A high-quality call doesn't require much bandwidth; 150 kilobytes of continuous upload/download is sufficient. However, this bandwidth must be constantly available for near-realtime communication. The bandwidth required for voice data transmission depends, among other things, on the codec used and its bit rate, as well as the payload per packet. ATTENTION: It is important that there is always sufficient bandwidth available in the network for telephony.

Network packet delay (NPD) describes the time it takes for a data packet to travel from its origin through the network to its destination. Each station along the path, such as switches, routers, firewalls, or jitter buffers, as well as the length of the path itself, increases this value.

In most cases, analog signals are not digitized losslessly during the encoding process. Instead, the analog signal is subjected to dynamic range reduction and data compression. Depending on the extent and method, this can lead to quality losses when converting the digital data stream back to analog signals. Not only can the sound quality be affected, but playback continuity can also be compromised. This results in a reduction in the bandwidth required to transmit the digital signal. Please note the codecs supported by peoplefone and, in some cases, the order in which they are used. We always recommend starting with the lowest " quality ."

In network engineering, jitter refers to the variance in the delay of data packets from the sender to the receiver from the application perspective. Jitter of more than 20 milliseconds can lead to quality problems in voice transmission, assuming a packet payload of 20 milliseconds of voice data.

Packet loss occurs when a sender sends packets that do not reach the receiver due to network issues. In practice, it is difficult to identify packet loss as the cause of a problem. Each codec handles packet loss differently. For example, with the same loss rate, the perceived voice quality of a codec with data compression might be better than that of a codec that uses the entire bandwidth.

An incorrect order of incoming data packets has a similar effect to packet loss in voice and video transmissions. If a packet arrives out of sequence, it is discarded by default by the end device. Obviously, it makes no sense to play voice data out of order.

Transcoding refers to the conversion of voice signals during the transition from a TDM ("time-division multiplexing") network to an IP network and vice versa. If the voice data has to be converted multiple times during a call – for example, because both participants are using IP to make internal calls, but the call is routed via the landline – the voice quality degrades with each transition. Ultimately, this can only be solved through the most intelligent call routing possible.

There are two causes of the infamous echo in VoIP calls: Acoustic echo occurs when feedback occurs between the speaker/earpiece and the microphone on a telephone. Line echo, on the other hand, occurs when impedance differs at the transition between a two- and four-wire network, during signal conversion between a TDM bus (time-division multiplexing) and the LAN, or when impedance differs between a headset and its adapter.





SIP stands for Session Initiation Protocol and is a network protocol commonly used to establish telephone calls via VoIP (Voice over IP). It is one of many signaling protocols that enable the establishment and termination of a call (session) between two or more participants, but is particularly widely used in IP telephony.