Changing network conditions pose severe problems to video streaming in the Internet. HTTP adaptive (HAS) is a technology employed by numerous services that relieves these issues adapting current conditions. It enables service providers improve resource utilization and Quality of Experience (QoE) incorporating information from different layers order deliver adapt its best possible quality. Thereby, it allows taking into account end user device capabilities, available quality levels,...

This paper addresses the challenge of assessing and modeling Quality Experience (QoE) for online video services that are based on TCP-streaming. We present a dedicated QoE model You Tube takes into account key influence factors (such as stalling events caused by network bottlenecks) shape quality perception this service. As second contribution, we propose generic subjective assessment methodology multimedia applications (like video) is crowd sourcing - highly cost-efficient, fast flexible...

HTTP Adaptive Streaming (HAS) is employed by more and video streaming services in the Internet. It allows to adapt downloaded quality current network conditions, thus, avoids stalling (i.e., playback interruptions) greatest possible extend. The adaptation of streams done switching between different representation levels, which influences user perceived stream. In this work, influence several parameters, namely, switch amplitude level difference), frequency, recency effects, on Quality...

The performance of YouTube in mobile networks is crucial to network operators, who try find a trade-off between cost-efficient handling the huge traffic amounts and high perceived end-user Quality Experience (QoE). This paper introduces YoMoApp (YouTube Performance Monitoring Application), an Android application, which passively monitors key indicators (KPIs) adaptive video streaming on smartphones. monitored KPIs (i.e., player state/events, buffer, quality level) can be used analyze QoE...

YouTube is changing the way operators manage network performance monitoring. In this paper we introduce YOUQMON, a novel on-line monitoring system for assessing Quality of Experience (QoE) undergone by HSPA/3G customers watching videos, using network-layer measurements only. YOUQMON combines passive traffic analysis techniques to detect stalling events in video streams, with QoE model map stallings into Mean Opinion Score reflecting end-user experience. We evaluate detection hundreds and...

Monitoring the Quality of Experience (QoE) undergone by cellular network customers has become paramount for ISPs, who need to ensure high quality levels limit customer churn due dissatisfaction. This paper tackles problem QoE monitoring, assessment and prediction in networks, relying on end-user device (i.e., smart-phone) QoS passive traffic measurements crowdsourced feedback. We conceive different models based supervised machine learning techniques, which are capable predict experienced end...

A quarter of the world population will be using smartphones to access Internet in near future. In this context, understanding quality experience (QoE) popular apps such devices becomes paramount cellular network operators, who need offer high-quality levels reduce risks customers churning for dissatisfaction. paper, we address problem QoE provisioning from a double perspective, combining results obtained subjective laboratory tests with end-device passive measurements and crowd-sourced...

WhatsApp is a very popular mobile messaging application, which dominates todays communication. Especially the feature of group chats contributes to its success and changes way people communicate. The group-based communication paradigm investigated in this work, particularly focusing on usage WhatsApp, chats, implications network traffic.

Current objective video quality metrics typically estimate for short sequences (10 to 15 sec) of constant quality. However, customers services usually watch longer videos which are more and delivered via adaptive streaming methods such as HTTP (HAS). A viewing session in a setting contains several different qualities over time. In order express this an overall score the whole session, temporal pooling have been proposed related work. Within paper, we set out compare performance prediction...

Since its introduction a few years ago, the concept of `Crowdsourcing' has been heralded as highly attractive alternative approach towards evaluating Quality Experience (QoE) networked multimedia services. The main reason is that, in comparison to traditional laboratory-based subjective quality testing, crowd-based QoE assessment over Internet promises be not only much more cost-effective (no lab facilities required, less cost per subject) but also faster terms shorter campaign setup and...

Today's packet-switched networks are subject to bandwidth fluctuations that cause degradation of the user experience multimedia services. In order cope with this problem, HTTP adaptive streaming (HAS) has been proposed in recent years as a video delivery solution for future Internet and being adopted by an increasing number services, such Netflix Youtube. HAS enables service providers improve users' quality (QoE) network resource utilization adapting stream current conditions. However,...

As stalling is the worst Quality of Experience (QoE) degradation HTTP adaptive video streaming (HAS), this work presents a stream-based machine learning approach, ViCrypt, which analyzes YouTube sessions in realtime from encrypted network traffic. The session subdivided into stream short time slots 1s length, while considering two additional macro windows each for current trend and whole ongoing session. Constant memory features are extracted traffic these three fashion, fed random forest...

Video streaming is the killer application of Internet today. In this article, we address problem real-time, passive Quality-of-Experience (QoE) monitoring HTTP Adaptive Streaming (HAS), from Internet-Service-Provider (ISP) perspective - i.e., relying exclusively on in-network traffic measurements. Given wide adoption end-to-end encryption, resort to machine-learning (ML) models estimate multiple key video-QoE indicators (KQIs) analysis encrypted traffic. We present ViCrypt, an ML-driven...

Online video games and cloud gaming are rapidly growing in pervasiveness. Their resource demands can put significant stress on the global communication infrastructure. And network conditions amongst chief factors that influence one's enjoyment while playing games. This makes it imperative for to be considered dimensioning, server placement or protocol development. For reason, this work we provide an introduction technical aspects of general their particular. understanding forms basis a rich...

Machine learning has found many applications in network contexts.These include solving optimisation problems and managing operations.Conversely, networks are essential for facilitating machine training inference, whether performed centrally or a distributed fashion.To conduct rigorous research this area, researchers must have comprehensive understanding of fundamental techniques, specific frameworks, access to relevant datasets.Additionally, data can serve as benchmark springboard further...

Changing network conditions like bandwidth fluctuations and resulting bad user experience issues (e.g. video freezes) pose severe challenges to Internet streaming. To address this problem, an increasing number of services utilizes HTTP adaptive streaming (HAS). HAS enables service providers improve Quality Experience (QoE) resource utilization by incorporating information from different layers. However, these adaptation possibilities also introduce new perceivable impairments such as the...

A quarter of the world population will be using smartphones to access Internet in near future. In this context, understanding Quality Experience (QoE) popular services such devices becomes paramount for cellular network operators, who need offer high quality levels reduce risks customers churning dissatisfaction. paper we study problem QoE provisioning smartphones, presenting results obtained from subjective lab tests performed five apps: YouTube, Facebook, Web browsing through Chrome,...

HTTP Adaptive Streaming (HAS) adapts the video quality to current network condition by switching between different layers. As HAS was shown perform better than classical streaming, it is becoming increasingly popular. Recent research showed that switch amplitude and time on layer have an impact Quality of Experience (QoE) HAS. However, those studies focused only adaptation two layers so far. This work extends these findings taking three into account. Thereby, especially intermediate user...

The concept of Quality Experience (QoE) Internet services is widely recognized by service providers and network operators. They strive to deliver the best experience their customers in order increase revenues avoid churn. Therefore, QoE increasingly considered as an integral part reactive traffic management cycle In addition, also constitutes a its own, which includes user behavior requirements. This letter describes this cycle, not taken into account yet, discusses interactions two cycles,...

The introduction of the QUIC (Quick UDP Internet Connections) transport protocol by Google aimed to improve Quality Experience (QoE) with web services compared prevailing Transport Control Protocol (TCP). Nowadays, has become default communicate between Chrome browser and servers accounts for an increasing share traffic. This work investigates whether promised QoE benefits are indeed noticeable end users or not. A measurement study was conducted YouTube video streaming in two mobile fixed...

Due to biased assumptions on the underlying ordinal rating scale in subjective Quality of Experience (QoE) studies, Mean Opinion Score (MOS)-based evaluations provide results, which are hard interpret and can be little meaningful. This paper proposes consider full QoE distribution for evaluating reporting results instead only using MOS values. The represented a concise way by parameters multinomial without losing any information about ratings, even keeps backward compatibility with previous,...