2.1.5.7. Parameters for Jitter Buffer Configuration 
To understand the delay parameters, it is important that the values
are aligned with the compensable network jitter. As is the case for all
IP gateways or terminals, the actual transmission times for voice packets
in HG 3500 and HG 3575 also fluctuate around a constant ideal value.
This deviation (jitter at transmitting end) is already taken into account
during implementation.
These parameters can be set via the AMO STMIB (for HG3575)
or the AMO CGWB (for HG 3500).
- Jitter Buffer Type: Select whether the jitter buffer should operate in static or adaptive mode. In adaptive mode, the jitter buffer adjusts the average delay according to the situation when the data is received. In this way, the jitter buffer tries to reduce as far as possible both the delay and the number of lost packets. In static mode, the average delay always remains the same.
- Average Delay for Voice (msec): With this parameter you can specify the average number of milliseconds for which an IP packet should be held in the jitter buffer in IP-based voice transmission. In the “adaptive“ jitter buffer, the value specified here only represents an initial value. 40 is the value recommended for most environments.
- Maximum Delay for Voice (msec): In the “static“ jitter buffer, this parameter is used to define how many milliseconds are allowed (before the jitter buffer begins to regulate the data stream) for an actual measured delay when IP packets arrive during voice transmission. In the “adaptive“ jitter buffer, the maximum number of milliseconds allowed for the average voice delay is entered in this field. If the actual measured delay is longer, packets are lost. 80 is the recommended value for most environments with the static jitter buffer; 120 is the recommended value for the adaptive jitter buffer. Either way, the value must be higher than that in the “Average Delay for Voice (ms)“ field.
- Minimum Delay for Voice (msec): If the “adaptive“ jitter buffer was selected, use this parameter to enter how many milliseconds are allowed for the minimum average voice delay. This means that in every case, the average delay value is higher than or equal to this value.
- Packet Loss / Delay Preference: In the case of adaptive jitter buffers, you can use values from 0 to 8 in this parameter to specify whether you would prefer packet loss or a longer delay in the case of large packet delays. 0 denotes minimum packet loss and acceptance of delays in the voice data stream, 8 denotes minimum delay in the voice data stream and acceptance of packet loss. 4 is the value recommended for most environments.
- Average Delay for Data (msec): With this parameter, you can specify the average number of milliseconds for which an IP packet should be held in the jitter buffer during data transfer. 60 is the value recommended for most environments.
- Maximum Delay for Data (msec): With this parameter you can specify the number of milliseconds allowed (before the jitter buffer begins to regulate transmission) for an actual measured delay when IP packets arrive during data transfer. 200 is the recommended value for most environments. A parameter setting does not make any difference for higher values (approx. 200 upwards) because packets then leave the buffer as soon as they are received in full. Values under 100 ms are possible, however they are not recommended for use.