Systems and methods for dynamic bandwidth allocation
Disclosed herein are various embodiments of systems and methods that may be utilized in a variety of videoconferencing applications. According to various embodiments, techniques may be utilized to dynamically allocate and adjust bandwidth utilization during a videoconferencing session. A data network may allow for the transmission of data between two or more endpoints. The data exchanged between the endpoints may include video data, audio data, control data, and status data. Control data may be utilized in various embodiments to operate a robotic videoconferencing endpoint. Accordingly, various components of a data network connecting videoconferencing endpoints may transmit data wirelessly.
1. A method of monitoring network statistics in a videoconferencing session, comprising:
initiating a videoconferencing session between a first endpoint and a second endpoint via a network;
transmitting data between the first endpoint and the second endpoint;
collecting a first plurality of data samples, each of the plurality of data samples comprising a quantifiable metric representing one or more network conditions;
discarding data samples from the first plurality of data samples that exceed a first percentile of the data samples in a network statistics data set;
adding data samples from the first plurality of data samples to network statistics dataset;
analyzing the network statistics dataset to estimate an available bandwidth of the network;
identifying a network disturbance;
discarding data samples collected for a specified period of time following the network disturbance; and
adjusting the bandwidth used for the videoconferencing session based on the estimated available bandwidth.
2. The method of claim 1 , wherein the plurality of data samples comprises a plurality of latency measurements.
3. The method of claim 1 , wherein the first percentile is approximately equal to the 90th percentile of data samples in the network statistics dataset.
4. The method of claim 1 , further comprising:
a second percentile, the second percentile being lower than the first percentile;
identifying a data sample below the first percentile and above the second percentile;
collecting a second plurality of data samples;
determining that a data sample from among the second plurality of data samples exceeds a third percentile; and
discarding the second plurality of data samples.
5. The method of claim 1 , further comprising:
a second percentile, the second percentile being lower than the first percentile;
identifying a data sample below the first percentile and above the second percentile;
collecting a second plurality of data samples;
determining that a data sample from among the second plurality of data samples is below a third percentile; and
adding the second plurality of data samples to the network statistics dataset.