Packet Loss Based Real Time Network Condition Estimation
Xiaoyu Cheng, BI Du-yan
Abstract
Xiaoyu Cheng, BI Du-yan
Abstract
Packet loss behaviors have been studied for many years. Traditional packet loss evaluation criteria are not sufficient to describe characteristics of losses: the loss rate is too simple while traditional loss episodes sometimes are not stable due to the randomness of loss events. In this paper, we propose an extended loss episode approach, merged loss episode based on the traditional definition, to estimate the network condition. Also we presented the corresponding algorithm for calculation. Simulation results show that compared with traditional loss episodes, the merged version could be more stable and effective to measure network performance on loss issues.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Packet loss behaviors have been studied for many years. Traditional packet loss evaluation criteria are not sufficient to describe characteristics of losses: the loss rate is too simple while traditional loss episodes sometimes are not stable due to the randomness of loss events. In this paper, we propose an extended loss episode approach, merged loss episode based on the traditional definition, to estimate the network condition. Also we presented the corresponding algorithm for calculation. Simulation results show that compared with traditional loss episodes, the merged version could be more stable and effective to measure network performance on loss issues.
Key concepts: Packet loss, Data loss, Computer science, Randomness, Simple (philosophy), Transmission loss, Power loss, Network packet