Four-wire analog trunk on Cisco
Channels of departmental and branch systems of conference calls have a common feature - work with four-wire ends of the tonal frequency.
In particular, in the energy sector this is the legacy of ADASE (Long-Range Automatic Communication Systems for Power Systems), which has been and continues to be used in dispatch and technological communication systems of energy enterprises.
Long-distance communication channels are used as trunks between the sites participating in the conference call, and the ACC BR equipment (Selector Communications Equipment Distribution Unit) is installed directly on the sites, working with four-wire terminations and providing connection to subscriber sets, coordinating reception and transmission levels and organizing the system interruption. "
Due to the fact that these communication systems have not been developed for a long time and are morally obsolete, with an increase in the number of sites for meeting participants, difficulties arise with the organization of four-wire transit - ADASE channel-forming equipment is absent at the site, and its installation is not practical.
But the rules are not set by us, but this system exists and continues to be operated. In addition, there are situations when the equipment on the site is already available and operates with 100% use of a possible resource, and the task is to organize a backup communication channel, or to include a remote regional department in the existing system, which further complicates the implementation, since the inclusion point in this case may be in the tens or even hundreds of kilometers from the site.
At the moment, you can find equipment of primary multiplexers that provide work with four-wire channels of PM (equipment like OPM-30, MP-30, NTS-1100, etc.). This option involves the installation of two sets of equipment (racks, processor boards, PM boards, power supplies, etc.) and the organization of an E1 or T1 channel between them, which will also require channel-forming equipment.
And if there is none of this, or the deployment is impractical, but is there Cisco equipment both at the point of presence of the ADACE channel terminations and at the site of the participant with organized data transfer via Ethernet? Moreover, Cisco routers are “voice” VK9 and are already used for office telephone channels. The situation under consideration is shown schematically in Fig. 1.
Two options are possible. The first option involves converting a four-wire circuit into a two-wire PSTN circuit and using Cisco voice modules such as VIC FXO / FXS with further voice transmission over VoIP protocols. At the same time, the used converters, in addition to implementing the differential circuit, must understand the PSTN signaling, but cannot transmit the signal to the four-wire circuit (the signaling is not transmitted in the four wire ends of the intercom channels, the connections are established by the operators at the communication nodes). That is, the device should receive a call signal from a 2-wire PSTN circuit and simply establish a 4-wire connection with separation of TX - RX speech channels and the ability to configure the required signal levels (-13 and + 4dB). This option is difficult to implement due to the lack of devices of this type, and the existing proposals bear a severe imprint of the “knee-high” assembly and failed to achieve stable operation from them. It is also impossible to use regular SUV, KTN and other channel pairing modules using a single-frequency code of interaction and control signals without allocating a signal channel, because there is simply nothing to accept this alarm, process and establish a connection. In the event of a broken connection, the system will automatically restore the channel (in fact, auto-dialing), which will require the organization of software “crutches” and further reduce the reliability of the circuit. There is still a whole range of issues requiring a solution in this implementation, but voiced more than enough to recognize this decision as unacceptable. It is also impossible to use regular SUV, KTN and other channel pairing modules using a single-frequency code of interaction and control signals without allocating a signal channel, because there is simply nothing to accept this alarm, process and establish a connection. In the event of a broken connection, the system will automatically restore the channel (in fact, auto-dialing), which will require the organization of software “crutches” and further reduce the reliability of the circuit. There is still a whole range of issues requiring a solution in this implementation, but voiced more than enough to recognize this decision as unacceptable. It is also impossible to use regular SUV, KTN and other channel pairing modules using a single-frequency code of interaction and control signals without allocating a signal channel, because there is simply nothing to accept this alarm, process and establish a connection. In the event of a broken connection, the system will automatically restore the channel (in fact, auto-dialing), which will require the organization of software “crutches” and further reduce the reliability of the circuit. There is still a whole range of issues requiring a solution in this implementation, but voiced more than enough to recognize this decision as unacceptable. process and establish a connection. In the event of a broken connection, the system will automatically restore the channel (in fact, auto-dialing), which will require the organization of software “crutches” and further reduce the reliability of the circuit. There is still a whole range of issues requiring a solution in this implementation, but voiced more than enough to recognize this decision as unacceptable. process and establish a connection. In the event of a broken connection, the system will automatically restore the channel (in fact, auto-dialing), which will require the organization of software “crutches” and further reduce the reliability of the circuit. There is still a whole range of issues requiring a solution in this implementation, but voiced more than enough to recognize this decision as unacceptable.
The second option is more standard, but, like the whole topic raised, it gives away some “exoticism”. This is the use of Cisco VIC E&M voice modules. Analog lines with E&M signaling use a six or eight wire circuit. Both two-wire and four-wire speech transmission schemes can be implemented, and dedicated lines of two or four conductors are used for signaling, depending on the type of E&M signaling used. The standard itself is considered obsolete, but is still supported by manufacturers. Cisco has two generations of two port voice modules VIC-2E / M and VIC2-2E / M. When choosing modules, you need to make sure that the router and IOS version support this type of VIC (for example, the 2800 series does not work with VIC-2E / M, while there will be no problems with the Cisco 1760).
E&M alarm has five types. We will not dwell on each type in detail, we only note that in our solution we will use the E&M Type V signaling system with immediate signal transmission. With such an alarm, switching of analog channels occurs without confirming readiness or checking the occupancy of voice ports. But this is also necessary, since the signaling circuits are not used at all (we simply have nothing to accept them), and the channel itself is supposed to be used in the “always on” mode. If necessary, the bandwidth is freed from voice traffic at the end of the conference call by the operator accompanying the meeting by programmatically disconnecting one of the voice interfaces.
Example of voice port configuration (VIC2-2E / M modules are installed in slots 0/2/0 of Cisco 2801 routers):
<- Participant site --->
interface FastEthernet0 / 0
ip address XXXX 255.255.255.252
h323-gateway voip bind srcaddr XXXX
dial-peer voice 101 pots
destination-pattern 101
port 0/2/0
dial-peer voice 102 voip
destination-pattern 102
session target ipv4: YYYY
codec g728
no vad
voice-port 3/0
operation 4-wire
type 5
signal immediate
input gain 14
output attenuation -6
no echo-cancel enable
timeouts call-disconnect 3
connection trunk 102
<- ADACE channel termination platform --->
interface Loopback0
ip address YYYY 255.255.255.255
h323-gateway voip bind srcaddr YYYY
dial-peer voice 102 pots
destination-pattern 102
port 0/2/0
dial-peer voice 101 voip
destination-pattern 101
session target ipv4: XXXX
codec g728
no vad
voice- port 0/2/0
operation 4-wire
type 5
signal immediate
input gain 14
output attenuation -6
no echo-cancel enable
timeouts call-disconnect 3
connection trunk 101 answer-mode
Answer-mode is set only on one side, i.e. one side initiates the trunk, and the other responds. The input gain and output attenuation parameters are set by applying a 800Hz test signal and adjusting the voice path to standard levels + 4dB reception and -13dB transmission, followed by voice verification.
ADACE systems use four-wire analog PM lines that do not have separate signaling circuits. The interaction signaling is transmitted in the speech channel at frequencies of 1200 Hz and 1600 Hz ± 5 Hz. Theoretically, such a trunk can also be used to turn on ADASE channels, but this mode of operation has not been tested.
The solution in question is implemented and used in existing networks.
In particular, in the energy sector this is the legacy of ADASE (Long-Range Automatic Communication Systems for Power Systems), which has been and continues to be used in dispatch and technological communication systems of energy enterprises.
Long-distance communication channels are used as trunks between the sites participating in the conference call, and the ACC BR equipment (Selector Communications Equipment Distribution Unit) is installed directly on the sites, working with four-wire terminations and providing connection to subscriber sets, coordinating reception and transmission levels and organizing the system interruption. "
Due to the fact that these communication systems have not been developed for a long time and are morally obsolete, with an increase in the number of sites for meeting participants, difficulties arise with the organization of four-wire transit - ADASE channel-forming equipment is absent at the site, and its installation is not practical.
But the rules are not set by us, but this system exists and continues to be operated. In addition, there are situations when the equipment on the site is already available and operates with 100% use of a possible resource, and the task is to organize a backup communication channel, or to include a remote regional department in the existing system, which further complicates the implementation, since the inclusion point in this case may be in the tens or even hundreds of kilometers from the site.
At the moment, you can find equipment of primary multiplexers that provide work with four-wire channels of PM (equipment like OPM-30, MP-30, NTS-1100, etc.). This option involves the installation of two sets of equipment (racks, processor boards, PM boards, power supplies, etc.) and the organization of an E1 or T1 channel between them, which will also require channel-forming equipment.
And if there is none of this, or the deployment is impractical, but is there Cisco equipment both at the point of presence of the ADACE channel terminations and at the site of the participant with organized data transfer via Ethernet? Moreover, Cisco routers are “voice” VK9 and are already used for office telephone channels. The situation under consideration is shown schematically in Fig. 1.
Two options are possible. The first option involves converting a four-wire circuit into a two-wire PSTN circuit and using Cisco voice modules such as VIC FXO / FXS with further voice transmission over VoIP protocols. At the same time, the used converters, in addition to implementing the differential circuit, must understand the PSTN signaling, but cannot transmit the signal to the four-wire circuit (the signaling is not transmitted in the four wire ends of the intercom channels, the connections are established by the operators at the communication nodes). That is, the device should receive a call signal from a 2-wire PSTN circuit and simply establish a 4-wire connection with separation of TX - RX speech channels and the ability to configure the required signal levels (-13 and + 4dB). This option is difficult to implement due to the lack of devices of this type, and the existing proposals bear a severe imprint of the “knee-high” assembly and failed to achieve stable operation from them. It is also impossible to use regular SUV, KTN and other channel pairing modules using a single-frequency code of interaction and control signals without allocating a signal channel, because there is simply nothing to accept this alarm, process and establish a connection. In the event of a broken connection, the system will automatically restore the channel (in fact, auto-dialing), which will require the organization of software “crutches” and further reduce the reliability of the circuit. There is still a whole range of issues requiring a solution in this implementation, but voiced more than enough to recognize this decision as unacceptable. It is also impossible to use regular SUV, KTN and other channel pairing modules using a single-frequency code of interaction and control signals without allocating a signal channel, because there is simply nothing to accept this alarm, process and establish a connection. In the event of a broken connection, the system will automatically restore the channel (in fact, auto-dialing), which will require the organization of software “crutches” and further reduce the reliability of the circuit. There is still a whole range of issues requiring a solution in this implementation, but voiced more than enough to recognize this decision as unacceptable. It is also impossible to use regular SUV, KTN and other channel pairing modules using a single-frequency code of interaction and control signals without allocating a signal channel, because there is simply nothing to accept this alarm, process and establish a connection. In the event of a broken connection, the system will automatically restore the channel (in fact, auto-dialing), which will require the organization of software “crutches” and further reduce the reliability of the circuit. There is still a whole range of issues requiring a solution in this implementation, but voiced more than enough to recognize this decision as unacceptable. process and establish a connection. In the event of a broken connection, the system will automatically restore the channel (in fact, auto-dialing), which will require the organization of software “crutches” and further reduce the reliability of the circuit. There is still a whole range of issues requiring a solution in this implementation, but voiced more than enough to recognize this decision as unacceptable. process and establish a connection. In the event of a broken connection, the system will automatically restore the channel (in fact, auto-dialing), which will require the organization of software “crutches” and further reduce the reliability of the circuit. There is still a whole range of issues requiring a solution in this implementation, but voiced more than enough to recognize this decision as unacceptable.
The second option is more standard, but, like the whole topic raised, it gives away some “exoticism”. This is the use of Cisco VIC E&M voice modules. Analog lines with E&M signaling use a six or eight wire circuit. Both two-wire and four-wire speech transmission schemes can be implemented, and dedicated lines of two or four conductors are used for signaling, depending on the type of E&M signaling used. The standard itself is considered obsolete, but is still supported by manufacturers. Cisco has two generations of two port voice modules VIC-2E / M and VIC2-2E / M. When choosing modules, you need to make sure that the router and IOS version support this type of VIC (for example, the 2800 series does not work with VIC-2E / M, while there will be no problems with the Cisco 1760).
E&M alarm has five types. We will not dwell on each type in detail, we only note that in our solution we will use the E&M Type V signaling system with immediate signal transmission. With such an alarm, switching of analog channels occurs without confirming readiness or checking the occupancy of voice ports. But this is also necessary, since the signaling circuits are not used at all (we simply have nothing to accept them), and the channel itself is supposed to be used in the “always on” mode. If necessary, the bandwidth is freed from voice traffic at the end of the conference call by the operator accompanying the meeting by programmatically disconnecting one of the voice interfaces.
Example of voice port configuration (VIC2-2E / M modules are installed in slots 0/2/0 of Cisco 2801 routers):
<- Participant site --->
interface FastEthernet0 / 0
ip address XXXX 255.255.255.252
h323-gateway voip bind srcaddr XXXX
dial-peer voice 101 pots
destination-pattern 101
port 0/2/0
dial-peer voice 102 voip
destination-pattern 102
session target ipv4: YYYY
codec g728
no vad
voice-port 3/0
operation 4-wire
type 5
signal immediate
input gain 14
output attenuation -6
no echo-cancel enable
timeouts call-disconnect 3
connection trunk 102
<- ADACE channel termination platform --->
interface Loopback0
ip address YYYY 255.255.255.255
h323-gateway voip bind srcaddr YYYY
dial-peer voice 102 pots
destination-pattern 102
port 0/2/0
dial-peer voice 101 voip
destination-pattern 101
session target ipv4: XXXX
codec g728
no vad
voice- port 0/2/0
operation 4-wire
type 5
signal immediate
input gain 14
output attenuation -6
no echo-cancel enable
timeouts call-disconnect 3
connection trunk 101 answer-mode
Answer-mode is set only on one side, i.e. one side initiates the trunk, and the other responds. The input gain and output attenuation parameters are set by applying a 800Hz test signal and adjusting the voice path to standard levels + 4dB reception and -13dB transmission, followed by voice verification.
ADACE systems use four-wire analog PM lines that do not have separate signaling circuits. The interaction signaling is transmitted in the speech channel at frequencies of 1200 Hz and 1600 Hz ± 5 Hz. Theoretically, such a trunk can also be used to turn on ADASE channels, but this mode of operation has not been tested.
The solution in question is implemented and used in existing networks.