Prodys is releasing new firmware upgrades for its family of IP codecs. Version 6 is incorporating many important improvements aimed at ensuring consistent audio quality across IP links despite variable network performance.

IP networks are designed to support non-real time connection oriented applications, not usually sensitive to delay variation (jitter) and lost packets. However audio over IP applications require timely packet delivery with low delay, jitter and packet loss values.

The quality of service of Internet networks varies under congestion and convergence in terms of jitter, lost and disordered packets and available bandwidth. Due to the requirements of audio over IP applications, some sort of QoS must be provided by the network to deal wih these impairments or, in case this is not possible, or this QoS mechanisms do not perform as expected, a set of tools to help the customer diagnose and achieve the best consistent quality in their audio over IP communications must be provided by the audio codec to compensate for this.

In this regard, Prodys is releasing a set of tools with its next firmware version. Among this tools are FEC (Forward Error Correction), Error Concealment and Active Recovery techniques.

FEC (Forward Error Correction) is a passive recovery mechanism, which adds redundancy to the audio stream so that the receiver is able to reconstruct the original stream in case of a packet loss. The number of lost packets which can be recovered will depend on the FEC scheme. Thus, Prodys offers two different schemes: First scheme is called 50% (intruduces 50% overhead) and it allows for recovery of single packet losses, and 100% (100% overhead) which allows for recovering of some some bursts of two consecutive packet losses. In order to offer backwards compatibility the sending device sends the original media stream, as if there were no FEC plus and additional FEC stream. This allows the media stream to still be used by receivers who are not FEC capable.

Active Recovery is based on Retransmission techniques. Unlike FEC (passive recovery), no redundant information is added to the original stream to help the receiver reconstruct the data flow in case of packet losses. On the contrary, Active Recovery does only send additional information whenever it is actually needed, thus keeping additional network overhead to a minimum. The disadvantage of this technique is longer end-to-end latency.

The final Audio Integrity Protection Technique introduced by Prodys in this firmware release is Error Concealment.

When compressed audio is transmitted through a packet switched data network, data can be lost due to channel errors and/or congestion. Techniques that post-process the received audio and conceal the errors in real-time are needed. These techniques are aimed at reducing the impact of packet losses on the audio quality in terms of glitches and dropouts.

Prodys has an advantage over other manufactures in that ALL hardware and ALL software development is done "in-house" (this includes the audio algorithms). This means that we have an implicit understanding of the how to implement transparent Error Concealment in the audio domain.

Once again, Prodys places itself ahead of the development on the audio over IP in the worldwide market, offering their customers the best and most intuitive tools to help them achieve the most consistent and highest quality audio over IP.