In the white paper, scenarios where edge computing can be applied in the future are given. Many application scenarios can already be realized. Only video analysis, content optimization and positioning technology scenarios are analyzed here.
1) Positioning technology
Figure 3 shows an example of location tracking. The location of a person or object is obtained through the global position system (GPS) or third-party positioning technology running on the MEC platform. Then, if necessary, it is returned to the core network Side. This kind of local positioning function is very effective for retailers, venues, stadiums, campuses or specific areas. First of all, the position feedback is very fast, and secondly, the accuracy is also guaranteed.
2) Video analysis
Figure 4 shows an example of video analysis. Taking surveillance as an example, cameras are currently widely used and can basically achieve seamless docking and blind-angle monitoring in parking lots, traffic thoroughfares, residential areas, and campuses. With the deployment of The increase in the number of cameras and the
With the improvement of video quality, the amount of surveillance video data is also gradually increasing. If such a large amount of video data is transmitted back through the core network to the centralized cloud platform for video analysis and processing, the round-trip delay will be very large. And if intelligence is deployed inside the camera Analysis tools, it will be difficult to deploy these functions due to the design size of the camera itself. Therefore, a better solution is to deploy the video analysis APP on the local MEC platform. Camera LoRa gateway in a certain area
3) Content optimization and caching
Figure 5 gives an example of content optimization based on RAN side sensing. Content optimization refers to dynamically optimizing content based on the information provided by the network, such as cell ID, cell load, link quality, data throughput rate, etc., so as to Improve QoE and network efficiency. As for video caching, when the terminal requests video playback, the resource may exist in uploading the recorded surveillance video to the MEC platform. After video analysis and processing, the results obtained can be retrieved and returned at any time. Passed to the core network. China Unicom’s edge cloud-based intelligent security commercial deployment solution demonstrated at the Shenzhen Hi-Tech Fair in November 2017 deployed artificial intelligence (AI) video analysis functions on the edge cloud, which greatly improved Time for result processing and delivery.
The local MEC platform downloads video resources from the local when playing again, saving bandwidth and core network processing time. This video caching function is suitable for the playback and viewing of popular TV series, popular movies and recent variety shows. It is of great help. At the same time, this model is also more suitable for areas such as university towns, residential areas or hot business districts with dense flow of people and large requests for video playback.