Error robustness scheme for perceptually coded audio based on interframe shuffling of samples

High-quality audio streaming over IP networks is a significant part of multimedia traffic in telecommunications networks in the future. Although there are a number of error concealment and correction methods developed for real-time multimedia streaming over unreliable packet-switched networks, many effective codec dependent error robustness schemes have not been utilized for the state-of-art audio compression standards developed mainly to compress audio for storage media. This paper describes how the idea of redistributing adjacent audio samples to different packets can be applied to the perceptual audio codecs, such as MP3 and AAC. The experiences with testing the concept using AAC are explained as well. The proposed approach is especially applicable with semi-reliable transport protocols and future networks providing flexible support for prioritized data traffic.