Empirical Conditional Mean: A New Method of Predicting Throughput in Uplink Data Network

Experience of live video streaming can be improved if future bandwidth can be predicted more accurately on the video uploader side. Thus follows a natural question, how to make predictions both easily and precisely in real networks. Researchers have developed many prediction algorithms in the literature, from where a simple algorithm comes out, Arithmetic Mean (AM). Based on that, we are purposing a new method called Empirical Conditional Mean (ECM), hoping to utilize more information in the past data to get a more accurate prediction without loss of practicality. Our experiments found that when the frequency of measuring the network throughput is low, the ECM algorithm performs better than the commonly used AM one, which means ECM can send more data out while not having a higher loss rate. Hence ECM can be helpful for those who especially have relatively limited networks to reach a more considerable balance between frame loss rate and video quality.