Predicting Perceptual Quality in Internet Television Based on Unsupervised Learning

Frnda, Jaroslav; Nedoma, Jan; Martínek, Radek; Fridrich, Michael

doi:10.3390/sym12091535

Cited by 4 publications

(3 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A brand-new platform for testing of new technologies (CPIT TL3) is currently being finalized in Technical University of Ostrava. The whole building will be split into blocks, covering various fields of research, spanning from smart factory or home care to automotive [32,33]. Laboratory dedicated for development of VLC technologies and their integration into everyday life will be also part of CPIT TL3.…”

Section: Discussionmentioning

confidence: 99%

Enhancements of SDR-Based FPGA System for V2X-VLC Communications

Danys¹,

Martínek²,

Jaroš³

et al. 2021

Computers, Materials &Amp; Continua

Self Cite

View full text Add to dashboard Cite

This pilot study focuses on a real measurements and enhancements of a software defined radio-based system for vehicle-to everything visible light communication (SDR-V2X-VLC). The presented system is based on a novel adaptive optimization of the feed-forward software defined equalization (FFSDE) methods of the least mean squares (LMS), normalized LMS (NLMS) and QR decomposition-based recursive least squares (QR-RLS) algorithms. Individual parameters of adaptive equalizations are adjusted in real-time to reach the best possible results. Experiments were carried out on a conventional LED Octavia III taillight drafted directly from production line and universal software radio peripherals (USRP) from National Instruments. The transmitting/receiving elements used multistate quadrature amplitude modulation (M-QAM) implemented in LabVIEW programming environment. Experimental results were verified based on bit error ratio (BER), error vector magnitude (EVM) and modulation error ratio (MER). Experimental results of the pilot study unambiguously confirmed the effectiveness of the proposed solution (longer effective communication range, higher immunity to interference, deployment of higher state QAM modulation formats, higher transmission speeds etc.), as the adaptive equalization significantly improved BER, MER and EVM parameters. The best results were achieved using the QR-RLS algorithm. The results measured on deployed QR-RLS algorithm had significantly better Eb/N0 (improved by approx. 20 dB) and BER values (difference by up to two orders of magnitude).

show abstract

Section: Discussionmentioning

confidence: 99%

Enhancements of SDR-Based FPGA System for V2X-VLC Communications

Danys¹,

Martínek²,

Jaroš³

et al. 2021

Computers, Materials &Amp; Continua

Self Cite

View full text Add to dashboard Cite

show abstract

“…PCA algorithm reduces the set of features/variables/attributes in a way that analyses the potential correlation between them. As a result of this method, a set of variables called principal components is provided, where none of these variables are cross‐correlated (Jaroslav et al, 2020). PFA method computes the correlation matrix showing the correlation coefficients between two features/variables/attributes.…”

Section: Analyzing and Synthesizing Datamentioning

confidence: 99%

A feature selection for video quality of experience modeling: A systematic literature review

Skaka-Čekić

Husić

2023

WIREs Data Min & Knowl

View full text Add to dashboard Cite

Quality of Experience (QoE) multidimensional concept is the key for successful delivery of multimedia services. Higher user requirements for new experiences such as augmented reality, virtual reality, and future 6G services set higher requirements for QoE. A more complex QoE space requires the use of data mining methods in order to process the data for better QoE prediction. The increased dimensionality of the QoE space becomes a limiting factor for achieving the desired QoE prediction accuracy. Existing studies considering the QoE multidimensional concept with approaches that overcome the challenge of increased QoE space dimensionality are of great importance for future research. Accordingly, this article aims to review the applications of Feature Selection (FS) methods in video QoE modeling. It provides a comprehensive overview of the existing studies with the categorization and review of applied FS methods with reference to the data collection and data modeling steps. The analysis included 71 studies which provides overview of the FS methods applications in video QoE modeling depending on the input Influence Factor (IF) dimension sizes, type of IFs, QoE prediction methods used and QoE evaluation type. Our review revealed the advantages of using FS methods in video QoE modeling, frequency of application of FS methods with potential of applying more FS methods in a series or a parallel, gives an overview of the achieved dimensionality reduction degree for different methods, and provides insights in opportunities for researchers for applying FS methods on complex multidimensional QoE space. This article is categorized under: Technologies > Data Preprocessing Algorithmic Development > Multimedia

show abstract

“…An unsupervised learning approach has been studied to assess the QoE in IPTV services along with self-organizing maps. That approach included a full-reference model, visual distortion, and packet loss considerations and was suitable for assessing the video quality in broadcasting TV systems [11]. Another study that investigated unsupervised learning for QoE measurement presented the real-time algorithm deployed on the server side and the offline no-reference assessment on the customer side.…”

Section: Introductionmentioning

confidence: 99%

Unsupervised Learning Data-Driven Continuous QoE Assessment in Adaptive Streaming-Based Television System

Andruloniw

Kowalik²,

Zwierzykowski

2022

Applied Sciences

View full text Add to dashboard Cite

The quality of experience (QoE) assessment of adaptive video streaming may be crucial for detecting degradations impacting customer satisfaction. In a telecommunication environment, eliminating failure points may be the highest priority. This study aims to assess the QoE level of the video played by the STB device connected to the production TV system. The evaluation has been based on the stalling effects, video quality changes, and the time related to the last decreased bitrate change occurrence. The two-phase continuous clustering approach has been studied to assess the QoE level based on the ACR scale. The number of devices with grades 1 or 2 is relatively low, but those devices generate significantly more events than adequately functioning devices. STBs try to play the highest possible bitrate, and there is no possibility of setting the intermediate bitrate level. The STB player does not have the button to set the quality level, usually available in pure over-the-top applications. Hence the bitrate fluctuations that can annoy customers appear for the lowest grades. The boundary cases can be easily assessed. The outcome should be challenged by the customers’ opinions to find the proper QoE threshold. Continuous clustering may allow telecom operators to assess customer satisfaction with their TV service.

show abstract

Predicting Perceptual Quality in Internet Television Based on Unsupervised Learning

Cited by 4 publications

References 28 publications

Enhancements of SDR-Based FPGA System for V2X-VLC Communications

Enhancements of SDR-Based FPGA System for V2X-VLC Communications

A feature selection for video quality of experience modeling: A systematic literature review

Unsupervised Learning Data-Driven Continuous QoE Assessment in Adaptive Streaming-Based Television System

Contact Info

Product

Resources

About