Packet reordering is not pathological network behavior

It is a widely held belief that packet reordering in the Internet is a pathological behavior, or more precisely, that it is an uncommon behavior caused by incorrect or malfunctioning network components. Some studies of Internet traffic have reported seeing occasional packet reordering events and ascribed these events to “route fluttering”, router “pauses” or simply to broken equipment. We have found, however, that parallelism in Internet components and links is causing packet reordering under normal operation and that the incidence of packet reordering appears to be substantially higher than previously reported. More importantly, we observe that in the presence of massive packet reordering transmission control protocol (TCP) performance can be profoundly effected. Perhaps the most disturbing observation about TCP's behavior is that large scale and largely random reordering on the part of the network can lead to self-reinforcingly poor performance from TCP     

Evaluation of bit error rate for packet combining with constellation rearrangement

In this paper, we propose a method for evaluation of the bit error rate (BER) for packet combining based on constellation rearrangement (CoRe). Such mapping diversity scheme, adopted in the high speed downlink packet access (HSDPA), uses Gray‐mapped constellations and is based on suboptimal accumulation of the reliability metrics generated in each of the transmissions. We present an exact model for the logarithmic likelihood ratios (LLR) obtained by means of the so‐called max‐log approximation, and we show that their conditional probability density functions (pdf) are piecewise Gaussian. We then present the derivation of the uncoded BER and illustrate it with simulation results that confirm our formulation. Finally, we propose simplifications which significantly reduce the complexity of the evaluation method and provide results with a very good accuracy; an extension to transmissions over faded channel is also presented. Copyright © 2007 John Wiley & Sons, Ltd.     

Packet reordering analysis for concurrent multipath transfer

Path diversity of different concurrent paths inevitably introduces out‐of‐order packet delivery, which leads to unnecessary retransmissions and a severely lowered data throughput. In order to analyze the fundamental behavior and influencing factors of packet reordering, several sets of simulation‐based multipath experiments are conducted with different aggregating paths. In addition, the relationship between reordering and number of paths are discussed. Furthermore, an end‐to‐end measurement study of packet reordering in concurrent multipath transfer is presented. To simply the analysis process, a cumulative distribution function of transmission delay for each packet is used. The model is proved to be an effective one by contrastive analysis. Finally, some solutions to handle it are provided with a further discussion. Copyright © 2013 John Wiley & Sons, Ltd.     

Optimized link state routing for quality‐of‐service provisioning: implementation,measurement, and performance evaluation

This paper provides a measurement‐based performance evaluation of the Optimized Link State Routing (OLSR) protocol. Two versions of OLSR, OLSR‐ETX and OLSR‐ETT, are implemented and evaluated on a mesh network that we built from off‐the‐shelf commercial components and deployed within our department building. OLSR‐ETX uses the Expected Transmission Count (ETX) metric, whereas OLSR‐ETT uses the Expected Transmission Time (ETT) metric as a means of assessing link quality. The paper describes our implementation process of the ETT metric using the plug‐in feature of OLSRd, and our calculation method of link bandwidth using the packet‐pair technique. A series of measurements are conducted in our testbed to analyze and compare the performance of ETX and ETT metrics deemed useful for quality of service. Our measurements show that OLSR‐ETT outperforms OLSR‐ETX significantly in terms of packet loss, end‐to‐end delay, jitter, route changes, bandwidth, and overall stability, yielding much more robust, reliable, and efficient routing. Copyright © 2012 John Wiley & Sons, Ltd.     

The effect of packet reordering in a backbone link on application throughput

Packet reordering in the Internet is a well-known phenomenon. As the delay and speed of backbone links continue to increase, what used to be a negligible amount of packet reordering may now, combined with some level of dropped packets, cause multiple invocations of fast recovery within a TCP window. This may result in a significant drop in link utilization and hence in application throughput. What adds to the difficulty is that packet reordering is a silent problem. It may result in significant application throughput degradation while leaving little to no trace. In this article we try to measure and quantify the effect of reordering packets in a backbone link that multiplexes multiple TCP flows on application throughput. Different operating systems and delay values as well as various types of flow mixes were tested in a laboratory setup. The results show that only a small percentage of reordered packets, by at least three packet locations, in a backbone link can cause significant degradation of application throughput. Long flows are affected most. Due to the potential impact of this phenomenon, minimization of packet reordering as well as mitigating the effect algorithmically should be considered.     

Feasibility and performance evaluation of a 6LoWPAN‐enabled platform for ubiquitous healthcare monitoring

Technology has revolutionized medical practices by enabling more convenient and non‐intrusive monitoring of patient's health, leading to next generation ubiquitous healthcare (u‐healthcare). The exploitation of the Internet protocol version 6 addressing space along with the miniaturization of electronic devices has fostered providing interoperability and connectivity of wearable sensor devices in wireless body area networks to the Internet of Things. In this paper, we propose to integrate the IPv6 over low power wireless personal area network (6LoWPAN) to the u‐healthcare monitoring system architecture. The main objective is to study the feasibility of the 6LoWPAN‐enabled platform in real‐world scenarios dealing with medical data. The performance evaluation of this platform is carried out initially through simulations using OMNet++ and then supported by an experimental study using sensor motes and a customized micro‐computing unit. Performance metrics such as throughput, end‐to‐end delay, packet error rate, and energy consumption are investigated under acute health conditions, where patient's health information has to be sent continuously and at maximum rate to the care provider. The obtained results show that the proposed 6LoWPAN solution fulfills the main quality of service requirements of u‐healthcare applications. Copyright © 2015 John Wiley & Sons, Ltd.     

Dimensioning the packet loss burstiness over wireless channels: a novel metric,its analysis and application

The packet loss burstiness over wireless channels is commonly acknowledged as a key impacting factor on the performance of networking protocols. An accurate evaluation of the packet loss burstiness, which reveals the characteristics and performance of the wireless channels, is crucial to the design of wireless systems and the quality‐of‐service provisioning to end users. In this paper, a simple yet accurate analytical framework is developed to dimension the packet loss burstiness over generic wireless channels. In specific, we first propose a novel and effective metric to characterize the packet loss burstiness, which is shown to be more compact, effective, and accurate than the metrics proposed in existing literature for the same purpose. With this metric, we then develop an analytical framework and derive the closed‐form solutions of the packet loss performance, including the packet loss rate and the loss‐burst/loss‐gap length distributions. Lastly, as an example to show how the derived results can be applied to the design of wireless systems, we apply the analytical results to devise an adaptive packetization scheme. The proposed packetization scheme adaptively adjusts the packet length of transmissions based on the prediction of the packet loss rate and loss‐burst/loss‐gap lengths of the wireless channel. Via extensive simulations, we show that with the proposed packetization scheme, the channel throughput can be enhanced by more than 10% than the traditional scheme.Copyright © 2012 John Wiley & Sons, Ltd.     

Misbehavior analysis of IEEE 802.11 MAC layer in mobile ad hoc network using stochastic reward nets

In this paper, we present a Petri net model for performance evaluation of IEEE 802.11 distributed coordination function as a popular media access control layer protocol in mobile ad hoc network. The goal of this evaluation is to examine this protocol under the existing of misbehavior nodes that selfishly try to grasp common channel in a neighbor area. The presented model consists of 2 separate models based on stochastic reward net (SRN), as a variation of stochastic Petri net. The first model, which is called one node operation model , is supposed for presenting all distributed coordination function operations in a single node such as collision avoidance, request to send/clear to send (RTS/CTS) handshake, and backoff mechanism. The next SRN model, all node operation model , is used for modeling nodes competition for occupying channel in a neighbor area. The models could be adjusted to a dynamic network with any number of nodes, dimension scale, and nodes' speed. For evaluation purpose, 4 distinct attack types implemented by modifying associated transitions in SRN models. The proposed SRN model has been quantified by deriving 2 performances metrics as Throughput and Delay . Both metrics are also compared to the value obtained from NS‐2 in terms of different number of nodes and 3 packet generation rates. Three additional metrics measuring the channel usage are also quantified in terms of different attack strategies using only presented SRN model.     

Considerations for application‐layer multipath transport control

Multipath transport faces a lot of challenges caused by path diversity, network dynamics, and service diversity. An effective end‐to‐end multipath transport control mechanism becomes essential to efficiently utilize multiple paths. On the base of the general framework of multipath transport system based on application‐level relay proposed in our previous work, this paper presents a multipath transport control mechanism supporting various applications with different transmission requirements. We propose a multipath transport protocol suite, which is extensible and suitable for various applications, and a multipath transport control model in which an application‐dependent splitting granularity named flow block is introduced. Two load distribution models are explored: the earliest idle path first load distribution for reliable data transmission to maximize the data throughput and the packet reordering‐controlled load distribution for real‐time data transmission to minimize the packet reordering thereby reducing end‐to‐end delay and packet loss rate of multipath transport. Simulation results show that the proposed models can effectively improve data throughput for applications with reliable transmission requirements and reduce the total packet loss rate of the destination for applications with real‐time transmission requirements.     

粒度自适应的多径流量分割算法

针对多径传输存在的负载均衡和分组乱序之间的矛盾,提出了一种粒度自适应的多径流量分割算法。该算法在“流分片”的基础上,通过对流量特性的分析,将网络流量区分为正常流和强势流,基于“区分服务”的思想,不同流量分别采用适应其流量特性的流片分割粒度。理论分析和仿真结果表明,在考虑路径时延估计误差的情况下,该算法在降低分组乱序的同时具有良好的负载均衡性能。     

Wide-area Internet traffic patterns and characteristics

The Internet is rapidly growing in number of users, traffic levels, and topological complexity. At the same time it is increasingly driven by economic competition. These developments render the characterization of network usage and workloads more difficult, and yet more critical. Few recent studies have been published reporting Internet backbone traffic usage and characteristics. At MCI, we have implemented a high-performance, low-cost monitoring system that can capture traffic and perform analyses. We have deployed this monitoring tool on OC-3 trunks within the Internet MCI's backbone and also within the NSF-sponsored vBNS. This article presents observations on the patterns and characteristics of wide-area Internet traffic, as recorded by MCI's OC-3 traffic monitors. We report on measurements from two OC-3 trunks in MCI's commercial Internet backbone over two time ranges (24-hour and 7-day) in the presence of up to 240,000 flows. We reveal the characteristics of the traffic in terms of packet sizes, flow duration, volume, and percentage composition by protocol and application, as well as patterns seen over the two time scales     

Analysis and performance evaluation of new coding options for space telecommand links ‐ Part I: AWGN channels

We investigate and compare, over the additive white Gaussian noise channel, different options for updating the error correcting code currently used in space mission telecommand links. Besides some more consolidated proposals, using low‐density parity‐check codes, we characterize and assess the performance of alternative schemes, based on parallel turbo codes and soft‐decision decoded Bose‐Chaudhuri‐Hocquenghem (BCH) codes. The analysis considers relevant metrics like the codeword error rate, the frame error rate, and the undetected frame error rate. The considered codes include binary and non‐binary low‐density parity‐check codes, parallel turbo codes and extended BCH codes, with different decoding algorithms. The complexity of the various schemes and possible limits for their application are considered and discussed. Several numerical examples are provided. International Journal of Satellite Communications and Networking. Copyright © 2014 John Wiley & Sons, Ltd.     

Stochastic analysis of Aloha in vehicular ad hoc networks

The aim of this paper is to study the Aloha medium access (MAC) scheme in 1D, linear networks, which might be an appropriate assumption for vehicular ad hoc networks. We study performance metrics based on the signal over interference plus noise ratio assuming power-law mean path-loss and independent point-to-point fading. We derive closed formulas for the capture probability. We consider the presence or the absence of noise and we study performance with outage or with adaptive coding. We carry out the joint optimization of the density of packet progress (in bit-meters) both in the transmission probability and in the transmission range. We also compare the performance of slotted and non-slotted Aloha. We show that in contrast to planar networks the density of packet progress per unit of length does not increase with the network node density.     

Multicast-based mobility: a novel architecture for efficient micromobility

Handover performance is very important when evaluating IP mobility protocols. If not performed efficiently, handover delays, jitters, and packet loss directly impact application performance. We propose a new architecture for providing efficient handover, while being able to coexist with other protocols. We propose a paradigm for multicast-based micromobility (M&M), where a visiting mobile is assigned a multicast address to use while moving within a domain. The multicast address is obtained using algorithmic mapping, and handover is achieved using multicast join/prune mechanisms. This paper outlines a framework for the design and evaluation of micromobility protocols. We define a suite of protocols (called candidate access router set) to enable multiple-access routers to receive traffic for the mobile node. By changing the number of such routers, timing, and buffering parameters, the protocol may be fine-tuned for specific technologies (e.g., 802.11) and handover scenarios. Extensive NS-2 simulations are used to compare M&M with other micromobility schemes-cellular Internet protocol (CIP) and handoff-aware wireless access Internet infrastructure (HAWAII). For proactive handover scenarios, our results show that M&M and CIP show lower handover delay and packet reordering than HAWAII. M&M, however, handles multiple border routers in a domain, where CIP fails. Also, for scenarios of reactive handover and coverage gaps M&M clearly outperforms CIP and HAWAII.     

A performance study for IPv4–IPv6 translation in IP multimedia core network subsystem

A Next Generation Network (NGN) is an advanced, packet‐based network that exploits broadband and QoS‐enabled transport technologies for enabling multimedia services. In NGNs, the principles and requirements of IP Multimedia Sub‐system (IMS), which are used to deliver the desired benefits, should be carefully examined and studied. Internet Protocol version 6 (IPv6) is adopted by Third‐generation Partnership Project (3GPP) to solve address storage problem and provide new features (e.g. plug‐and‐play and mobility) for IMS. However, in the early stage of IPv6 deployment, the existing Voice‐over‐IP (VoIP) networks support Internet Protocol version 4 (IPv4) only. For IPv4–IPv6 interworking between IMS and the existing VoIP networks, the IMS‐Application Level Gateway (IMS‐ALG) and the Transition Gateway (TrGW) are proposed to translate Session Initiation Protocol (SIP) and Real‐time Transport Protocol (RTP) packets, respectively. In this paper, we focus on the IPv4–IPv6 translation for RTP packets, which is the bottleneck of VoIP performance. Specifically, we developed a TrGW called National Information and Communications Initiative (NICI)‐TrGW. In NICI‐TrGW, we perform IPv4–IPv6 translation at the Linux kernel and adjust the header room of kernel‐level packet buffer for each packet to reduce memory‐copy operations. We evaluate the performance of NICI‐TrGW and the existing solutions by using the SmartBits. Our study indicates that NICI‐TrGW outperforms the existing solutions in terms of three different output measures including packet loss rate, maximum throughput, and average latency. Copyright © 2009 John Wiley & Sons, Ltd.     

Characterizing user‐perceived impairment events using end‐to‐end measurements

Measures of quality of service (QoS) must correlate to end‐user experience. For multimedia services, these metrics should focus on the phenomena that are observable by the end‐user. Metrics such as delay and loss may have little direct meaning to the end‐user because knowledge of specific coding and/or adaptive techniques is required to translate delay and loss to the user‐perceived performance. Impairment events, as defined in this paper, are observable by the end‐users independent of coding, adaptive playout or packet loss concealment techniques employed by their multimedia applications. Time between impairments and duration of impairments are metrics that are easily understandable by a network user. Methods to detect these impairment events using end‐to‐end measurements are developed here. In addition, techniques to identify Layer 2 route changes and congestion events using end‐to‐end measurements are also developed. These are useful in determining what caused the impairments. End‐to‐end measurements were conducted for about 26 days on 9 different node pairs to evaluate the developed techniques. Impairments occurred at a high rate on the two paths on which congestion events were detected. On these two paths, congestion occurred for 6–8 hours during the day on weekdays. Impairments caused by route changes were rare but lasted for several minutes. Copyright © 2005 John Wiley & Sons, Ltd.     

Beyond folklore: observations on fragmented traffic

Fragmented IP traffic is a poorly understood component of the overall mix of traffic on the Internet. Many assertions about the nature and extent of fragmented traffic are anecdotal rather than empirical. In this paper we examine the causes and attributes of measured fragment traffic, in particular, the effects of NFS, streaming media, networked video games, tunneled traffic, and the prevalence of packet fragmentation due to improperly configured machines. To understand the prevalence, causes, and effects of fragmented IP traffic, we have collected and analyzed seven multiday traces from four sources. These sources include a university commodity access link, two highly aggregated commercial exchange points, and a local NAP. Although there is no practical method of ascertaining whether any data provide a representative sample of all Internet traffic, we include data sources that cover several different types of WAN with traffic from commercial entities, educational and research institutions, and large government facilities. The dominant causes of fragmentation are streaming media and tunneled traffic. Although rumored to be the main impetus for IP packet fragmentation, NFS is not among the top ten causes.     

Performance evaluation of random power capture on mobile communications

Power assignment schemes are man‐made methods to enhance the capture effect of radio communications. In a previous study, Wen and Yang investigated the combined capture effect of the fixed power assignment scheme, Rayleigh fading, and near–far effect on the performance of packet radios. The performance analysis was limited to an infinite population environment. This assumption is reasonable for a conventional packet radio system with a vast service area. However, for a cellular mobile system, a finite population model should be used. In this paper, we analyse the combined natural and man‐made capture effect on the performance of a cellular system with finite population in each cell. A random power assignment scheme is adopted to produce the man‐made capture. The system throughput and delay are carried out by a Markov model. Some numerical calculations are used to demonstrate the degree of performance improvement. Copyright © 2001 John Wiley & Sons, Ltd.     

Smoothing and transporting video in QoS IP networks

Real‐time traffic such as voice and video, when transported over the Internet, demand stringent quality of service (QoS) requirements. The current Internet as of today is still used as a best effort environment with no quality guarantees. An IP‐based Internet that supports different QoS requirements for different applications has been evolving for the past few years. Video streams are bursty in nature due to the instant variability of the video content being encoded. To help mitigate the transport of bursty video streams with minimal loss of information, rate‐adaptive shapers (RASs) are usually being used to reduce the burstiness and therefore help preserve the desired quality. When transporting video over a QoS IP network, each stream is classified under a specific traffic profile to which it must conform, to avoid packet loss and picture quality degradation. In this paper we study, evaluate and propose RASs for the transport of video over a QoS IP network. We utilize the encoding video parameters for choosing the appropriate configuration needed to support the real‐time transport of Variable Bit Rate (VBR) encoded video streams. The performance evaluation of the different RASs is based on the transport of MPEG‐4 video streams encoded as VBR. The performance is studied based on looking at the effect of various parameters associated with the RASs on the effectiveness of smoothing out the burstiness of video and minimizing the probability of packet loss. Copyright © 2005 John Wiley & Sons, Ltd.     

Network traffic prediction based on INGARCH model

In this paper, we introduce the integer-valued generalized autoregressive conditional heteroscedasticity (INGARCH) as a network traffic prediction model. As the INGARCH is known as a non-linear analytical model that could capture the characteristics of network traffic such as Poisson packet arrival and long-range dependence property, INGARCH seems to be an adequate model for network traffic prediction. Based on the investigation for the traffic arrival process in various network topologies including IoT and VANET, we could confirm that assuming the Poisson process as packet arrival works for some networks and environments of networks. The prediction model is generated by estimating parameters of the INGARCH process and predicting the Poisson parameters of future-steps ahead process using the conditional maximum likelihood estimation method and prediction procedure, respectively. Its performance is compared with those of three different models; autoregressive integrated moving average, GARCH, and long short-term memory recurrent neural network. Anonymized passive traffic traces provided by the Center for Applied Internet Data Analysis are used in the experiment. Numerical results show that the proposed model predicts better than the three models in terms of measurements used in prediction models. Based on the study, we can conclude the followings: INGARCH can capture the characteristics of network traffic better than other statistic models, it is more tractable than neural networks (NNs) overcoming the black-box nature of NNs, and the performances of some statistical models are comparable or even superior to those of NNs, especially when the data is insufficient to apply deep NNs.