Emulation of point cloud streaming over 5G network

Emulation of point cloud streaming over 5G network

연구 분야: Networking

학회: International Journal of Information Technology

Realistic digital representations of three dimensions (3D) objects and environments are now achievable because to recent developments in computer graphics, enabling real-time user interactions. Creating effective compression techniques and technologies that may take into account varied application limits has become a crucial problem due to the rising need for various point clouds. Future wireless networks are expected to undergo a paradigm change as a result of the 3rd Generation Partnership Project’s (3GPP) 5th Generation advanced development. In this paper, we propose a complete system for streaming 3D high density point cloud data using a web-based streaming server with Hyper-Text Transfer Protocol version 2 (HTTP/2) protocol enabled, and compare results in two scenarios over Wireless Fidelity (WiFi) and over 5G standalone network. Results have shown great outperformance over conventional work by decreasing inter-frame latency for large point cloud streaming by 16.8% for the case of 4 million points using http/2 over wifi, and by 71.51% over emulated 5G network. Streamed packets were also captured showing an increased frame rate of the same sample by 20.7% and 353% for the cases of WiFi and 5G networks, respectively.