spacenetlab / spacertc_project Goto Github PK

User-perceived latency is important for the quality of experience (QoE) of wide-area real-time communications (RTC). This paper explores a futuristic yet important problem facing the RTC community: can we exploit emerging mega-constellations to facilitate low-latency RTC globally? We carry out our quest in three steps. First, through a measurement study associated with a large number of geo-distributed RTC users, we quantitatively expose that the meandering routes over the client-cloud segment and the inter-cloud-site segment in existing cloud-based RTC architecture are critical culprits for the high latency issue suffered by wide-area RTC sessions. Second, after analyzing the low-latency potential enabled by mega-constellations, we formulate the dynamic RTC latency minimization problem under the integrated space-ground environment, and propose SpaceRTC, a satellite-cloud cooperative framework that adaptively selects relay servers upon satellites and cloud sites to build an overlay network which enables diverse close-to-optimal paths, and then judiciously allocates RTC flows in the network to facilitate low-latency interactions. Finally, we implement a hardware-in-the-loop experimental environment based on public constellation information and RTC trace, and extensive experiments demonstrate that SpaceRTC can deliver near-optimal interactive latency, with up to 64.9% latency reduction as compared with other state-of-the-art cloud-based solutions under representative videoconferencing traffic.

License: MIT License