Streaming media service rate prediction method and device
A technology of service rate and prediction method, which is applied in the field of streaming media and can solve problems such as inaccurate prediction results
Patent Information
- Authority / Receiving Office
- CN · China
- Current Assignee / Owner
- Publication Date
- 2017-02-22
Smart Images

Figure 1 
Figure 2 
Figure 3
Abstract
Description
technical field
[0001] The invention relates to the technical field of streaming media, in particular to a streaming media service rate prediction method and device. Background technique
[0002] With the rapid development of mobile communication technology and mobile terminal technology, users urgently need to be able to receive Internet information and services anytime, anywhere while moving. Moreover, with the continuous improvement of mobile network performance and terminal processing capabilities, users have more requirements for multimedia services such as high-quality video and audio. On this basis, the streaming media transmission service based on the HLS (HTTP Living Streaming) protocol came into being. The streaming media transmission service based on the HLS protocol can better adapt to the existing network, provide users with a near real-time playback experience, and is widely used in scenarios such as video on demand and live broadcast.
[0003] Specifically, ...
Examples
Embodiment 1
[0024] Embodiment 1 of the present invention provides a streaming media service rate prediction method. The streaming media service rate prediction method is applicable to a streaming media transmission service system based on the HLS protocol, which is not limited in this embodiment of the present invention. Specifically, such as figure 1 As shown, it is a schematic flow chart of the streaming media service rate prediction method in Embodiment 1 of the present invention, and the streaming media service rate prediction method may include the following steps:
[0025] Step 101: Obtain service rate impact data of the previous data sampling time period at the current moment.
[0026] Optionally, the service rate impact data may include: service data transmission success rate, service data retransmission rate, service data out-of-sequence rate, service drop times, service average rate, and the like.
[0027] Among them, the specific meaning of each service rate impact data can be...
Embodiment 2
[0117] Based on the same inventive concept, Embodiment 2 of the present invention provides a streaming media service rate prediction device. The streaming media service rate prediction device may be a network side device such as a video server, which is not limited in this embodiment of the present invention; in addition, It should be noted that, for the specific implementation of the streaming media service rate prediction device, refer to the relevant description in the first method embodiment above, and the repetition will not be repeated, as shown in Figure 4 As shown, the streaming media service rate prediction device may include:
[0118] The training unit 41 can be used to train multiple sets of sample data in the previous set period of time according to the set LS-SVM rate prediction algorithm to obtain the LS-SVM rate prediction model, wherein each set of samples The data includes the actual service rate of the first data sampling time period, and the service rate im...