Page 369 - From GMS to LTE

VoLTE, VoWifi and Mission Critical Communication 355

messages before they forward them to the next network element and finally to the
terminating device. A good example is the SIP Invite message to begin a call. When it is
sent by the originating device to the P‐CSCF, the IPSec tunnel is terminated and a
‘P‐Asserted‐Identity’ SIP header is added. This is necessary as the originating device
cannot be trusted to insert its real identity. The network inserts this header to ensure
that the terminating device is informed of the originator’s true identity (phone number)
unless the originator has requested the network to hide the phone number.
At the S‐CSCF and the TAS the Invite message is then completely reconstructed
before being forwarded to the other end. Information that is removed is, for example,
the manufacturer name, model name, software version and the International Mobile
Equipment Identity (IMEI) that is contained in the ‘User Agent’ and ‘Contact’ SIP head-
ers and in the SDP ‘originator, o=’ parameter.
If the voice packets are not sent directly between the two devices but are required to
traverse a gateway in the network, which is quite common in VoLTE networks as fur-
ther described below, IP addresses and port numbers for the media stream are also
changed before the message is forwarded to the terminating device. In addition only
speech codecs and bandwidths supported by the network are forwarded. Other SIP
header and SDP parameters are also checked before they are forwarded to ensure that
only valid content is forwarded and to prevent direct exchange of data that is not ses-
sion‐related between the two ends. This is done to prevent SIP messaging being mis-
used to transfer data free of charge.
In summary it can be noted that the Invite and many other messages received at the
terminating side have little resemblance to the corresponding message that was origi-
nally sent out.

5.3.11 DTMF Tones
Another voice call feature still required in VoLTE today is Dual‐Tone Multi‐Frequency
(DTMF) tones as these tones are still used for interacting with a voice mail system or for
sending a conference bridge ID and password over an established voice connection. In
analog fixed telephone networks DTMF tones were generated by the phone itself and
sent as an audible tone over the speech channel. In GSM and UMTS networks, DTMF
tones are sent as signaling messages to the Mobile Switching Center. There, the mes-
sages are processed and converted into an audible tone in the media gateway. The
VoLTE implementation of DTMF tones is a mix of in‐band transmission and digital
signaling messages. Instead of sending a message to the other end to produce a tone
using a signaling connection, VoLTE embeds the signaling message in the RTP (Real‐
time Transport Protocol) media flow by replacing RTP speech packets with RTP DTMF
signaling messages, as shown in Figure 5.10. Usually, 20 ms of speech data are contained
in each RTP packet that is sent over UDP. Therefore to send DTMF tones, a DTMF
signaling message has to be sent every 20 milliseconds instead of a voice packet. GSMA
IR.92 points to 3GPP TS 26.114 Annex G [16], which in turn points to RFC 4733 [17]
for implementation details.
When the RTP DTMF signaling message arrives at a terminating VoLTE device, it is
the device’s responsibility to produce an audible tone for the user. If the terminator is not
a VoLTE device, a media gateway is required to transcode the speech path into a codec
suitable for the terminating network and the terminating device. As a consequence the