General TCP State Inference Model From Passive Measurements Using Machine Learning Techniques

Hagos, Desta Haileselassie; Engelstad, Paal E.; Yazidi, Anis; Kure, Øivind

doi:10.1109/access.2018.2833107

Cited by 24 publications

(18 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Passive measurement methodology is a technique of tracking the behavior and characteristics of packet streams where the network is not influenced by injecting extra traffic. More details on the two types of network measurement technique categories (i.e., active and passive) are briefly described in [12]. The RTT seen by a TCP segment is defined as the time a sender waits until it receives a corresponding ACK from the receiver before it sends more data packets.…”

Section: B Passive Rtt Monitoring Methodology and Trace Analysismentioning

confidence: 99%

“…The background traffic for all our experiments are generated using the iperf [8], an open source TCP streaming benchmark, traffic generator on an emulated LAN link where we run each TCP variant by adding a configurable variation of the emulation parameters bandwidth (in Mbit/s), delay (in ms), jitter (in ms) and packet loss (%) within a flow. The values of configuration parameters of the emulator for our samples collection are presented in Table I. The cross-traffic variability and verification of the popular Linux-based network emulator we used, Network Emulator (NetEm) [14], are thoroughly addressed in [12]. Verification of the Emulator: Given that the software emulator is not precise, we can ask: can we trust the network emulator for all the variations of bandwidth, delay, jitter and packet loss values introduced by the emulator irrespective of the measurement we get from TCP stream?…”

Section: B Passive Rtt Monitoring Methodology and Trace Analysismentioning

confidence: 99%

“…(iii) The TCP active probes used in active measurements (such as ping messages) are blocked by firewalls etc. For more details about the difference between active and passive measurement techniques, we refer the reader to our previous work [12].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Deep Learning Approach to Dynamic Passive RTT Prediction Model for TCP

Hagos

Engelstad

Yazid

et al. 2019

2019 IEEE 38th International Performance Computing and Communications Conference (IPCCC)

Self Cite

View full text Add to dashboard Cite

The Round-Trip Time (RTT) is a property of the path between a sender and a receiver communicating with Transmission Control Protocol (TCP) over an IP network and over the public Internet. The end-to-end RTT value influences significantly the dynamics and performance of TCP, which is by far the most used communication protocol. Thus, in communication networks, RTT is an important network performance variable. By measuring the traffic at an intermediate node, a network operator or service provider can estimate the RTT and use the estimation to study and troubleshoot the per-connection characteristics and performance. This paper aims at improving the accuracy and timeliness of the RTT estimation, to help network operators improving their analysis. We propose and evaluate a novel deep learning-based model capable of dynamically predicting at real-time the RTT between the sender and receiver with high accuracy based on passive measurements collected at an intermediate node, taking advantage of the commonly used TCP timestamps. We validate extensively our prediction methodology in a controlled experimental testbed and in a realistic scenario on the Google Cloud platform. We show that our model, which is based on classical deep learning algorithms, gives reasonably effective state-of-the-art performance results across multiple TCP congestion control variants. We also show that the model works well for transfer learning. Even though the RTT prediction model was trained on an emulated network, it performs well also when applied to a realistic scenario setting, as demonstrated in our experimental evaluation.

show abstract

Section: B Passive Rtt Monitoring Methodology and Trace Analysismentioning

confidence: 99%

Section: B Passive Rtt Monitoring Methodology and Trace Analysismentioning

confidence: 99%

See 1 more Smart Citation

A Deep Learning Approach to Dynamic Passive RTT Prediction Model for TCP

Hagos

Engelstad

Yazid

et al. 2019

2019 IEEE 38th International Performance Computing and Communications Conference (IPCCC)

Self Cite

View full text Add to dashboard Cite

show abstract

“…The important part of TCP identification is to construct the probability function. The previous methods infer the back-off factor β for the identification of TCP algorithms [13]. For example, the β of Cubic is 0.7 [2], for Reno is 0.5.…”

Section: Problem Formulationmentioning

confidence: 99%

“…According to the way of data collection, TCP identification can be mainly divided into two categories, namely active detection and passive measurement, where the former relies on observing the behaviors of injected redundant packets and passive measurement relies on the observations at intermediate nodes, without affecting the current network traffic. The passive measurement has minimum effects on network which is more practical than active detection, so is commonly used in recent work [6], [12], [13].…”

Section: Introductionmentioning

confidence: 99%

Passive TCP Identification for Wired and Wireless Networks: A Long-Short Term Memory Approach

Chen

et al. 2019

2019 15th International Wireless Communications &Amp; Mobile Computing Conference (IWCMC)

View full text Add to dashboard Cite

Transmission control protocol (TCP) congestion control is one of the key techniques to improve network performance. TCP congestion control algorithm identification (TCP identification) can be used to significantly improve network efficiency. Existing TCP identification methods can only be applied to limited number of TCP congestion control algorithms and focus on wired networks. In this paper, we proposed a machine learning based passive TCP identification method for wired and wireless networks. After comparing among three typical machine learning models, we concluded that the 4-layers Long Short Term Memory (LSTM) model achieves the best identification accuracy. Our approach achieves better than 98% accuracy in wired and wireless networks and works for newly proposed TCP congestion control algorithms.

show abstract

An Adaptive Measurement Method for Flow Traffic in Software Defined Networking

Huo¹,

Jiang

Zhu

et al. 2019

Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering

View full text Add to dashboard Cite

In Software Defined Networking (SDN), the fine-grained measurements are crucial for network management and design. However, the measurement overhead and accuracy are contradiction, how to accurately measure the network traffic with low overhead has become a hot topic. Artificial Intelligence (AI) has been used to predict the traffic in networks. Then, we propose an AI-based Lightweight Adaptive Measurement Method (ALAMM) for traffic measurement in SDN with low overhead and high measurement accuracy. Firstly, we use measurements in the front to train the AI-based traffic prediction model and utilize the model to predict traffic in SDN. Then, we obtain the sequence of sampling points by judging the change of traffic prediction and send the measurement primitive to switches to obtain coarse-grained measurements. At last, we utilize the interpolation theory to fill the coarse-grained measurement and propose an optimization function to optimize the fine-grained measurement. Simulation results show that the ALAMM is feasible, and the measurement overhead of ALAMM is low.

show abstract

General TCP State Inference Model From Passive Measurements Using Machine Learning Techniques

Cited by 24 publications

References 31 publications

A Deep Learning Approach to Dynamic Passive RTT Prediction Model for TCP

A Deep Learning Approach to Dynamic Passive RTT Prediction Model for TCP

Passive TCP Identification for Wired and Wireless Networks: A Long-Short Term Memory Approach

An Adaptive Measurement Method for Flow Traffic in Software Defined Networking

Contact Info

Product

Resources

About