Visual quality assessment after network transmission incorporating NS2 and Evalvid

Area of Science:

• Computer Science
• Multimedia Systems
• Network Engineering

Background:

• Multimedia transmission over networks faces challenges in maintaining quality due to various influencing factors.
• Existing simulation tools may lack comprehensive evaluation capabilities for multimedia data.
• Understanding the interplay of source and network parameters is crucial for optimizing video streaming.

Purpose of the Study:

• To develop a novel simulation and evaluation platform, myEvalvid, for multimedia transmission.
• To investigate the impact of key video stream parameters on transmission performance.
• To establish relationships between these parameters and resulting quality distortions.

Main Methods:

• Utilized the Evalvid tool integrated within NS2 (Network Simulator version 2) to create the myEvalvid platform.
• Conducted simulations analyzing the effects of Group of Pictures (GOP) pattern, compression parameters, packet length, and packet error rate.
• Evaluated video quality by comparing reconstructed images at the receiver with original images.

Main Results:

• Different video stream parameters were found to cause varying levels of distortion.
• The study identified specific relationships between source/network parameters and video quality degradation.
• Simulation results demonstrated the effectiveness of the developed evaluation models.

Conclusions:

• The myEvalvid platform provides a robust environment for simulating and evaluating multimedia transmission.
• The designed video transmission and quality evaluation model accurately assesses performance in complex network conditions.
• Optimizing parameters such as GOP pattern and packet error rate is essential for enhancing video streaming quality.