significant reduction in the amount of data sent across the network. The reduction is estimated to be 85%, while preserving the original data structure.

P4-To-VHDL: Automatic generation of high-speed input and output network blocks

High-performance embedded architectures typically contain many stand-alone blocks which communicate and exchange data; additionally a high-speed network interface is usually needed at the boundary of the system. The software-based data processing is typically slow which leads to a need for hardware accelerated approaches. The problem is getting harder if the supported protocol stack is rapidly changing. Such problem can be effectively solved by the Field Programmable Gate Arrays and high-level synthesis which together provide a high degree of generality. This approach has several advantages like fast development or possibility to enable the area of packet-oriented communication to domain oriented experts. However, the typical disadvantage of this approach is the insufficient performance of generated system from a high-level description. This can be a serious problem in the case of a system which is required to process data at high packet rates. This work presents a generator of high-speed input (Parser) and output (Deparser) network blocks from the P4 language which is designed for the description of modern packet processing devices. The tool converts a P4 description to a synthesizable VHDL code suitable for the FPGA implementation. We present design, analysis and experimental results of our generator. Our results show that the generated circuits are able to process 100 Gbps traffic with fairly complex protocol structure at line rate on Xilinx Virtex-7 XCVH580T FPGA. The approach can be used not only in networking devices but also in other applications like packet processing engines in embedded cores because the P4 language is device and protocol independent.

Low-Rate DDoS Attack Detection Using Expectation of Packet Size

Low-rate Distributed Denial-of-Service (low-rate DDoS) attacks are a new challenge to cyberspace, as the attackers send a large amount of attack packets similar to normal traffic, to throttle legitimate flows. In this paper, we propose a measurement—expectation of packet size—that is based on the distribution difference of the packet size to distinguish two typical low-rate DDoS attacks, the constant attack and the pulsing attack, from legitimate traffic. The experimental results, obtained using a series of real datasets with different times and different tolerance factors, are presented to demonstrate the effectiveness of the proposed measurement. In addition, extensive experiments are performed to show that the proposed measurement can detect the low-rate DDoS attacks not only in the short and long terms but also for low packet rates and high packet rates. Furthermore, the false-negative rates and the adjudication distance can be adjusted based on the detection sensitivity requirements.

The IX Operating System

The conventional wisdom is that aggressive networking requirements, such as high packet rates for small messages and μs-scale tail latency, are best addressed outside the kernel, in a user-level networking stack. We present ix , a dataplane operating system that provides high I/O performance and high resource efficiency while maintaining the protection and isolation benefits of existing kernels. ix uses hardware virtualization to separate management and scheduling functions of the kernel (control plane) from network processing (dataplane). The dataplane architecture builds upon a native, zero-copy API and optimizes for both bandwidth and latency by dedicating hardware threads and networking queues to dataplane instances, processing bounded batches of packets to completion, and eliminating coherence traffic and multicore synchronization. The control plane dynamically adjusts core allocations and voltage/frequency settings to meet service-level objectives. We demonstrate that ix outperforms Linux and a user-space network stack significantly in both throughput and end-to-end latency. Moreover, ix improves the throughput of a widely deployed, key-value store by up to 6.4× and reduces tail latency by more than 2× . With three varying load patterns, the control plane saves 46%--54% of processor energy, and it allows background jobs to run at 35%--47% of their standalone throughput.

Energy Prediction for Teleoperation Systems That Combine the Time Domain Passivity Approach with Perceptual Deadband-Based Haptic Data Reduction.

We study the combination of the perceptual deadband (PD)-based haptic packet rate reduction scheme with the time domain passivity approach (TDPA) for time-delayed teleoperation and propose a novel energy prediction (EP) scheme that deals with the conservative behavior of the resulting controller. The PD approach leads to irregular packet transmission, resulting in degraded system transparency and reduced teleoperation quality when the PD approach is combined with the TDPA. The proposed method (PD+TDPA+EP) adaptively predicts the system energy during communication interruptions and allows for larger energy output. This achieves less conservative control and improves the teleoperation quality. Evaluation of the displayed impedance shows that the PD+TDPA+EP method achieves improved system transparency, both objectively and subjectively, when compared with related approaches from literature. According to a subjective user study, the PD+TDPA+EP method allows for a high packet rate reduction (up to 80 percent) without noticeably distorting the perceived interaction quality. We also show that the PD+TDPA+EP method is preferred over related approaches from literature in a direct comparison test. Thus, with the proposed PD+TDPA+EP method, a high data reduction and a high teleoperation quality are simultaneously achieved for time-delayed teleoperation.

Pktgen: Measuring performance on high speed networks

Pktgen is a tool for high-speed packet generation and testing. It runs in the Linux kernel, and is designed to accommodate a wide range of network performance tests. Pktgen consists of a packet generator, a receiver, and a protocol that defines the format for test packets. This paper describes the design of pktgen, and discusses its usage as a capable performance testing tool. The design is focused on performance, and in order to generate packets at high packet rate, pktgen takes advantage of multicore systems and multi-queue features on modern network cards. To support realistic heterogeneous traffic patterns, pktgen can generate multiple simultaneous UDP flows. Pktgen supports measurements of throughput, latency, packet inter-arrival time, and packet delay variation with a wide-range of fine-grained user specified rates.We evaluate and compare pktgen to other tools, and show that it is more than an order of magnitude faster than current Linux applications such as Iperf and Netperf, and has the same upper limit on performance as special-purpose high-speed tools, such as DPDK and Netmap, when it comes to throughput measurements with user-specified rate.

On the impact of mobile node radio architectures on heterogeneous wireless network standards: A performance analysis of LTE–eHRPD mobility

With the availability of various heterogeneous radio access technology standards, how to support multiple heterogeneous radio interfaces in a single device and provide seamless mobility between them is one of key issues in the design and implementation of modern multi-radio mobile devices (like smartphones). The radio architectures of such multi-radio mobile devices can have great impacts on achievable mobility performance. This paper presents a case study that investigates into the impacts of radio architecture design on mobility performance achievable in wireless standard implementation, with a particular focus on heterogeneous radio access mobility between Long Term Evolution (LTE) and enhanced High Rate Packet Data (eHRPD). We present an in-depth overview of handover procedures in standards and their achievable performance enhancements from the perspective of device radio architectures. We consider three radio architectures (single transmission/reception, single transmission dual reception, and dual transmission/reception architectures) and provide a performance analysis to compare the three architectures in terms of handover delay and energy consumption. We also discuss supportability for advanced flow-precision mobility under the different radio architectures, and present a comparison with another vertical handover technology.

A Study on High Rate Shrew DDOS Attack

Denial of Service attacks is frequently presenting an increasing threat to the global inter-networking infrastructure in networking area and scenario. The algorithm for Transfer Control Protocol (TCP) congestion control is highly efficient for the various networking areas and operations as well its internal assumption of end-system cooperation results is well prone to attack by high-rate flows. A Shrew attack uses the concept of a low-rate burst which is carefully designed to use the TCP’s retransmission timeout mechanism in an unfair way and can affect the bandwidth of a TCP flow in a smooth manner without coming into appearance as an intruder. A Shrew attack has further classifications such as a low rate shrew attack or a high rate shrew attack. A high rated shrew attack uses the concept of timely sending high rate packet stream in low frequency. Such attack can affect the performance of a network to a large extent.

Downlink Throughput Enhancement of a Cellular Network using Two-Hop User-Deployable Indoor Relays

In this paper, we propose two-hop user-deployable indoor relays for CDMA systems with signal bandwidth 5 MHz or less. We assume concurrent relaying wherein macro base stations (BS) are allowed to transmit when a relay is forwarding. The two-hop indoor relay is studied using a generalized model for outdoor-to-indoor communication in a frequency-selective channel in the presence of co-channel interference. We propose an equalize-and-forward relay that employs multiple-antenna interference-aware equalization before forwarding. We present a unified solution to minimize interference (due to concurrent transmissions) at the relay-served destination and the BS-served destination, using power control and exploiting spatial reuse. A systematic method to enable indoor relays with the proposed interference management solutions is presented that is applicable to existing CDMA systems, without additional requirements in Access Terminal (AT)/User Equipment. Simulation results in the context of High Rate Packet Data (HRPD) CDMA network shows that the ATs served through the indoor relays experience throughput improvement by at least a factor of two, whereas the ATs served by the BS experienced a gain of around 6% in spite of concurrent relaying. The results also show an overall system throughput gain upto 30% when compared to the existing network with only single hops.

QFQ: Efficient Packet Scheduling With Tight Guarantees

Packet scheduling, together with classification, is one of the most expensive processing steps in systems providing tight bandwidth and delay guarantees at high packet rates. Schedulers with near-optimal service guarantees and <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">O</i> (1) time complexity have been proposed in the past, using techniques such as timestamp rounding and flow grouping to keep their execution time small. However, even the two best proposals in this family have a per-packet cost component that is linear either in the number of groups or in the length of the packet being transmitted. Furthermore, no studies are available on the actual execution time of these algorithms. In this paper we make two contributions. First, we present Quick Fair Queueing (QFQ), a new <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">O</i> ( 1) scheduler that provides near-optimal guarantees and is the first to achieve that goal with a truly constant cost also with respect to the number of groups and the packet length. The QFQ algorithm has no loops and uses very simple instructions and data structures that contribute to its speed of operation. Second, we have developed production-quality implementations of QFQ and of its closest competitors, which we use to present a detailed comparative performance analysis of the various algorithms. Experiments show that QFQ fulfills our expectations, outperforming the other algorithms in the same class. In absolute terms, even on a low-end workstation, QFQ takes about 110 ns for an enqueue()/dequeue() pair (only twice the time of DRR, but with much better service guarantees).

Scalable software architecture for distributed MMORPG traffic generation based on integration of UrBBaN-Gen and IMUNES

We present a scalable software architecture for distributed traffic generation capable of producing Massively Multiplayer Online Role-Playing Game (MMORPG) packet flows in a statistically accurate manner for thousands of concurrent players. The main challenge, to achieve truly massive scale traffic generation, has been achieved by introducing kernel based virtualization, pioneered by the network simulator/emulator IMUNES, into the User Behaviour Based Network Traffic Generation (UrBBan-Gen, introduced in our earlier work). The UrBBan-Gen software architecture consists of four modules: Service repository, Control function and user interface, Behaviour process, and Traffic generation process. IMUNES has been integratedinto the virtualization part of the Traffic generation process,which has resulted in two improvements: 1) increasing thenumber of generated packet flows while accurately replicating the required statistical properties, and, 2) introducing the ability to run various network scenarios in simulated, as well as real networks, under realistic traffic loads. With respect to the traffic generation capabilities of the previous version of UrBBan-Gen, which was based on Linux containers, the IMUNES based solution demonstrates higher scalability, lower packet loss rates, and lower CPU load for both the UDP traffic at high packet rate and “thin” TCP traffic flows typical for MMORPGs.

Comparative of Multiplexing Policies for Online Gaming in Terms of QoS Parameters

This letter compares different policies for multiplexing the traffic of online games. In order to achieve bandwidth savings and to alleviate the high packet rate, headers are compressed and a number of native packets are included into a bigger one, using PPPMux and an L2TP tunnel. Small and controlled delays and jitter are added due to retention at the queue of the multiplexer. The policies are compared using real traffic traces of a popular game, and the results show that the savings are significant, while the impairments are not severe.

Combining Contact Models With Perceptual Data Reduction for Efficient Haptic Data Communication in Networked VEs

In cooperative networked virtual environments, haptic rendering enables joint haptic interaction with virtual objects and, thus, shared touch experiences among multiple users. In case of an underlying packet-based communication network (e.g., the Internet), minimizing the end-to-end delay, with the goal of preventing instability of the involved control loops, results in high packet rates. Previously proposed perceptual data reduction approaches address this challenge and satisfy the strict delay constraints. However, significantly increased packet rates still occur during contact events. We present a novel event-based coding scheme based on a distributed haptic-rendering framework that integrates model-based distributed haptic rendering with perceptual data reduction. We also present a comprehensive haptic contact model for signals with multiple degrees of freedom. Furthermore, we show that the integration of event-triggered force transient models from event-based haptics into our local contact model is instrumental for generating convincing haptic feedback. Psychophysical experiments reveal that the approach presented herein allows us to push the data reduction performance beyond what is normally achievable by perceptual data reduction schemes alone while significantly improving the quality of haptic contact feedback.