Intelligent aerial video streaming: Achievements and challenges

Deploying aerial infrastructure to support traditional terrestrial wireless communications has enhanced the maturity of 5G networks and beyond. This improvement has certainly benefited video streaming services whose data makes up the bulk of mobile traffic and continues to soar. Realising this fashion, artificial intelligence is considered as one of the important tools, which can effectively deal with various problems to improve and optimise many aspects of video streaming systems. However, there is still a lack of a systematic investigation of recent works in this field to facilitate researchers with a comprehensive reference framework to advance the development of the technology. Realising this shortcoming, this paper has dedicatedly investigated the recent achievements in using AI and the future challenges of the system. First, the basic knowledge and overall architecture of the aerial video streaming (AVS) system are given. Subsequently, achievable utilities are thoroughly evaluated by highlighting each system performance mathematically and analysing recent technical works to address them. Next, typical application scenarios are acquired to demonstrate the advantages to human life and the necessity of their implementation. Finally, open challenges are discussed to drive future technological developments towards video streaming over aerial infrastructure.