A survey on analyzing encrypted network traffic of mobile devices

Bhatia, Ashutosh; Agrawal, Ankit; Bahuguna, Ayush; Tiwari, Kamlesh; Haribabu, K.; Vishwakarma, Deepak

doi:10.1007/s10207-022-00581-y

Cited by 6 publications

(5 citation statements)

References 79 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, the academic and industrial communities have paid much attention to traffic classification [12], [28]- [32]. Traditional techniques for traffic classification, like port-based and deep packet inspection (DPI), are no longer adequate due to the exponential growth in encrypted traffic [33]- [35].…”

Section: Related Workmentioning

confidence: 99%

“…This approach aims to obtain predetermined patterns as signatures for every protocol and application. Subsequently, it classifies traffic by comparing the data in the packet's payload to the predefined signatures [6], [12], [41]. Although this method has high accuracy in classifying unencrypted traffic, it is computationally intensive.…”

Section: B Deep Packet Inspection (Dpi) Approachesmentioning

confidence: 99%

“…It is crucial to consider the temporal characteristics of the data and use a suitable technique for the traffic classification task. Despite the limitations of CNNs, they can still be used to extract features from traffic flow data [1], [12], [17], [18], [21]. Even though deep learning extracts features without human intervention, imbalanced learning data can result in poor results in this approach [26].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

CBS: A Deep Learning Approach for Encrypted Traffic Classification With Mixed Spatio-Temporal and Statistical Features

Seydali,

Khunjush,

Akbari

et al. 2023

IEEE Access

View full text Add to dashboard Cite

With the rapid advancement of the internet and online applications, traffic classification has become an increasingly significant topic in computer networks. Managing network resources, improving service quality, and enhancing cybersecurity are critical. Due to traffic encryption techniques, traditional traffic classification approaches have become ineffective and inaccurate. Therefore, the scientific community considers deep learning a high-performance approach for classifying encrypted traffic. This paper proposes an encrypted traffic classification approach, CBS, based on a deep learning technique. CBS can classify encrypted traffic at two levels using 1D-CNN, attention-based Bi-LSTM, and SAE deep network models. The proposed model classifies traffic types and applications based on a comprehensive set of session and packet-level features. CBS accurately distinguishes traffic classes using spatial, temporal, and statistical features extracted from packet content relationships, temporal relationships between packets in a session, and statistical characteristics of a work session. A traffic data augmentation technique based on a GAN network is employed to mitigate the impact of data imbalance on traffic classes. The proposed platform's performance is evaluated on the public ISCX VPN-Non VPN 2016 dataset. The results demonstrate that the platform accurately and efficiently identifies applications and classifies encrypted traffic. Compared to state-of-the-art methods, the proposed traffic classification model improves precision by 21.3%, accuracy by 13.1%, recall by 18.11%, and F1 score by 19.79%.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: B Deep Packet Inspection (Dpi) Approachesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

CBS: A Deep Learning Approach for Encrypted Traffic Classification With Mixed Spatio-Temporal and Statistical Features

Seydali,

Khunjush,

Akbari

et al. 2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Within the framework of a "Personal Interactive Transmission Session," we are referring to a kind of communication that is customised for individual users, wherein information is sent in a dynamic and real-time manner [4]. During these sessions, examining the TCP stream entails analysing the attributes, security implications, and performance factors of the data being communicated [5]. The objective is to ensure a reliable and protected interactive encounter, taking into account aspects such as the integrity of data, the delay in transmission, and the general efficiency of managing personalised communication via a TCP connection [6].…”

Section: Introductionmentioning

confidence: 99%

Examining Potential Threats of Eavesdropping in TCP Stream of Personal Interactive Transmission Session

Abubakar,

Muhammed Najmuddin,

Mohd Alwi

et al. 2024

IJPCC

View full text Add to dashboard Cite

This study establishes that Personal Interactive Transmission Sessions conducted using TCP streams provide exceptional convenience and immediacy. However, they also provide noteworthy challenges and considerations, especially in terms of security, integrity, and performance. This study investigates the security weaknesses present in potential eavesdropping threats on TCP Streams during personal interactive transmission sessions. An experimental analysis was carried out, by doing a comprehensive analysis of network traffic using Wireshark. The investigation reveals a notable vulnerability in the management of user-provided data, together with the possible risks of eavesdropping, emphasizing the urgent need for enhanced security measures. The findings underscore the need of proactive cybersecurity strategies, such as frequent security audits, vulnerability assessments, and the implementation of robust encryption methods, to safeguard user privacy and ensure data integrity. To enhance risk mitigation, data protection, and user confidence in the digital environment, organizations should prioritize addressing potential eavesdropping threats during transmission sessions that involve sensitive personal data.

show abstract

“…The advent of the 5G era and rapid development of cellular vehicle-to-everything (C-V2X) technologies have realized a massive number of vehicles equipped with advanced intelligent devices (e.g., high-definition (HD) players, 3D navigation equipment, vehicle mixed reality (MR) glasses), which bring significant mobile data traffic and service explosion [1]. Furthermore, the global mobile data traffic is expected to reach 77 exabytes per month by 2022, as reported by Cisco, where 79 percent of this data comes from mobile video [2]. Additionally, innovative mobile video technologies have inspired a wide range of video streaming applications in vehicular networks, becoming one of the most popular and indispensable services.…”

Section: Introductionmentioning

confidence: 99%

Intelligent Caching for Mobile Video Streaming in Vehicular Networks with Deep Reinforcement Learning

Luo

Liwang

2022

Applied Sciences

View full text Add to dashboard Cite

Caching-enabled multi-access edge computing (MEC) has attracted wide attention to support future intelligent vehicular networks, especially for delivering high-definition videos in the internet of vehicles with limited backhaul capacity. However, factors such as the constrained storage capacity of MEC servers and the mobility of vehicles pose challenges to caching reliability, particularly for supporting multiple bitrate video streaming caching while achieving considerable quality of experience (QoE). Motivated by the above challenges, in this paper, we propose an intelligent caching strategy that takes into account vehicle mobility, time-varying content popularity, and backhaul capability to improve the QoE of vehicle users effectively. First, based on the mobile video mean opinion score (MV-MOS), we designed an average download percentage (ADP) weighted QoE evaluation model. Then, the video content caching problem is formulated as a Markov decision process (MDP) to maximize the ADP weighted MV-MOS. Owing to the prior knowledge of video content popularity and channel state information that may not be available at the road side unit in practical scenarios, we propose a deep reinforcement learning (DRL)-based caching strategy to solve the problem while achieving a maximum ADP weighted MV-MOS. To accelerate its convergence speed, we further integrate the prioritized experience replay, dueling, and double deep Q-network technologies, which improve the performance of DRL algorithm. Numerical results demonstrate that the proposed DRL-based caching strategy significantly improves QoE, and achieves better video delivery reliability compared to existing non-learning approaches.

show abstract

A survey on analyzing encrypted network traffic of mobile devices

Cited by 6 publications

References 79 publications

CBS: A Deep Learning Approach for Encrypted Traffic Classification With Mixed Spatio-Temporal and Statistical Features

CBS: A Deep Learning Approach for Encrypted Traffic Classification With Mixed Spatio-Temporal and Statistical Features

Examining Potential Threats of Eavesdropping in TCP Stream of Personal Interactive Transmission Session

Intelligent Caching for Mobile Video Streaming in Vehicular Networks with Deep Reinforcement Learning

Contact Info

Product

Resources

About