Benefits of IMS Data Channels for Voice Enrichment in 4G/5G: A Simulation-Based Approach

Benefits of IMS Data Channels for Voice Enrichment in 4G/5G: A Simulation-Based Approach. As mobile communication becomes more advanced and focused on user experience, there is growing demand for services that enrich voice and video calls in real time such as transcription, translation, emotion detection and file sharing. Traditional networks, though reliable, cannot guarantee the performance required for these new features. This thesis focuses on a promising technology for future mobile services: the IMS Data Channel. It allows real-time exchange of data (like text, files or emotional feedback) during a voice or video call. The IMS Data Channel is an extension of the IP Multimedia Subsystem (IMS), which telecom operators already use to manage services over 4G, 5G and Wi-Fi. With this architecture, advanced services can run directly inside the call without using external apps, offering better control, security and quality. This thesis is structured into three main parts. The first part provides a theoretical overview of how IMS Data Channel works, based on international standards. It introduces the main components involved and explains how secure, real-time data sessions are established and controlled. The focus is on how telecom operators can offer smart services during calls, while maintaining reliability and quality of service, thanks to a standardized, operator-managed infrastructure. The second part implements a real-time simulation of use cases using WebRTC and Artificial Intelligent services. The test environment is based on the Janus WebRTC Gateway and integrates Amazon services such as Transcribe, Translate and Comprehend. A Python Flask server acts as middleware, connecting the WebRTC clients to Amazon Web Services via software Application Programming Interfaces for real-time data exchange. In the simulated scenario, user A speaks or sends text messages in Italian. User B receives real-time English transcription, translated audio and text, as well as visual feedback based on sentiment analysis. The third part investigates the system's behavior under unstable network conditions, which are common in best-effort internet environments. Controlled degradations such as jitter, artificial congestion and packet loss are introduced to evaluate the impact on real-time performance. Metrics such as Round-Trip Time (RTT), time to first transcription and packet loss are collected and analyzed under various conditions. Results show that best-effort networks fail to guarantee stable performance. Unlike WebRTC, which relies on public internet conditions and cannot ensure consistent performance, the IMS Data Channel operates within the operator’s managed network and follows international standards designed to guarantee quality of service. By supporting features like resource reservation and traffic prioritization, it maintains low latency and service stability even under stress. These capabilities make the IMS Data Channel a promising candidate to support next generation of Artificial Intelligent enhanced communication services.