| 摘要 |
A method and apparatus for requesting bandwidth in a subscriber station is disclosed. The method dynamically changes the size of the bandwidth request based on the prediction of the number of packets needed to be transmitted. The average delay experienced by packets in a queue is measured, and this information is then used to change the coefficient of a prediction equation. When the experienced average delay is below the agreed upon QoS latency parameter or delay target, the method reduces the size of the bandwidth requests by making the prediction equation more conservative. On the other hand, when the experienced delay is above the agreed upon latency, the algorithm will make the prediction equation more aggressive, increasing the bandwidth requests and reducing the latency for future packets. By modifying the prediction equation based on the measured delay, the method is able to select the optimal point for achieving time delay requirements while preserving air-link resources. |
