| 摘要 |
<p>Adaptive redundancy is implemented (either switched on or off) for a voice over internet protocol (VoIP) packet connection by a basic two step approach. A first step of adaptive redundancy implementation involves monitoring relevant sources or indicators to determine when the need for redundancy (e.g., redundancy coding) may arise. The monitored sources can be one or more of a transmission buffer(s); a receiving buffer(s): sender and/or receiver statistics. An example of such statistics are the statistics which can be obtained from RTCP protocol, e.g., number of packets/data bytes sent, number of packets/data bytes lost, jitter, etc. A second step of adaptive redundancy implementation involves the triggering of redundancy implementation events when the monitored sources reach or obtain certain threshold value(s) (threshold). Examples of such thresholds (one or more of which may be utilized) include: a number of packets (or bytes of data) in an transmission buffer; a number of lost packets (or bytes of data) calculated from a reception buffer; and, requisite sender and receiver statistics (such as number of packets/data bytes sent and/or being lost has reached a maximum/minimum amount). Examples of the second step implementation events thusly triggered include one or more of the following: ( 1 ) dropping or removing a number of packets (or bytes of data) in the transmission buffer, and (2) triggering a signaling message to the sending client and/or receiving client to switch on/off redundancy coding. The first and second steps of redundancy implementation can be repeated as needed. Scenarios of adaptive redundancy implementation include both network-initiated implementations (e.g.. implementations initiated by a radio access network (RAN)) and mobile-initiated implementations (e.g.. implementations initiated by a wireless station).</p> |
