| In recent years,the interactive live streaming industry has developed rapidly,and more and more young people are using interactive live streaming as an indispensable part of life and entertainment.However,with the increasing number of users,in order to provide users with better service,many interactive live streaming platforms have to face the shortage of bandwidth resources during the peak period of interactive live streaming.The simplest solution is undoubtedly adding more bandwidth resources,but this increases the cost and causes a waste of bandwidth resources during off-peak hours.How to effectively use bandwidth resources under the condition of limited bandwidth resources will be an important problem for the interactive live streaming industry.In order to improve the utilization rate of interactive live streaming bandwidth resources,this paper designs the user Qoe model for the broadcast and the client,and proposes a bandwidth resource scheduling approach based on the model.By setting the bit rate selection strategy in the case of the least impact on the user Qoe with reducing the video bit rate of the broadcast and the video bit rate of the client to reduce the bandwidth consumption of the platform,and solve the shortage of bandwidth resources during the peak period of the interactive live streaming.The main work of this paper is as follows:(1)For the interaction between the broadcast and the client user in the process of interacting with the service or application,the subjective feelings of the service generated by the user are respectively designed with the Qoe model of the broadcast and the client.The Qoe model of the broadcast mainly considers the factors such as video bit rate,live effect,delay and so on.The client Qoe model is mainly aimed at the bit rate of the adaptive bit rate video stream,bit rate oscillation,delay,and carton.The assessment and the weights are based on the corresponding influencing factors,it is visually and effectively reflect the overall recognition of the services used by the users in a certain objective environment.(2)A bit rate selection strategy is formulated based on the designed Qoe model.In the case that the bandwidth of the interactive live streaming platform is shortage,the ratio of the delta of the user's Qoe per the delta of the occupied bandwidth resource is calculated according to the optional bit rate,and the user with the smallest ratio is selected to reduce the bandwidth resource that occupied by the user.As muchas possible,the most bandwidth resources are obtained without affecting the user's Qoe,ensuring that the total live Qoe of the interactive live streaming platform keeps growing while providing the best service to more users.In this paper,it is recommended that the threshold be set to 80% of the bandwidth resources required by all broadcasters or viewers,and under this condition,the total QoE of the broadcast in interactive live streaming platform is reduced by only 2%,while the total QoE of the client is reduced by about 11%. |
