To Be A Top Sales

Definition

As for Huawei Access equipment, many voice features, like POTS Access, MGCP Voice Feature and so on. This text mainly introduce MoIP on Huawei OLT, ONT, DSLAM.

The term "modem" is abbreviated from modulator and demodulator. Modem service is a data service that is widely applied on the PSTN network. In its earlier application, modem service is mainly used for point-to-point dialing and Internet dialup. Later, the service scope is expanded to cover point of service (POS) machine connection, alarming, and lottery machine connection.

Modem over Internet Protocol (MoIP) is a technology for providing modem services over an IP network or between an IP network and a PSTN network.

The dialup mode of modems can be pulse dialup and tone dialup.

In the traditional modem dialup scenario, PSTN users directly dial numbers, and the media stream passes through only the narrowband channel.

MoIP differs from the traditional modem dialup service in that MoIP users are next generation network (NGN) access gateway(AG) users, and that modem negotiation and media stream transmission between the AG and the trunk gateway are performed based on IP.

MoIP Services 

Modem services are usually transmitted using the V.32/V.34/V.90/V.92 protocol. Different modem protocols support different ranges of rates. In practical service application, the transmission rate is usually the result negotiated by the two modems at both ends.

V.32 supports a maximum rate of 14.4 kbit/s.

V.34 supports a maximum rate of 33.6 kbit/s. V.34 is the modem protocol most commonly used. It defines a negotiation process that is the same as the high-speed fax process. One application example of V.34 is the POS machine service.

V.90/V.92 supports a maximum rate of 56 kbit/s and is usually used for Internet dialup services.

Modem services are classified into high-speed modem services and low-speed modem services. The negotiation process of high-speed modem services is the same as that of high-speed fax services.

Based on the service rate, services with rates lower than or equal to 14.4 kbit/s are low-speed modem services, and services with rates higher than 14.4 kbit/s are high-speed modem services.

Based on the modem's requirements on the network, modems that require the network device involved to disable the echo cancelation function are high-speed modems, and modems that do not require so are low-speed modems.

Principle of MoIP 

Just like VoIP, MoIP supports the AG-based PSTN-IP-PSTN network structure, and also supports the PSTN-IP network structure.

MoIP Transmission Modes MoIP has two transmission modes:

Transparent transmission mode, which is also called the voice-band data (VBD) transparent transmission mode. In this mode, the AG uses the G.711 mode to encode and decode the modem signals, and processes the signals like processing common Real-time Transfer Protocol (RTP) data. In other words, the AG does not process the modem modulation signals, which are transparently transmitted on the IP network through VoIP channels. Figure 1 shows the MoIP transparent transmission mode.

Figure 1 MoIP transparent transmission mode

Principle of MoIP

High-speed modem: with voice activity detector (VAD) and echo canceller (EC) disabled.

Low-speed modem: with VAD disabled and EC enabled.

Redundancy mode, which is also called the relay mode.

NOTE: Currently, Huawei AGs support only the transparent transmission mode for MoIP. 

 Enhanced MoIP

What is FoIP?

Do you know FoIP ( Voice feature) in Huawei OLT and Huawei DSLAM(MA5616 and MA5600)? FoIP is Fax over Internet Protocol, a fax service provided on an IP network or between an IP network and a traditional PSTN. Figure 1 shows the network application of FoIP. 

Figure 1 FoIP network application

 Basic Process 

The basic process of fax can be described as follows:

The fax transmitting machine scans a page to obtain image information.

The fax transmitting machine digitalizes and compresses the image signals.

The fax transmitting machine modulates the image signals into analog signals, and transmits the signals to the fax receiving machine through common subscriber lines (as defined in the T.30 protocol).

Commonly Used Protocols for Fax 

The following protocols are commonly used for fax:

T.30: It is a fax protocol based on the PSTN network. T.30 defines in detail the process for transmitting fax signals on a PSTN network. It also defines the modulation mode (V.17/V.21/V.27/V.29/V.34) and transmission format (HDLC) of data, and the physical standard for fax signals. The T.30 fax messages and data can be transmitted transparently between AGs. This is called the T.30 transparent transmission mode. The quality of fax in this mode may not be high due to packet loss, delay, and packet disorder on the IP network.

T.38: It is a protocol that defines the process for carrying fax services over a packet IP network. Serving as a supplementary protocol to T.30, T.38 implements packet encapsulation based on T.30 in order to adapt to IP applications. In T.38 fax service, the T.38 redundancy mechanism is used to ensure service quality.

Typical Fax Process Defined in T.30 

T.30 defines 5 phases for a typical fax process, as shown in Figure 2

Phase A: A call for fax is established. This phase is similar to a telephone call establishment phase.

Phase B: The earlier phase of the packet transmission process. In this phase, the devices at both ends perform capability negotiation and training.

Phase C: This is the packet transmission phase and also the packet transmission control phase.

Phase D: The later phase of the packet transmission process. In this phase, packets are verified, errors are corrected, and multiple pages are continuously transmitted.

Phase E: The call for fax is released.

Typical fax process defined in T.30

Introduction to the R2 Feature

R2 signaling is a type of channel associated signaling (CAS) and it is also the international standard signaling based on E1 digital networks. Timeslot 16 in the R2 signaling is reserved for transmitting the signaling of the voice channel.

Actually, there is no agreed standard for R2 signaling. ITU-T Recommendation Q.400-Q.490 define the R2 signaling standards but different countries and regions have developed their own standards.

Purpose

The Huawei MA5600T/MA5603T/MA5608T connects an R2 PBX to the next generation network, achieving transformation of the public switched telephone network (PSTN) to the NGN.

R2 Principles

R2 signaling is inter-office channel associated signaling, and applies to international/national networks. R2 signaling is specified on both analog and digital transmission systems. R2 signaling contains line signaling and register signaling. Line signaling is available in three forms: DC line signaling, inband single-frequency pulse line signaling, and digital line signaling. For multi-end route transmission, the link-by-link forwarding mode is used. Register signaling can be transmitted in multi-frequency compelled (MFC) mode and dual tone multiple frequency (DTMF) mode.

Line Signaling

Line signaling is primarily used for monitoring the occupation, release and congestion states of a trunk line. Line signaling is classified into analog line signaling and digital line signaling. The Huawei access network devices MA5600T/MA5603T/MA5608T supports only the digital line signaling which will be described in detail in the following paragraph.

The digital line signaling uses timeslot 16 of the PCM for transmitting line signaling at a rate of 2048 kbit/s. To transmit line signaling of 30 voice channels, 16 frames form a multiframe. Timeslot 16 of frame 0 in the multiframe is used for multiframe synchronization. The first four bits of timeslot 16 in frame 1 correspond to the first voice channel, while the last four bits correspond to the 16th voice channel and so on.

Register Signaling

Register signaling is the signaling transmitted over a voice channel after the line signaling occupies the voice channel. Register signaling, including the selection signaling and service signaling, is used to transmit the control signals for a voice channel connection, such as managing the telephone network and selecting the route and the called party. The MFC register signaling will be described in detail.

The MFC register signaling includes the forward signaling and backward signaling, both of which are consecutive. The forward signaling is used to transmit the address information and control the indication information, while the backward signaling is used for acknowledgement and control. When transmitting a number, the transmit end stops transmitting the forward signaling only after receiving the acknowledgement from the backward signaling. Similarly, the receive end stops transmitting the backward signaling only after confirming that the transmission of the forwarding signaling is stopped. A control period can be divided into four steps, which is listed as follows:

Step 1: The user side sends the forward signaling.

Step 2: The network side receives the forward signaling and returns the backward signaling.

Step 3: The user side receives the backward signaling and stops sending forward signaling.

Step 4: The network side finds that the transmission of forward signaling is stopped, and stops sending the backward signaling.

The MFC register signaling uses the arithmetic frequency with 120 Hz as the common difference. This section takes the definition in Q.441 as an example, the forward signaling uses the high-frequency group (1380 Hz to 1980 Hz) and selects two out of six frequencies (1380 Hz, 1500 Hz, 1620 Hz, 1740 Hz, 1860 Hz, and 1980 Hz) for encoding. Up to 15 signaling combinations can be formed. The backward signaling uses the low-frequency group (780 Hz to 1140 Hz) and selects two out of four frequencies (780 Hz, 900 Hz, 1020 Hz, and 1140 Hz) for encoding. Up to six signaling combinations can be formed. To expand the signaling capability, the forward signaling is divided into group I forward signaling and group II forward signaling, while the backward signaling is divided into group A backward signaling and group B backward signaling.

Register signaling must always start with group I forward signaling. Group I forward signaling includes the information of the country code, echo suppressor indicator (I-11) and address signal (number: 1-9).

Group II forward signaling is the calling party's category signaling sent by an outgoing R2 register. It is used to reply to backward signal A-3 (the address-complete signal) or A-5 (the request signal for a calling party's category), and send the national or international calling information.

Group A backward signaling is used to acknowledge the group I forward signaling and under some conditions, group II forward signaling, such as acknowledging the calling party's type and group II forward signals.

Any group B backward signaling acknowledges the group II forward signaling and is always preceded by the address-complete signal A-3. Signal A-3 indicates that the incoming R2 register has received all the required forward signals from the outgoing R2 register.

Register signaling is the in-band signaling (the frequency is within the voice frequency band). Therefore, the register signaling is transmitted over the voice channel.

R2 Application in Access Networks

This section takes H.248 protocol as an example. A PBX is connected to the next generation network (NGN) through R2 signaling. Interfaces between an MG and an MGC use the H.248 protocol, and interfaces between an MG and a PBX use the R2 protocol. Figure 1 shows the networking of connecting the R2 PBX to the NGN network.

To connect a PBX to the NGN network through R2 signaling, MGs need to perform the conversion between R2 signaling and H.248 signaling.

In the upstream direction, the MA5600T/MA5603T/MA5608T terminates R2 signaling transmitted from the PBX, converts R2 signaling to H.248 signaling, and sends the converted signaling to the softswitch.

In the downstream direction like GPON service board, the MA5600T/MA5603T/MA5608T terminates H.248 signaling transmitted from the softswitch, converts H.248 signaling to R2 signaling, and sends the converted signaling to the PBX.

Figure 1 Network example of connecting the R2 PBX to the NGN network

R2 Standards and Protocols Compliance

The reference standards and protocols of this feature are as follows:

draft-manyfolks-megaco-caspackage-01.txt

draft-laha-megaco-cas-mntc-00.txt

draft-ietf-megaco-r2package-03.txt

ITU-T H.248.25 Gateway control protocol: Basic CAS packages

ISDN System Structure

Figure 1 shows the ISDN system structure of voice feature on Huawei OLT/DSLAM products..

Figure 1 ISDN System Structure

The ISDN users include the BRA users and PRA users.

The BRA users can connect the ISDN telephone with the NT1 directly, or connect the common telephone through the TA. On the MG side, the BRA users access to the network through the BRA port. Connect the NT1 and MG with the ordinary telephone line.

The PRA users access the network through the E1 port on Huawei E1 service processing board like SSN2PQ1A with the PBX. Connect the PBX and the gateway with the E1 cable.

ISDN Call Control Process

For the H.248 protocol or the MGCP protocol, the ISDN uses primitive Q931 for call control. Between the gateway and the NT1, a Layer 2 link that complies with the Q921 protocol is set up between the gateway and the PBX to carry the Q931 message. The IUA link is set up between the gateway and the softswitch to carry the Q931 message.

For the SIP protocol, the gateway converts the ISDN message to the SIP message and then the ISDN uses the SIP protocol for call control. The gateway between the ISDN terminal and the IMS implements signaling conversion and service control. Specifically, the gateway translates the DSS1 signaling to the SIP signaling and then sends the signaling to the IMS. Generally, the IMS transparently transmits the SIP signaling to the peer ISDN user. Then, the peer ISDN user translates the SIP signaling to the DSS1 signaling and then sends the signaling to the ISDN terminal. This process enables two ISDN terminals to be connected with each and therefore the ISDN service is provisioned.

Figure 2 shows the ISDN call control process.

Figure 2 ISDN call control process 

For the H.248 protocol or the MGCP protocol, the gateway does not process the primitive Q931 but takes out the primitive terminal Q931 from the Q921 message, encapsulates the Q.931 to the IUA message, and then sends to the softswitch. Resources are not assigned to the primitive Q931.

For the SIP protocol, the gateway converts the ISDN message to the SIP message.

The call process includes two sections: call setup and call disconnection.

The call setup process is as follows:

The host hooks off and initiates a call setup.

The softswitch responds "SETUP_ACK", and applies more call information, such as the called number.

The calling party dials, and the number is carried by the primitive IMFORMATION to the softswitch.

The softswitch responds "CALL PROCEEDING", and the call is setting up.

The softswitch applies sending setup to the called party to set up a call.

After receiving the call, the called party starts ringing and sends "ALERTING". If the "ALERTING" reaches the calling party, the call is connected.

The called party hooks off and sends "CONNECT". If the "CONNECT" reaches, the call is connected.

The calling party responds "CONNECT_ACK". The call setup is complete.

The call disconnection process is as follows:

One party hooks on, and sends "DISCONNECT".

The softswitch sends "DISCONNECT" to the other party, and sends "RELEASE" to the party who hooks on.

The party who hooks on finishes the call disconnection, and sends "RELEASE_COMPLETE" to the softswitch.

After receiving the disconnection, the other party sends "RELEASE" to the softswitch.

The softswitch responds "RELEASE_COMPLETE".

What is ISDN?

The integrated services digital network (ISDN) is a CCITT standard, providing integrated transmission service for voice, video, and data. The ISDN enables the voice, video, and data to be transmitted on the data channel simultaneously.

The ISDN supports two types of services:

Basic rate interface (BRI): provides a rate of 144 kbit/s, including two B channels and one D channel. The rates of B and D channels are 64 kbit/s and 16 kbit/s, respectively.

Primary rate interface (PRI): provides a rate of 2.048 Mbit/s, including 30 B channels and one D channel. The rates of both B and D channels are 64 kbit/s.

ISDN networks support B and D channels, shown in Figure 1

B channels are used for carrying services.

D channels are used for transmitting call control signaling as well as maintenance and management signaling.

 Figure 1 ISDN channels 

Purpose

The ISDN access on the media gateway provides integrated transmission services, such as voice, video, and data for the users.

ISDN Reference Model 

Figure 1 shows the ISDN reference model.

The ISDN user accesses the MA5600T/MA5603T/MA5608T through the U reference point. The actual terminal on the user side may support NT 1, NT 2, and TE 1 functions at the same time. When VoIP is used for upstream transmission, the IUA protocol is used to load the Q.931 call signaling of the ISDN between the MG and MGC, and the H.248 protocol or MGCP signaling is used to control the media connection on the MG.

ISDN Protocol Stack Model 

Figure 2 shows the mapping relationship between the ISDN protocol and OSI model. Layers in the ISDN protocol stack map the physical layer, data link layer, and network layer in the following OSI model.

ISDN physical layer: For users, the ISDN physical layer is on S reference point or T reference point. This layer has the following major functions: coding, full-duplex transmission, channel multiplexing, port activation and depolarization, feeding, and termination identification. This layer can multiplex multiple links at the data link layer and use AMI, 4B3T and 2B1Q for coding.

ISDN data link layer: ISDN does not define Layer 2 protocols dedicated to B channels. Any Layer 2 protocols can be used between two communicating devices after negotiation as long as they can transparently transmit data on B channels. The link access procedure on the D channel (LAPD) protocol defined in Q.921 (a reliable transport protocol) is used for D channels, which is mainly used to carry messages and data generated by Layer 3 entities.

ISDN network layer: ISDN does not define Layer 3 protocols dedicated to B channels. Layer 3 protocol Q.931 for D channels is mainly used to control and manage connection setup and release on B channels.

Figure 2 Mapping relationship between the ISDN protocol stack model and OSI model

ISDN Protocol Processing Model 

Figure 3 and Figure 4 show the ISDN protocol processing model. ISDN involves the following protocols:

Q.921: Defines LADP. It is a reliable transport protocol.

Q.931: Defines the procedure of processing and controlling messages and state machines that are used for calls (including circuit switching calls and packet switching calls) between the user-side device and network-side device.

SCTP: A transport protocol in the SIGTRAN protocol stack, which is a reliable transport protocol on top of protocols (such as IP) providing unreliable transmission services. SCTP transmits acknowledged, error-free, and repetition-free data and ensures real time transmission to some extent.

IUA: ISDN Q.921 user adaption layer protocol

Figure 3 H.248 protocol processing model

Exchange between protocols

For the H.248 protocol:

Data transmission from ISDN user terminals to voice boards uses Q.921 and Q.931.

Voice boards terminate Q.921 messages and send Q.931 messages to the CPU of the control board using the master-slave serial port communications protocol. Then the CPU of the control board uses IUA to packetize Q.931 signaling carried on SCTP links and sends the packed signaling to the MGC through the LAN switch. In this case, Q.921 is not used between the AG and MGC; instead, Q.931 and IUA are used.

The MGC converts the IUA packets containing Q.931 signaling packed by using IUA to Q.931 signaling. Also, the MGC sends Q.931 signaling to the peer end through SCTP links. This is the entire process of ISDN call signaling.

For the SIP protocol:

Data transmission from ISDN user terminals to voice boards uses Q.921 and Q.931.

Standards of the POTS Line Interface 

Major standards of the voice line interface are as follows:

ITU-Q552: It defines transmission specifications of the Z interface.

ES 201970: It defines basic hardware features of the voice interface.

YD751 - Network Entry Checking Methods for Telephone Exchange Equipment: It describes the voice interface standards defined by China.

The voice technologies have gone through a long period of development. Almost every country has its own standards. The preceding standards are related to the basic features of boards. Special features must be tailored to meet requirements of different countries. For example, British Telecom (BT) has its own standards, namelyBTNR315. Some of the requirements in the BTNR315 are very special and can be met only by special boards.

Basic Features of the Z Interface

Basic features of the Z interface supported by the voice interface board of the MA5600T/MA5603T/MA5608T are summarized as follows:

Battery feeding (B)

Batter feeding refers to the supply of the voltage and the current to terminals (such as telephones) to ensure the normal operation of terminals.

When the telephone is in the on-hook state, the voltage of the MA5600T/MA5603T/MA5608T board is generally 48 V. When the telephone is in the off-hook state, the MA5600T/MA5603T/MA5608T board supports the constant-current feeding at 20 mA, 25 mA, or 30 mA. The feeding current can be configured according to the actual requirement.

The off-hook feeding of the port can be automatically adjusted. If the length of the loop is short, the port is fed with the constant current. If the length of the loop is long, the port automatically adjusts the loop current based on the preset threshold. This design ensures the compliance with the related standards and optimizes the power consumption of the port.

If the feeding current is 25 mA and the voltage is -48 V, the feeding current is equal to or larger than 25 mA when the loop resistance is less than 1200 ohm, and the feeding current is larger than 18 mA when the loop resistance is 1800 ohm.

MA5600T/MA5603T/MA5608T boards also support the 40-mA feeding current. The 40-mA feeding current increases the power consumption of ports and thus is not recommended. If the 40-mA feeding current is configured, the number of ports configured with the 40-mA feeding current cannot exceed five on each board.

 Ringing (R)

Ringing refers to the supply of the ring current to telephones so that telephones can ring to inform subscribers of incoming calls. Huawei OLT boards are designed with the balanced ringing feature.

The concept of the balanced ringing is put forward based on the concept of the traditional imbalanced ringing. The traditional imbalanced ringing is classified into two types: (1) In a subscriber line, A line is 0 V and B line is -48 V DC overlaid with the 75 Vrms AC signals. (2) In the subscriber line, A line is -48 V and B line carries the 75 Vrms AC signals. In the case of the balanced ringing, both A and B lines of the subscriber line have the AC signals. The AC signals of the A and B lines are of the same frequency and opposite phases, that is, differential signals. The frequency of signals in the case of the balanced ringing can be set to 16 Hz, 25 Hz, or 50 Hz.

The amplitude of the ringing current can reach up to 70 Vrms. The amplitude of the ringing current on a terminal can exceed 35 Vrms if the line impedance is 1400 ohm (5-km lines with the core diameter of 0.4 mm) and the terminal impedance is 4000 ohm. The amplitude of the 50-Vrms ringing current is configurable. This configuration is mainly applicable to the short loop with a length less than one kilometer, aiming to substantially reduce the power consumption of ringing on the ports.

The DC offset provided by boards can reach 20 V, which ensures reliable ringing when the distance is long.

The break-make ratio of the ringing current can be configured to meet requirements of different carriers in the world.

Over-voltage protection (O)

          Over-voltage protection is one of the interface protection measures. 

Supervision (S)

          Supervision refers to the detection of telephone state, such as on-hook, off-hook, and off-hook in the ringing state. The terminal state can be learned through detection. The terminal state detection is the basis of some calls. 

Code/Decode (C)

          Coding/Decoding refers to the process that analog signals of the subscriber line are converted into digital signals and compressed according to the A/U law. 

H - Hybrid circuit

          Hybrid circuit refers to the conversion from the 2-wire analog interface to the 4-wire digital interface on the board and implementation of the balanced matching with the impedance of the subscriber line. 

Test (T)

Interface Impedance, Transmission Specifications, and Gain 

The voice interface board of the MA5600T/MA5603T/MA5608T supports the configuration of the interface impedance and gain.

At present, sixteen common interface impedances can be configured, see as theTable 1:

The interface transmission gain is also configurable. The send gain is generally in the range of +4 dB and -6 dB and the receive gain is in the range of 0 dB to -12 dB. The gain can be configured at the step of 0.5 dB.

The transmission specifications of the boards are fully compliant with the ITU-Q522 test requirements. If the interface impedance is not one of the preceding eight types, independent software can be developed to support the interface impedance.

Digit Collection 

The voice interface board of the MA5600T/MA5603T/MA5608T supports the pulse-based digit collection.

Old-fashioned telephones generally adopt the pulse dialing mode, while new telephones adopt the DTMF dialing mode. Most telephones support the pulse dialing mode.

The service boards support the pulse-based digit collection at the speed of 8 pps to 12 pps. The break-make ratio is in the range of 50% and 80%. The interval of pulses is configurable and is in the range of 100 ms and 2 s. The default interval of pulses is 300 ms.

The DTMF digit collection is completed by the DSP instead of the service boards.

Charging Signals

Service boards support three charging modes, namely polarity reverse, 12/16KC, and counter impulse delivery.

Polarity reverse: The voltage polarity between A and B lines of the subscriber line is reversed. Some terminals detect this type of reverse for charging purpose.

12/16KC: The service board sends the 12000 Hz/16000 Hz sine AC signals at a specific interval to the terminals.

Counter impulse delivery: The service board sends pulse signals to the terminals. Charging is implemented based on the pulse signals.

All ports of the service board support both the fast and slow polarity reverse features. Fast polarity reverse is generally completed within 3 ms, which meets the time requirements of polarity reverse of some telephones. Slow polarity reverse is generally completed within 80 ms. It can substantially reduce the interference to the line during the polarity reverse and is compatible with the DSL transmission on the same line.

The service board supports the 12/16KC charging. In the 12/16KC charging mode, the amplitude of the 12/16KC signals is configurable. The amplitude can be set to 0.45 Vrms, 0.775 Vrms, 1 Vrms, 1.5 Vrms, 2 Vrms, or 2.5 Vrms. The maximum value is 2.5 Vrms (200 ohm). In addition, the break-make ratio of KC signals is also configurable. By default, the Make duration is 100 ms and the Break duration is 300 ms. Both the Make duration and the Break duration range from 10 ms to 500 ms.

The service board also supports the counter impulse delivery charging. Some attributes of this charging mode, such as the pulse width and number of pulses sent per minute, are configurable.

Current Reduction of Locked Ports

When a phone connected to a port is in off-hook state for a long time but the conversation is not going on, the service board can lower the current of the port to less than 12 mA to reduce the power consumption of the port.

Short Loop Feeding

When the length of the line is short, the service board uses the low voltage for feeding to reduce the power of the port. When the length of the line becomes long, the service board automatically uses the voltage higher than the previous low voltage to meet the application requirement.

Power Cut-off

Feeding of ports that are not allocated with numbers can be cut off to reduce the power consumption of the ports.

On-Hook Transmission

Service boards support the on-hook and off-hook transmission functions, such as the caller identification display service and the fixed network short message service.

Ringer Equivalence Number

What is POTS access?

Plain Old Telephone Service (POTS) is the traditional basic telephony service provided using twisted-pair copper lines. POTS Voice like Huawei HG8110F     ( 1*FE+1*POTS) interface features are the features implemented on the voice interface. The purpose is to provide the standard-compliant voice interface that has the reliable protection capability and intelligent energy-saving function.

Ringing Feature

A subscriber's phone rings, indicating that an incoming call is waiting on the line. The POTS board plays the ring tone to plain old telephone service (POTS) subscribers, and terminals play the ring tone to integrated services digital network (ISDN) subscribers. This section describes the ring tone played by the POTS board.

Ringing Break-Make Ratio of the Ring Tone Played by the POTS Board

Ringing The break-make ratio varies depending on countries. For example, the break-make ratio in China is 1:4, that is, the ring tone is played in the cycle of one-second ringing followed by four-second mute. Figure 1 shows a six-segment ring with break-make ratio 0.4:0.2:0.4:0.2:0.4:2.6, which cycles in the following mode: ringing for 0.4 seconds and mute for 0.2 seconds twice followed by 0.4-second ringing and 2.6-second mute.

Ringing Application

In some countries, break-make ratio is different for different services. For example, break-make ratio 1:4 is used for local calls, 1:2 for toll calls, and 0.4:0.2:0.4:4 for intra-Centrex calls.

POTS Interface Protection

Introduction

Service boards are connected to user terminals through subscriber cables. Some subscriber cables may be routed under the ground or overhead and some may be routed in parallel with the mains AC power cables. In these cases, a high voltage may be generated on subscriber cables because of the lightning attack, contact with power lines, and induction of power lines. The high voltage may damage the ports on service boards. Therefore, service boards must be equipped with the protection capability to prevent occurrence of the preceding problems.

Version Support

The MA5600T/MA5603T/MA5608T fully complies with the requirements specified in the ITU K20/K45. The protection standards formulated by Huawei are stricter than the requirements specified in the ITU K20/K45. For example, the ITU requires that ports can withstand a voltage of 1500 V when the level-1 protection is not available. Tests show that Huawei boards generally can withstand a voltage of 4000 V when the level-1 protection is not available.

POTS IP SPC

The semi-permanent connection (SPC) exclusively occupies a constant voice channel to meet the communication requirements and ensure the communication quality for special and key access users. To configure an IP SPC, configure the data (including the local IP address, local UDP port, remote IP address, and remote UDP port), set up an IP direct connection between the two ends of the voice service. In this manner, the voice media data can be directly transmitted to the peer end.

Networking Description 

Figure 1 shows the networking of a POTS IP SPC. An IP SPC is set up between the POTS port on AG 1 and POTS port on AG 2. Users under these two POTS ports are permanently online and can communicate with each other without dialing a number. Figure 1 Networking of a POTS IP SPC 

Application Scenarios A POTS IP SPC is mainly used in the following two scenarios:

Special modems are connected to a POTS port and these modems require direct communication without dialing and require online permanently.

H.248 Voice feature on Huawei Access Devices has  H.248-Based VoIP, H.248-Based MoIP, H.248-Based FoIP. VoIP, MoIP and FoIP are similar, here is a detailed introduction of H.248-Based VoIP for an example. 

Figure 1 illustrates the principle of the call establishment and release in the H.248-based VoIP service on Huawei OLT, ONT(like HG8326R), DSLAM.

Figure 1 Principle of the VoIP feature that supports the H.248 protocol

Figure 2 illustrates the basic flow of a call establishment and release. Figure 2 H.248-based call flow 

MG-0 detects the offhook of user A0, and notifies the MGC of the offhook event through the Notify command.

After receiving the offhook event, the MGC sends a digitmap to MG-0, requests MG-0 to play the dial tone to user A0, and at the same time checks for the digit collection event.

User A0 dials a telephone number, and MG-0 collects the digits according to the digitmap issued by the MGC. Then, MG-0 reports the result of digit collection to the MGC.

The MGC sends the Add command to MG-0 for creating a context and adding the termination and RTP termination of user A0 into the context.

After creating the context, MG-0 responds to the MGC. The response contains the session description that provides the necessary information for the peer end to send the packet to MG-0, such as the IP address and UDP port number.

The MGC sends the Add command to MG-1 for creating a context and adding the termination and RTP termination of user A1 into the context, and then issues the IP address/UDP port number of user A0 to user A1.

After creating the context, MG-1 responds to the MGC. The response contains the session description that provides the necessary information for the peer end to send the packet to MG-1, such as the IP address and UDP port number.

MG-1 detects the offhook of user A1, and then reports the offhook event to the MGC. The softswitch (MGC) sends the Modify command to stop the ring back tone of user A0 and the ringing of user A1.

The MGC sends the session description of MG-1 to user A0 through the Modify command. Then, the conversation is set up between users A0 and A1.

MG-0 detects the onhook of user A0, and notifies the MGC of the onhook event through the Notify command.

The MGC sends the Modify command to MG-0 and MG-1 respectively to modify the RTP mode to receive-only.

The MGC sends the Modify command to MG-1 requesting MG-1 to play the busy tone to user A1, and at the same time checks for the onhook event.

The MGC sends the Subtract command to MG-0, requesting MG-0 to release the resources that are occupied by the call of user A0.

MG-1 detects the onhook of user A1, and notifies the MGC of the onhook event through the Notify command.

The MGC sends the Subtract command to MG-1, requesting MG-1 to release the resources that are occupied by the call of user A1.

The call between users A0 and A1 is terminated, and all the resources occupied by the call are released.

What is H.248 on the equipment of Huawei MA5600T/MA5603T/MA5608T, Huawei MA5616, and MA5622A...?

H.248 is a media gateway control protocol through which the media gateway controller (MGC) controls the media gateway (MG) so that interoperability is implemented between different media. ITU-T issued the first version of this protocol in June 2000.

Purpose Compared with MGCP, H.248 has the following merits:

Supports more types of access technologies, and is more thorough and complete in standardization

Compensates for the deficiency of MGCP in descriptiveness, is applicable to larger networks and has better extensibility and flexibility

Carried on various protocols, such as UDP/SCTP (MGCP is carried on UDP)

Mechanism of the H.248 Protocol 

A termination ID identifies a termination that is going to register or deregister a service. The termination ID of each termination is unique. During service configuration, the termination ID corresponding to each termination must be configured on the MG and the MGC. The root termination ID represents an entire MG. The ServiceChange command executed on the root termination ID is effective on an entire MG. The wildcarding principle is that the ALL wildcard (*) can be used but the CHOOSE wildcard ($) cannot be used.

Registration Mechanism of the H.248 Interface The MG sends the ServiceChangeRequest command to inform the MGC that a user or a group of users are about to register or deregister service. After this command is executed successfully, the termination status is changed to InService or OutOfService. In addition, the MGC can unsolicitedly send the ServiceChangeRequest command to request the MG to register or deregister service for a user or a group of users.

Figure 1 shows the registration flow of the MG.

Description of the flow:

The MG sends the ServiceChangeRequest command to the MGC. In the command, TerminationId is Root, Method is Restart, and ServiceChangeReason is 901 (cold boot, registering for the first time after power-on), 902 (warm boot, through command lines), or 900 (in other cases).

The MGC sends the Reply message to the MG indicating the successful registration.

The MGC sends the Modify command to the MG requesting the MG to detect the offhook of all users (al/of).

The MG responds to the MGC with the Reply message.

Heartbeat Mechanism of the H.248 Interface

After the registration is successful, the MG and the MGC maintain communication by sending each other the heartbeat message Notify (it/ito). By default, the heartbeat message is sent every 60s. The sending interval can be set within the range of 5-655s.

After the MG sends the first heartbeat message to the MGC, if the MG does not receive the heartbeat response from the MGC before the preset interface heartbeat timer (for example, the length of three sending intervals) times out, the MG sets the interface status to "wait for response". Then, the MG keeps initiating a registration with the MGC. If dual-homing is configured, the MG initiates registration with the two MGCs alternatively. The registration is initiated once every 30s, every three trials of registration are one round, and every registration message is re-transmitted 7 times. Therefore, 24 registration messages in total are transmitted within 90s. Then, the MG starts the next round of registration with the other MGC.

Deregistration Mechanism of the H.248 Interface

Figure 2 shows the unsolicited deregistration flow of the MG.

Figure 2 Unsolicited deregistration flow of the MG

Description of the flow:

The MG sends the ServiceChangeRequest command to the MGC. In the command, TerminationId is Root, Method is Forced, and ServiceChangeReason is 905 ("905" indicates that the termination is taken out of service because of maintenance operation, and now the MG uses "905" to initiate a deregistration request through command lines).

The MGC sends the Reply message to the MG indicating a successful deregistration.

Figure 3 shows the flow of the MGC unsolicitedly deregistering the MG.

Figure 3 Unsolicited deregistration flow of the MGC

Description of the flow:

The MGC sends the ServiceChangeRequest command to the MG. In the command, TerminationId is Root, Method is Forced, and ServiceChangeReason is 905.

The MG responds to the MGC with the Reply message. The MA5600T/MA5603T/MA5608T (MG) supports the registration and deregistration of not only an entire MG but also a single termination. The service status of a single user can be changed through the registration and deregistration of a single termination.

Authentication Mechanism of the H.248 Interface Authentication is a security mechanism through which the MGC authenticates the legality of the MG user. The purpose of authentication is to prevent unauthorized entities from establishing illegal calls or interfering with legal calls through the H.248 or MGCP protocol. Authentication can be implemented only when it is also supported by the softswitch interconnected with the MG.

In H.248, the implementation of authentication complies with RFC2402.

MD5 is adopted as the encryption algorithm.

Figure 4 shows the authentication flow.

The basic flow is as follows:

The MG sends the ServiceChange command to register with the MGC. The command contains the digital signature of the MG.

After receiving the ServiceChange command, the softswitch verifies the MG and sends a reply.

The softswitch sends the Modify message to the MG. The message contains the required algorithm ID and random number.

The MG verifies the message sent by the softswitch and sends a reply.

The softswitch authenticates the MG periodically.

On Huawei access products such as MA5600T/MA5603T/MA5608T, MA5616, and MA5622A, there are three MGCP voice principles:  MGCP-Based VoIP, MGCP-Based MoIP, MGCP-Based FoIP. 

MGCP-Based VoIP

Figure 1 illustrates the principle of the call establishment and release in the MGCP-based VoIP service.

Figure 1 Principles of the call establishment and release in the MGCP-based VoIP service

Figure 2 illustrates the basic flow of a call establishment and release process. Figure 2 MGCP-based call flow 

AG-0 detects the offhook of EP0, and notifies the MGC of the offhook event through the Notify command.

After the MGC receives the offhook event, the MGC sends a digitmap to AG-0, requests AG-0 to play the dial tone to EP0, and at the same time checks for the digit collection event.

User EP0 dials a telephone number, and AG-0 collects the digits according to the digitmap issued by the MGC. Then, AG-0 reports the result of digit collection to the MGC.

The MGC sends the CRCX (CreateConnection) command to AG-0 requesting AG-0 to create a connection at endpoint EP0.

AG-0 allocates resources for creating this connection and sends a response to the MGC. The response contains the session description that provides the necessary information for the peer end to send the packet to AG-0, such as the IP address and UDP port number.

The MGC sends the CRCX command to AG-1 requesting AG-1 to create a connection at endpoint EP1.

AG-1 allocates resources for creating this connection and sends a response to the MGC. The response contains the session description that provides the necessary information for the peer end to send the packet to AG-1, such as the IP address and UDP port number.

AG-1 detects the offhook of EP1, and sends the Notify command to the MGC. The softswitch (MGC) sends the MDCX (ModifyConnection) command to stop the ring back tone of EP0 and the ringing of EP1.

The MGC sends the session description of AG-1 to EP0 through the MDCX command. Then, the conversation is set up between EP0 and EP1.

AG-0 detects the onhook of EP0, and notifies the MGC of the onhook event through the Notify command.

The MGC sends the MDCX command to AG-0 and AG-1 respectively to modify the RTP resource to receive-only.

The MGC sends the MDCX command to AG-1 requesting AG-1 to play the busy tone to EP1, and at the same time checks for the onhook event.

The MGC sends the DCLX (DeleteConnection) command to AG-0, requesting AG-0 to release the resources that are occupied by the call of EP0.

AG-1 detects the onhook of EP1, and notifies the MGC of the onhook event through the Notify command.

The MGC sends the DCLX command to AG-1, requesting AG-1 to release the resources that are occupied by the call of EP1.

The call between EP0 and EP1 is terminated, and all the resources occupied by the call are released.

MGCP-Based MoIP 

MoIP refers to the modem service provided on the IP network or between the IP network and the traditional PSTN network. According to different control devices, MoIP can be classified as softswitch-controlled MoIP and auto-switching MoIP.

Softswitch-Controlled MoIP The basic flow of the softswitch-controlled MoIP service is as follows:

Establish a call. If the MoIP service is configured on the softswitch, the softswitch sends a command to the MG instructing the MG to detect the modem event.

The calling party and called party start communicating with each other.

During the call, when the MG detects the ANS or ANSAM modem start event (low-speed modem signal), or detects the ANSBAR or ANSAMBAR modem start event (high-speed modem signal), the MG sends the event to the softswitch.

According to the event, the softswitch sends a command instructing the MG to switch the DSP channel of the calling and called parties to the low-speed or high-speed modem mode.

According to the command sent by the softswitch, the MG switches the DSP channel to the corresponding modem mode. At this stage, the MG adopts the encoding format and port number specified by the softswitch.

The settings of echo cancellation (EC), voice activity detection (VAD), and DSP working mode are as follows:

After the modem data is transmitted, if the conversation proceeds, the DSP working mode does not automatically switch from the modem mode to the voice mode, because the modem end event is not issued. As a result, the quality of the voice service may be affected.

Low-speed modem: EC-ON, VAD-OFF, DSP working mode-modem mode

High-speed modem: EC-OFF, VAD-OFF, DSP working mode-modem mode

Auto-Switching MoIP The basic flow of the auto-switching MoIP service is as follows:

Set up a conversation.

The MGs at both ends check for the modem event on the IP side and the TDM side. When the modem event is detected, if the modem transmission mode is configured as auto-switching, the coding mode is switched to G.711 (the a/μ law is configurable), and the DSP parameters are modified according to the modem mode (high-speed/low-speed) detected.

When the modem service is terminated, the call is released.

MGCP-Based FoIP 

FoIP refers to the fax service provided on the IP network or between the IP network and the traditional PSTN network. The fax machine can be regarded as a special modem. In the FoIP negotiation, the modem negotiation is performed before the fax negotiation.

According to the transmission protocol adopted, there are two modes of fax services carried on the IP network: the T.30 transparent transmission mode and the T.38 mode. According to different control devices, FoIP can be classified as softswitch-controlled FoIP and auto-switching FoIP.

Softswitch-Controlled FoIP

The fax service can be classified into high-speed fax and low-speed fax. The softswitch-controlled low-speed fax service supports the T.30 transparent transmission mode and the T.38 mode. The basic service flow is as follows:

Configure the fax service and fax flow on the MGs and the softswitch.

After the voice channel is set up, the softswitch instructs the MG to detect the fax event and modem event.

When detecting the fax event, the MG reports the event to the softswitch. The event can be a low-speed modem event (ANS or ANSAM) or a low-speed fax event (V.21Flag).

According to the preset fax flow, the softswitch instructs the MGs at both ends to change the DSP channel working mode to the T.30 transparent transmission mode or T.38 mode.

The fax starts.

After the fax is complete, if the MG detects the fax end event, the MG reports the event to the softswitch.

The softswitch instructs the MGs at both ends to change the DSP channel working mode to the voice mode.

The voice service proceeds.

The softswitch-controlled high-speed fax service supports the T.30 transparent transmission mode. The basic service flow is as follows:

Configure the fax service and fax flow on the MGs and the softswitch.

After the voice channel is set up, the softswitch instructs the MG to detect the fax event and modem event.

When detecting a fax event, the MG reports the event to the softswitch. The event can be a high-speed modem event (ANSBAR or ANSAMBAR) or a low-speed fax event (V.21Flag; if the peer end is a low-speed fax machine or the network quality is poor, the fax speed is automatically decreased and this event is reported).

According to the preset fax flow, the softswitch instructs the MGs at both ends to change the DSP channel working mode to T.30 transparent transmission mode.

The fax starts.

After the fax is complete, if the MG detects the fax end event, the MG reports the event to the softswitch.

The softswitch instructs the MGs at both ends to change the DSP channel working mode to the voice mode. The voice service proceeds.

Auto-Switching FoIP

The auto-switching fax service supports the T.30 transparent transmission mode and the T.38 mode. The basic service flow is as follows:

Configure the auto-switching fax service on the MGs at both ends.

Establish a call and use the voice service.

The MG checks for the fax event on the IP side and the TDM side. When detecting the fax event, the MG changes the DSP channel working mode to the T.30 transparent transmission mode or the T.38 mode.

After the fax is complete, when the MG detects the fax end event, the MG changes the DSP channel working mode to the voice mode.

The voice service proceeds.

Common Fax Protocols

Two protocols are usually used for implementing the fax service on the packet voice network: the ITU-T Recommendation T.30 and ITU-T Recommendation T.38.

T.30 is based on the PSTN network. T.30 particularly defines the flow for transmitting fax signals on the PSTN network. It also defines the modulation mode (V.17/V.21/V.27/V.29/V.34) and transmission format (HDLC) of data, and the physical standard for fax signals. The T.30 fax messages and data can be transmitted transparently between MGs. This is called the T.30 transparent transmission mode. The quality of the fax in this mode may not be high due to packet loss, latency, and disorder on the IP network.

T.38 is a real-time fax mode based on the IP network. In this mode, the MG terminates the T.30 signals sent from the fax machine, and transmits the data to the peer MG in the T.38 mode. The peer MG then receives the T.38 packets and converts the packets into T.30 signals. The merit of the T.38 fax is that the fax packets have a redundancy processing mechanism and do not strictly rely on the quality of the network (the fax service can be processed even when a 20% packet loss occurs on the network). The demerit is that the DSP chip needs to participate in parsing the T.30 signals. Because there are various types of terminals on the network, the compatibility problem may arise.Figure 1 illustrates the principle of the T.38 fax.

What is MGCP?

Do you know what is MGCP, which is a voice feature on Huawei OLT, Huawei ONT,Huawei DSLAM product. Defined by IETF, MGCP is a protocol that specifies a call control mechanism in which call control and service bearing are separated. Call control is independent of the media gateway (MG) and is processed by the MGC. Therefore, MGCP is actually a master-slave protocol. The MG establishes various service connections under the control of the MGC.

MGCP provides the following commands:

NotificationRequest: The MGC sends this command to request the MG to detect a specified event, such as an offhook event or onhook event. After detecting such an event, the MG notifies the MGC. Through this command, the MGC can also instruct the MG to play signal tones, such as the dial tone and busy tone.

Notify: After the MG detects the specified event as instructed by the MGC, the MG sends this command to notify the MGC of the detected event.

CreateConnection: The MGC sends this command to instruct the MG to create a media connection. The command contains the instruction or suggestion on the bearing parameters and connection parameters.

ModifyConnection: The MGC sends this command to instruct the MG to modify the bearing parameters and connection parameters of an established media connection.

DeleteConnection: The MGC sends this command to instruct the MG to delete an established media connection. The MG can also voluntarily delete a connection. This means that, when the MG discovers that system resources are insufficient or the system is faulty, the MG can delete the connection and at the same time send this command to notify the MGC. Therefore, this command is bi-directionally available between the MGC and the MG.

AuditEndpoint and AuditConnection: The MGC sends the commands to check the status of a specified endpoint and connection.

RestartInProgress: The MG sends this command to notify the MGC that the MG or a certain endpoint managed by the MG is not available or is becoming available. This command is usually triggered by a system fault or restart.

MGCP also provides the following features:

Encoding in the text format

Adopting the Session Description Protocol (SDP) to describe the connection parameters of the media stream

Introducing the concept of event package

Adopting the wildcard to describe endpoints and events

What is SIP of voice feature on Huawei access products? Session Initiation Protocol (SIP), defined in RFC 3261, is used for setting up, modifying, and terminating sessions with one or more participants. The session can be a multimedia meeting, distance learning, or Internet telephony. SIP can be used for initiating sessions or inviting a member to join a session that has been set up otherwise. SIP transparently supports the mapping of names and the redirecting service, which facilitates the implementation of ISDN service, intelligent network, and personal mobile service. Once the session is set up, media streams are simply transmitted at the bearer layer through the Real-time Transport Protocol (RTP).

SIP is a text-based protocol put forth by IETF for IP phone/multimedia conferencing. It is a light-weight signaling protocol and has the following features:

Minimum status: One conference call or phone call can contain one or multiple requests or transactions. The proxy server can work in the stateless mode.

Irrelevance with lower layer protocols: SIP has minimum assumption of the lower layer protocols. The lower layer protocols can provide reliable or unreliable services to the SIP protocol layer, which can be packet or byte stream services. On the Internet, the SIP protocol layer can use the UDP or TCP protocol, and UDP is preferred. When UDP is not available, TCP is used.

Text-based: SIP adopts the text-based UTF-8 coding format and uses the ISO 10646 character set, which makes it easy to realize programming languages such as Java. This feature brings about merits such as easy commissioning, flexibility, and extensibility. The length of message, however, may also increase. For this reason, the message format is particularly designed so that the SIP messages are easy to parse.

Robustness: The robustness of SIP is demonstrated in several facets. For example, the proxy server does not need to maintain the call status, subsequent requests and re-transmission can adopt different routes, and the response message is transmitted in the self-routing mode.

Extensibility: The extensibility of SIP is demonstrated in several ways. Unidentifiable header fields can be ignored, the user can specify the message content that the SIP server must understand, new header fields can be introduced easily, and status codes are encoded in the layered coding mode.

Readiness to support IN services: Working with the end system, SIP and other call control extended protocols can support most services in Capability Set 1 and Capability Set 2 of ITU-T.

Position of the SIP Protocol on the Network

The SIP protocol is a signaling control protocol at the application layer. In the five-layer TCP/IP model, SIP is an application layer protocol. In the seven-layer OSI model, SIP is a session layer protocol. Figure 1 shows the position of the SIP protocol on the network.

The SIP protocol is independent from transmission protocols but is carried on different transmission protocols, such as, UDP, TCP, TLS, and SCTP. SIP is always carried on the UDP for its efficient transmission.

Figure 1 Position of the SIP protocol on the network

The SIP protocol must work together with other protocols to complete multimedia calls. Figure 2 shows positions of the SIP and other protocols. Figure 2 Positions of the SIP and other protocols 

The SIP protocol works with the Real-time Transmit Protocol (RTP), Real-Time Transport Control Protocol (RTCP), domain name server (DNS), Resource ReServation Protocol (RSVP), and Session Description Protocol (SDP) to complete multimedia calls.

RTP: A protocol defined by RFC 3550 for transmitting E2E real-time data. It provides the following functions for a series of E2E real-time data transmission services: payload type identification, sequence number arranging, timestamp, and transmission monitoring.

RTCP: A protocol that controls transmission of real-time media streams.

RSVP: A protocol that preserves network resources.

SDP: A text-based application layer protocol for describing multimedia sessions.

Sigcomp: A mechanism defined by RFC 3320 and used by application layer protocols to compress messages before they are sent to the network.

Advantages of the SIP Protocol

SIP will revolutionize the mode of communication service provisioning and the users' habit of communication consumption. An innovating communication mode integrating video phone service, messaging, Web service, e-mail, synchronous browsing, and conference call will be introduced to the telecommunication industry. Adopting SIP as the control layer protocol has the following advantages:

Based on an open Internet standard, SIP has inherent benefits in the integration and interoperability of voice and data services. SIP can implement across-media and across-device call control, and supports various media formats. SIP also supports dynamic adding and deleting of media streams, which make it easier to support richer service features.

SIP is intelligently extensible to the service and terminal side such as Huawei HG8245A , reducing the network load and facilitating the provisioning of service.

SIP supports mobile functions at the application layer, including the dynamic registering mechanism, location management mechanism, and redirecting mechanism.

SIP supports features such as presence, fork, and subscription, which facilitates development of new services.

In inter-board link aggregation, one or more ports on two boards are added to a LAG. Inter-board link aggregation provides protection for boards configured with aggregated links. In single homing, only one upstream device is available, without differentiating between primary and slave devices.

Service Requirements

The bandwidth between the access device uplink ports and the upper-layer switch is required to increase in load sharing mode. In addition, link protection is required. If one link becomes faulty, the upstream bandwidth is decreased, and the Internet access rate is reduced. However, the Internet access service must not be interrupted, and the access rate reduction slightly degrades user experience.

Board-level protection must be implemented in the LAG. If one board becomes faulty, services are not interrupted.

Networking

Figure 1 shows an example of single-homed inter-board link aggregation: Two GIU boards on the Huawei access device are used for upstream transmission , and the access device is interconnected with another device. One LAG is configured on the two GIU boards of the access device, and the other LAG is configured on two boards of the interconnected device.

Figure 1 Networking of single-homed inter-board link aggregation

Prerequisite

Interconnected devices, hardware, and ports support LAGs.

The two aggregated ports do not have a static MAC address. To check whether an aggregated port has a static MAC address, run the display mac-address command.

Context

Figure 2 shows the flowchart for configuring a LAG.

Figure 2 Configuration flowchart

Data Plan

Table 1 lists data plan for configuring single-homed inter-board link aggregation.

Table 1 Data plan for configuring single-homed inter-board link aggregation

Procedure

(Mandatory) Create a LAG and select an aggregation type.

(Optional) Add a LAG member port. NOTE: When adding a port to or deleting a port from a LAG, if this port has connected to the peer device, run the shutdown(Ethernet) command to deactivate this port or disconnect the optical fiber from this port to prevent a link loop.

(Optional) Select a load bearing type.

(Optional) Set the system priority and port priority.

(Optional) Select a link revertive mode.

(Optional) Query LAG information.

Run the link-aggregation command to add multiple uplink Ethernet ports to the same LAG to protect ports and share load between the ports. The port with the smallest port ID is the master port.

Perform this step when the LAG bandwidth or link reliability is required to improve. To do so, run the link-aggregation add-member command to add an Ethernet port to a LAG.

This step is required only when the LAG works in LACP aggregation mode.

Configuring the maximum active links in a LAG implements traffic allocation in load non-sharing mode. For example, M+N links have been configured in a LAG. Then, run the link-aggregation max-link-number command to specify N active links. The remaining M links are standby ones. If an active link is interrupted, a standby link automatically changes to the active one.

This step is required only when the LAG works in LACP aggregation mode.

LACP system priority: Run the lacp priority system command to set the LACP system priority of the access device.

LACP port priority: LACP port priority must be used with the maximum number of links. If a port is required preferentially for carrying services, set its priority higher. Run the lacp priority port command to change the link priority so that the standby link and the active link can be switched over.

This step is required only when the LAG works in LACP aggregation mode. Run the lacp preempt command to set whether traffic is switched back to the original link when the original link recovers.

Run the display link-aggregation command to query the LAG information, including the master port, number of links, aggregation type (manual or LACP aggregation), and maximum number of links.

Result

The bandwidth between the access device uplink ports and the upper-layer Huawei switch is increased in load sharing mode. In addition, link protection is implemented. If one board in the LAG becomes faulty, services are not interrupted.

Example The following configurations are used as an example to configure single-homed inter-board link aggregation:

The access device transmits data upstream using two GIU boards.

Uplink ports 0/19/0 and 0/20/0 on the active and standby GIU boards, respectively, are added to an inter-board LAG.

Packets are forwarded to these ports based on source and destination MAC addresses.

The LAG works in LACP aggregation mode.

Configuration File

In intra-board link aggregation, two or more ports on a board are added to a LAG.

Service Requirements

The bandwidth between the access device uplink ports and the upper-layer switch is required to increase in load sharing mode. In addition, link protection is required. If one link becomes faulty, the upstream bandwidth is decreased, and the Internet access rate is reduced. However, the Internet access service must not be interrupted, and the access rate reduction slightly degrades user experience.

Port-level protection must be implemented in the LAG. If one uplink port becomes faulty, services are not interrupted. This configuration does not take effect if the board fails.

Networking

Figure 1 shows an example of intra-board link aggregation: A GIU on the access device is used for upstream transmission, and the access device is interconnected with another device. One LAG is configured on the GIU board of the access device, and the other LAG is configured on one board of the interconnected device.

Figure 1 Networking of intra-board link aggregation

Prerequisite

Interconnected devices, hardware, and ports support LAGs(n*FE/GE- Ethernet Service Board /10GE).

The two aggregated ports do not have a static MAC address. To check whether an aggregated port has a static MAC address, run the display mac-address command.

Context

Figure 2 shows the flowchart for configuring a LAG.

Figure 2 Configuration flowchart

Table 1 lists data plan for configuring intra-board link aggregation.

Table 1 Data plan for configuring intra-board link aggregation

Procedure

(Mandatory) Create a LAG and select an aggregation type.

(Optional) Add a LAG member port. NOTE: When adding a port to or deleting a port from a LAG, if this port has connected to the peer device, run the shutdown(Ethernet) command to deactivate this port or disconnect the optical fiber from this port to prevent a link loop.

(Optional) Select a load bearing type.

(Optional) Set the system priority and port priority.

(Optional) Select a link revertive mode.

(Optional) Query LAG information.

Run the link-aggregation command to add multiple uplink Ethernet ports to the same LAG to protect ports and share load between the ports. The port with the smallest port ID is the master port.

Perform this step when the LAG bandwidth or link reliability is required to improve. To do so, run the link-aggregation add-member command to add an Ethernet port to a LAG.

This step is required only when the LAG works in LACP aggregation mode.

Configuring the maximum active links in a LAG implements traffic allocation in load non-sharing mode. For example, M+N links have been configured in a LAG. Then, run the link-aggregation max-link-number command to specify N active links. The remaining M links are standby ones. If an active link is interrupted, a standby link automatically changes to the active one.

This step is required only when the LAG works in LACP aggregation mode.

LACP system priority: If the access device like Huawei MA5608T is dual homed to two convergence devices, the access device determines the primary and standby LAGs. Run the lacp priority system command to set the LACP system priority of the access device to be higher than that of the peer device.

LACP port priority: LACP port priority must be used with the maximum number of links. If a port is required preferentially for carrying services, set its priority higher. Run the lacp priority port command to change the link priority so that the standby link and the active link can be switched over.

This step is required only when the LAG works in LACP aggregation mode. Run the lacp preempt command to set whether traffic is switched back to the original link when the original link recovers.

Run the display link-aggregation command to query the LAG information, including the master port, number of links, aggregation type (manual or LACP aggregation), and maximum number of links.

Result

The bandwidth between the access device uplink ports and the upper-layer switch is increased in load sharing mode. In addition, link protection is implemented. If one port in the LAG becomes faulty, services are not interrupted.

Example The following configurations are used as an example to configure intra-board link aggregation:

The access device transmits data upstream using a GIU board.

Uplink ports 0/19/0 and 0/19/1 on the GIU board are added to an intra-board LAG.

Packets are forwarded to these ports based on source MAC addresses.

The LAG works in LACP aggregation mode.

Configuration File

LAG Setup Process on Huawei Access Network 

The process of setting up an LACP LAG is as follows:

Interconnected devices send LACPDUs to each other.

Interconnected devices determine the actor (indicating the party that is dominant in negotiation) based on LACP system priorities and system IDs.

Interconnected devices determine active ports (ports carrying traffic) based on the port LACP priorities and port IDs of the Actor.

Exchanging LACPDUs Between Interconnected Devices

An LACP LAG is set up on interconnected devices, and member ports are added to the LAG. The member ports are enabled with LACP, enabling interconnected devices to exchange LACPDUs to each other, as shown in Figure 1.

Figure 1 Exchanging LACPDUs between interconnected devices of an LACP LAG

Determining the Actor

Interconnected devices receive LACPDUs from each other. Use the LAN switch shown in Figure 2 as an example to describe the process. After receiving LACPDUs from the access node, the LAN switch such as Huawei S3700-28TP-SI-AC checks and records information about the access node and compares system priorities. If the system priority of the access node is higher than that of the LAN switch, the access node acts as the actor.

If the system priority of the access node is the same as that of the LAN switch, the party with a smaller MAC address in the system ID functions as the actor.

Figure 2 Determining the actor

Selecting Active Ports

As shown in Figure 3, interconnected devices select active ports based on the priorities of ports on the actor after the actor is selected.

After the same active port is selected, selected links in the LAG are determined, and traffic is distributed in load sharing across the links.

Figure 3 Selecting active ports

LAG Link Switchover and Reversion

Switching Condition 

In an LACP LAG, a link switchover is triggered if a device at one end detects one of the following events:

LACP detects a link failure.

Negotiated LACP status is changed.

An active port becomes unavailable.

The board is faulty.

Switchover Process

When any of the preceding trigger conditions is met, the link switchover is performed in one of the following processes:

For load non-sharing LAGs:

The faulty link is disabled.

The standby link with the highest priority is selected to replace the faulty selected link.

The standby link with the highest priority becomes the selected link and then forwards data.

For load sharing LAGs:

The faulty link is disabled.

Traffic is reallocated to member links using the load sharing algorithm.

Switchover Reversion Process 

If an LACP LAG works in load sharing mode and the active port before switchover recovers, the original faulty link is enabled, and traffic is reallocated using the load sharing algorithm.

If an LACP LAG works in load non-sharing mode, you can set the link revertive mode to revertive or non-revertive.

When an LACP load non-sharing LAG is set to revertive, services are switched back to the link of a higher priority after this link recovers, as shown in Figure 1.

When an LACP load non-sharing LAG is set to non-revertive, services are not switched back to the selected link after this link recovers, but are still transmitted on the current link.

As shown in Figure 1, ports 1 and 2 are member ports in an LACP LAG, port 1 is the active port (working in the selected link), and port 2 is the inactive port (working in the standby link).

After port 1 fails, port 2 takes over as the active port.

After port 1 recovers, port 1 works as an active port, and port 2 still works as an inactive port.

Figure 1 Reversing a switchover (LACP LAG)

Local Tone 

About basic conception of Local Tone in voice service, which is the functions of Huawei access equipment, Huawei OLT(upstream service board, GPON service board), Huawei DSLAM, here providing a specific explanation, hope can be helpful.

The softswitch or AG can play tones to terminals. Tones played by the AG are called local tones. Before enabling the AG to play local tones, load a tone file to the AG. For parameter tones, configure them on the AG.

Tone File Signal tone standards vary depending on country requirements. As shown in Figure 1, customize a tone file, load the tone file to the AG, and the AG plays the customized tones, such as dial tone, ring back tone, and busy tone during calls.

Figure 1 Tone played by the AG using the tone file

Voice files, generally named voice.efs, are stored in the flash memory of a control board. A voice file describes the tone playing types supported by the DSP. The description covers the signal tone type, frequency, duration, and level. After the system is initialized, tone playing parameters are configured on the DSP. When the system requests to play a tone for a subscriber, the DSP reads the configuration and generates the desired signal tone on a real-time basis.

The signal tones specified in a voice file can be parameter tones, waveform tones, or announcements.

Parameter tones are simple tones, including dialing tones, busy tones, and ring back tones. The system issues parameter tone attributes, such as the frequency, energy, duration, and cadence, to the DSP. The DSP then generates parameter tones.

Waveform tones are simple tones. The system records these tones into a voice file and stores the file in the DSP. When the system needs to play a waveform tone for a subscriber, the DSP plays the recorded voice data for the subscriber.

Announcements are audio messages played to subscribers, such as "The subscriber you dialed is busy. Please call later." The system records announcements into a voice file and stores the file in the DSP. When the system needs to play an announcement for a subscriber, the logic or DSP plays the recorded voice data for the subscriber.

In addition to voice file recording, the system supports parameter tone customization. A customized parameter tone takes effect in the next tone playing, which does not require service board resetting.

Customized Parameter Tones

Users can customize a parameter tone by specifying a series of cadences and tone playing rules. Each cadence defines the signal frequency, energy, and duration.

In a parameter tone, cadences can be played in one of the following ways:

Sequential play: Each cadence is played once in sequence according to the break-make ratio.

Continuous play: All cadences are cyclically and continuously played according to the break-make ratio.

Cyclic play: All cadences are cyclically played and the number of cycles can be defined.

A cadence contains the following items:

Frequency: A cadence can contain only one frequency, two frequencies, three frequencies, or four frequencies.

Energy: The energy obtained when the POTS gain is 0 dBm and the line length is 0 km is used in this section.

Play mode: A cadence can be played only once or repeatedly played according to the break-make ratio. If the cadence is configured to repeatedly play, the number of plays is not limited.

Figure 2 shows ring back tone, reject response tone, and special dial tone diagrams.

A ring back tone is continuously played with a single cadence. The cadence contains one frequency, which is 425 Hz. The energy is -10 dBm. The cadence is played according to the break-make ratio. That is, the cadence is played for 1250 ms and stopped for 3750 ms.

The reject response tone is cyclically played with a single cadence. The number of cycles is 3. The cadence contains two frequencies, which are 300 Hz and 450 Hz, respectively. The energy is both -20 dBm. The cadence is played according to the break-make ratio. That is, the cadence is played for 200 ms and stopped for 200 ms.

The special dial tone is continuously played with two cadences, cadence 1 and cadence 2.

Cadence 1 contains one frequency, which is 420 Hz. The energy is -10 dBm. The cadence is played only once and the duration is 1000 ms. Cadence 2 contains two frequencies, which are 380 Hz and 420 Hz, respectively. The energy is both -10 dBm. The cadence is played only once and the duration is 1000 ms. Figure 2 Example of generating parameter tones 

Feature Dependencies and Limitations

Users only can customize parameter tones. If the AG is required to play a waveform tone or announcement, a voice file is required.

If a parameter tone is both customized and configured using a voice file, the customized configuration preferentially takes effect.

Accounting

This section describes accounting on pay phones using coins or IC cards.

Introduction The AG supports three types of accounting: polarity reversal accounting, 12/16KC accounting, and polarity reversal pulse accounting.

Polarity reversal accounting: The pay phone starts to account immediately after detecting a voltage reversal between A and B wires.

12/16KC accounting: It is also called KC accounting. The pay phone performs accounting based on 12 kHz or 16 kHz high frequency signals sent by the POTS board.

Polarity-reversal pulse accounting: The pay phone performs accounting based on standard pulse signals sent by the POTS board.

Polarity Reversal Accounting Polarity reversal refers to the voltage reversal between wires. For example, if the voltage between A and B wires is Vtp, the reversed voltage is -Vtp. After detecting a polarity reversal signal, the pay phone starts to account. The pay phone stops accounting when the voltage between A and B wires changes from -Vtp to Vtp, as shown in  Figure 1. The polarity reversal is classified into hard polarity reversal and soft polarity reversal.

Hard polarity reversal: also called quick polarity reversal. Specifically, the voltage is reversed in a short period of time, shorter than 3 ms in general. The hard polarity reversal brings great interference on lines.

Soft polarity reversal: also called slow polarity reversal. Specifically, the voltage is reversed in a long period of time, longer than 80 ms in general. The soft polarity reversal brings small interference on lines.

Figure 1 Hard and soft polarity reversal 

12/16KC Accounting

When detecting that both the calling and called parties enter a session, the softswitch requests the AG to send 12 kHz or 16 kHz high frequency pulse signals to the pay phone. The pay phone performs the accounting once after detecting such a pulse signal. For example, if the carrier charges subscriber 1 cent for one-minute call, the softswitch sends such a pulse signal at the interval of one minute. If the pay phone receives three such pulse signals after the call ends, the subscriber is charged 3 cents. After detecting that the call ends, the softswitch requests the AG to stop sending such pulse signals to the AG.

Polarity Reversal Pulse Accounting The polarity reversal pulse accounting rule is the same as the polarity reversal accounting rule. The difference lies in that the POTS board(Huawei XDSL Service Board)  sends standard pulse signals to pay phones. As shown in Figure 2, the pulse width is Tw and there are a total of two pulse signals.

Figure 2 Polarity reversal pulse accounting

VAG

VAG The virtual access gateway (VAG) is a solution in which one AG device can be simulated into multiple AG devices. Different VAGs can connect to different IMSs/softswitches,as shown in Figure 1.

The purposes of the VAG are as follows:

Differentiated service

Virtual operation

Sharing load on the IMS/softswitch

Simulate multiple logical AGs on a physical AG, and enable different logical AGs to connect to different customer groups. In this way, different convergence rates are provided for different customer groups to meet the identified service requirements.

In the AG networking application, certain small-scale operators do not buy independent AG devices such as Huawei MA5683T , but rent the AG devices from other operators to provide the service. For the renters, they may rent out a device to multiple operators to improve the device utilization rate. Simulate multiple logical AGs on a physical AG, and enable different logical AGs to connect to different softswitches. In this way, the wholesale service requirements can be met.

One IMS/softswitch has limited capacity. If subscribers connected to different VAGs are distributed to multiple IMS/softswitches, the workload on one IMS/softswitch becomes lighter.

Figure 1 VAG networking

The VAG call process and the traditional call process are similar. The main differences are as follows:

The DSP distributes resources. After a subscriber picks up the phone, the AG applies for DSP resources. If DSP resources are assigned to the VAG, the VAG obtains DSP resources from the shared resource pool. If DSP resources are not assigned to the VAG, the VAG obtains DSP resources from the exclusive resource pool.

The IP addresses in the IP packets during the connections and conversations of the subscribers under the same VAG are the media IP address corresponding to the VAG.

Each VAG can be configured and managed independently. For the softswitch, each VAG is an independent AG, and can be configured with related attributes separately, such as authentication, ringing mapping, and terminal layered mode. When configuring a subscriber, specify a VAG for the subscriber. Subscribers under different VAGs can share the same terminal ID. The total number of subscribers configured on all VAGs cannot exceed the maximum number configured in the system.

To configure a VAG, add a VAG interface (MG interface), and enter the VAG interface mode to configure interface parameters by running commands. Each VAG can be configured with independent signaling IP address and media stream IP address (the signaling IP address and media stream IP address in the same VAG can be different) in different VLANs.

Local Digitmap

A digitmap is a dialing scheme configured on an AG. The AG collects digits dialed by calling parties based on the digitmap. With the digitmap, the AG reports a group of digits each time, reducing the number of signaling exchanges between the AG and the softswitch, thereby improving efficiency. The digitmap configured on the AG is called local digitmap. An H.248-compliant AG can use the local digitmap or digitmap issued by the softswitch. A SIP-compliant AG must use the local digitmap

Definition

After a subscriber dials a number, the AG matches digits of the number. If the matching is successful, the AG sends the collected digits to the softswitch. The softswitch analyzes the number and corresponding services. If the matching fails, the AG discards the digits.

On the NGN, terminals send called numbers to the softswitch on the control layer through the AG on the access layer. Numbers can be transmitted in in-band or out-band mode.

In-band transmission occupies IP media stream resources. In this mode, the AG transfers numbers in RTP or RFC 2833 format to the upper layer network. Such processing is not related to digitmaps.

In out-band transmission mode, the AG transfers numbers through signaling streams. The out-band transmission has the following two modes:

Reporting a number digit by digit: The AG reports a digit to the softswitch after a subscriber dials a digit. The number of signaling exchanges between the AG and softswitch is determined by the number of digits contained in a dialed number. Such processing is not related to digitmaps.

Reporting a complete number: The AG reports a complete number dialed by the subscriber to the softswitch using one signaling message. This type of processing lowers the burden on the softswitch and is preferentially selected. To report a complete number, the AG must know when to collect digits and when to report digits. The digitmap determines whether dialed digits are valid.

Digitmap Example

The following digitmap is used as an example:

[2-8]xxxxxxx | 13xxxxxxxxx | 0xxxxxxxxx | 9xxxx | 1[0124-9]x | * | # | x.# | [0-9*#].T

This digitmap consists of nine character strings.

The first character string "[2-8]xxxxxxx" indicates that the matched number must contain eight digits, the first digit must be one of 2 to 8, and remaining seven digits can be any of 0 to 9.

The second character string "13xxxxxxxxx" indicates that the matched number must contain 11 digits, the first digit must be 1, the second digit must be 3, and remaining nine digits can be any of 0 to 9.

The third character string "0xxxxxxxxx" indicates that the matched number must start with 0 and contain 10 digits.

The fourth character string "9xxxx" indicates that the matched number must start with 9 and contain 5 digits.

The fifth character string "1[0124-9]x" indicates that the first digit must be 1, the second digit cannot be 3, and the third digit can be any of 0 to 9.

The sixth character asterisk (*) indicates that the matched character must be an asterisk (*).

The seventh character pound sign (#) indicates that the matched character must be a pound sign (#).

The eighth character string "x.#" indicates that the AG matches one of 0 to 9 for multiple times or even does not match any digits and stops the matching until detecting a pound sign (#).

The ninth character string "[0-9*#].T" indicates that the AG starts a timer, matches digits 0 to 9 for multiple times or even does not match any digits, asterisk (*), or pound sign (#), and stops the matching only after the timer expires.