Chapter 4 - Overview of Signal Units

Chapter 4 Overview of Signal Units

		Overview of the Signal Units

		SS7 is a packet-switching network and uses data packets just like X.25 and other packet-switching technologies. A packet contains all the information necessary to route data through a network, without establishing a connection to the destination.

		There are three basic methods of switching in a network: circuit switching, message switching, and packet switching. Circuit switching uses a physical connection between two entities for transmitting a data stream. The circuit remains connected until both entities have completed transmission. A good example of circuit-switching networks is the Public Switched Telephone Network (PSTN).

		Message switching came about in the 1970s and 1980s and uses a message structure to route data through a network. The data is accompanied by an address and a message (which serves as an instruction to the receiver). The data is sent in its entirety and does not include any error-checking schemes or flow control.

		Packet switching arranges the data into a packet or a group of packets and transmits it in the form of a complete packet, providing all the information needed to route and process the received data. Included in packet-switching networks are network management and error detection/correction.

		Packet switching is a more efficient way of networking, and it makes better use of facilities. In the event of network failures or any other problems in the network, the packet protocol can dynamically change the routing for a particular destination and can provide higher reliability through error checking and correction.

		Usually, the packet-switching networks use different types of packets depending on the function. For example, if sending a data packet in an X.25 network, the packet type is called an information frame. The information frame has a distinct format and provides parameters specific to the transfer of data through the network.

		If a packet is received in error and the packet must be retransmitted, a supervisory frame is used to inform the originator of the data packet that the data was in error and it needs to retransmit the errored packet.

		The supervisory frame contains the parameters needed to inform another node of an error, but it does not support the transmission of any data. This packet then serves a very unique purpose and cannot be used for anything else.

		SS7 uses three different structures of packets, called signal units. These signal units provide three different levels of service in the SS7 network. The SS7 protocol uses all three signal units for transmission of network management information, depending on the level of management. Information is sent using only one type of signal unit.

		Another unique factor about SS7 networks is the source of information. In most networks, we are sending data from a user to another user. In SS7 networks, the user is the telephone network. The information is control and signaling information from telephone company switches and computers, which must be shared from one device to another. This makes the SS7 network a machine-to-machine network, rather than a user-to-user network. There is very little human intervention in this network, because most of the procedures and processes are automated and do not require any operator control.

		A signal unit is nothing more than a packet, but in SS7, there are many applications, requiring different packet structures and capabilities. The applications found in the SS7 network vary from standard networks. There are circuit-related applications and non-circuit related applications. These are the two basic foundations used to identify the functions within the network.

		Circuit-related applications are directly related to the connection and disconnection of telephone circuits used to connect telephone subscribers. These circuits can be analog voice trunks or digital data circuits. They are located in a separate network outside of the signaling network and are used for the sole purpose of connecting telephone subscribers to other telephone subscribers.

		The SS7 network does not have anything to do with the voice and data in these circuits other than identifying the type of data and voice transmission to take place (for example, data rates and encoding methods used at the voice interfaces).

		Non-circuit related applications consist of all other traffic in the SS7 network. To support these circuit-related functions of the PSTN, telephone switches must be able to communicate with one another. Whether they are requesting information from a database stored in a central computer system or invoking a feature in a remote telephone switch, there needs to be a protocol for all other aspects of the telephone network not related to a specific circuit.

		Network management information is another type of communication which must be supported within the network. This is the automated part of the network, which allows signaling points within the network to automatically recover from failures and signaling point outages. Network management is completely autonomous in SS7 networks.

		All signal units rely on the services of the Message Transfer Part (MTP) for routing, network management and link management, and basic error detection and correction. Without MTP, the signal units are worthless. MTP is a lower level protocol and is found in all signal units.

		Anytime information is to be transferred through the network, from one signaling point to another, the Message Signal Unit (MSU) is used. It is called the "message" signal unit because SS7, like many protocols, uses data messages to convey information to another entity in the network. Information is considered control information or network management information.

		The MSU provides the fields of the MTP protocol, as well as an additional two fields: the service indicator octet (SIO) and the service information field (SIF).

		The SIO is used by level three to identify the type of protocol used at level four (i.e., ISUP or TCAP) and the type of standard. A standard can be a national standard or an international standard. If the protocol at level four is based on the ITU-TS standard, then the SIO field would indicate this as an international standard. If the protocol is any other type of protocol (such as ANSI), the SIO field would indicate this as a national protocol.

		This information is used by level three and the message discrimination function to determine the type of signal unit, the protocol, and how it should be decoded. There are also spare bits in the SIO field which can be used for priorities in the ANSI standard, or for other functions in private networks.

		The SIF is used to transfer control information, as well as the routing label used by level three. This field can contain up to 272 octets and is used by network management, ISUP, TCAP, MAP, and any other protocols which may be developed over time.

		Not all information in this field is considered level-four information. For example, in the case of network management information, it is level three. The same is true if SCCP is used to transport TCAP. The SCCP portion of the message is found in the SIF field of the MSU. SCCP is considered to be level-three information.

		Link status information is carried using another signal unit type called the Link Status Signal Unit (LSSU). The LSSU is actually used by level three at one node to transmit status regarding the link on which it is being carried to its adjacent node. The LSSU is never used to carry link status messages through the network. It is only used to communicate this status between two adjacent signaling points.

		When no traffic is being sent and the network is idle, the Fill-In Signal Unit (FISU) is sent to provide constant error checking on the link. This allows the SS7 network to maintain its high reliability, because even though no information is being sent, the signaling points can still perform error detection on the FISU to determine if the link is beginning to deteriorate.

		In addition to the FISU transmission, the MTP protocol is constantly monitoring the status of the link. The MTP is used in all three signal units. These signal units are explained in more detail in the following section.

		Fill-in Signal Unit (FISU)

		The lowest level signal unit that is, one that provides the lowest level of service is the FISU. The FISU acts as a flag in TDM-based SS7 networks. When there is no payload to be delivered and the network is idle, FISUs are sent (see Figure 4.1).

		This is different than in any other network, where flags are transmitted. A flag is usually a one-byte pattern, consisting of a 0, six 1s, and followed by a 0 (01111110). This one-byte pattern is used to maintain clock synchronization in many asynchronous networks. There is no "intelligence" in this type of pattern, however, so it serves no other purpose.

		In the event that a data link begins to degrade, there is no indication that the link can carry traffic any longer until a transmission is attempted. By this time, it is too late. The transmission will fail and the link will have to go through diagnostics.

		Figure 4.1 The FISU consists of the components necessary for routing (per level-three MTP) and is sent during idle periods, instead of flags. By sending the FISU, a signaling point can verify the integrity of a link by checking the FCS field for errors.

 

In the TDM SS7 networks, in order to maintain a high level of reliability, the FISU is used. This signal unit does not provide any information, but it does contain a minimal amount of information. The sequence numbers, for instance, can be used to acknowledge a previous signal unit.

		The most significant field in the FISU is the Frame Check Sequence (FCS) field. This field is used by level three to determine if there are any errors in the FISU. This is based on the bits in all the fields of the FISU. The FCS is found in all signal units and is used to transport the remainder of the CRC-16 equation performed by the transmitting signaling point.

		The CRC-16 is the error-checking mechanism implemented by all transmitting signaling points when transmitting a signal unit. The purpose is to provide the remainder of the CRC equation to the receiver of a signal unit. The receiver then uses the same CRC equation and compares it to the value received.

		By using this field for error checking in the FISU, the MTP can constantly evaluate the status of any link, even during periods of idle traffic. In the event that a link has degraded to a point where it is causing too many errors, the link can be taken out of service by the MTP link management function before it is needed for actual traffic.

		The FISU can also be used to acknowledge a previously received signal unit. This is done by sending an FISU with a backward sequence number equal to the forward sequence number of the signal unit being acknowledged. In other words, the backward sequence number identifies the sequence number of the last good signal unit received.

		In the event that a signal unit is received and rejected by the MTP at level two, the FISU can be used to send back a negative acknowledgment. A negative acknowledgment requires the use of a backward indicator bit (BIB). Usually, the BIB and the forward indicator bit (FIB) are of the same value. However, when there is a negative acknowledgment, the BIB is toggled, assuming an opposite value from the FIB. This signifies a retransmission request.

		The receiver of a FISU or any other signal unit with opposite values in the indicator bits examines the backward sequence number (BSN) to determine which signal units need to be retransmitted. This is more efficient than using specialized supervisory frames, as other protocols use.

		When there are no errors, the indicator bits maintain the same value; that is, both the forward indicator bit (FIB) and the backward indicator bit (BIB) are exactly the same. When a retransmission occurs, the retransmitted signal unit is sent with the indicator bits set to the same value.

 

When the retransmission is acknowledged, the originator of the retransmission request toggles the FIB to match the BIB, and maintains this value until another retransmission is required. This procedure is explained in full detail in Chapter 5.

		The FISU is never retransmitted if it is found in error. In fact, when FISUs are sent through the network, the sequence numbers do not increment (forward sequence numbers [FSNs]). There is no reason to retransmit these signal units because they do not provide any information. They are used only to monitor the integrity of a signaling link.

		The FSN assumes the value of the last MSU sent by the transmitting signaling point, and stays at the same value until another MSU or LSSU is transmitted. The BSN follows the same procedure, unless used to acknowledge a previously received MSU.

		The length indicator field in the FISU is always set to zero. The length indicator identifies the type of signal unit being received. The length is that of the information field, which does not exist in the FISU. The total length of an FISU is static, at 48 bits long.

		Although many drawings and publications will show both an opening and closing flag, there is only one opening flag and no closing flag. The opening flag of one signal unit is the closing flag of the previous signal unit. This is defined in Telcordia TR-NWT-000246.

		In TCP/IP networks, the FISU is not used. A "keep-alive" signal is still needed, however, to make sure all entities on the network are still in service. The TCP/IP protocol does report when nodes fail; however, it may take several seconds before a nodes status gets reported. This is not acceptable for SS7, so M2UA protocol provides a mechanism for constantly checking nodes on the network much as the FISU does.

		Link Status Signal Unit (LSSU)

		The LSSU is sent between two signaling points to indicate the status of the signaling link on which it is carried. Therefore, the LSSU is only of significance between two signaling points and does not get broadcast through the network (see Figure 4.2).

		Figure 4.2 The LSSU is used by level-three MTP to send link status information to an adjacent signaling point. The LSSU is not broadcast to any other signaling points.

 

When a link is determined to have failed, the signaling point that detects the error condition is responsible for alerting its adjacent signaling point that the link is no longer available. The types of error conditions that warrant this procedure are alignment problems.

		Alignment of a link means that all signal units received are of the correct length, and there are no ones-density violations. A ones-density violation occurs when the bit stream has more than five consecutive 1s, considered by the protocol as a flag. With link management, and level-two functionality, this should never occur. Yet when it does, the link must be taken out of service and realigned.

		Realignment is the procedure used by level two and level three to correct a link problem. The real problem is usually within a processor at either end of the link. Therefore, the processor that is at fault must be corrected. The first step is to remove all traffic from this link. The LSSU is sent to the adjacent signaling point to inform the adjacent node that all traffic should be removed from the link and the link is being realigned. No acknowledgment is required; this is simply an information signal unit.

		There are two reasons that this procedure is necessary. To begin with, the two signaling points run independently of one another. Each end of a signaling link has its own processor. This processor and its accompanying software are what provide the functionality of levels two and three. In the event the processor should fail or be unable to process any more MSUs, only that processor would be aware of the trouble. The adjacent processor thinks that the link is working fine and is capable of carrying traffic. It is for this reason that the LSSU is necessary.

		When level two determines there is a problem (as notified by level three), it will transmit an LSSU identifying the problem. The status indicators in the status field of the signal unit identify the specific status of the link; i.e., the link is in alignment or there is a processor outage at the originating node.

		The fact that the link is capable of sending this LSSU indicates the ability to send lower level traffic, despite the inability to process upper level traffic. Depending on the status of the link, the receiving signaling point may send a network management message in the backward direction (to its adjacent signaling points), indicating the inability to reach a particular signaling point. This would occur only if the route to a certain destination became inaccessible because of the link failures.

		This means that the LSSU works in conjunction with other network management functions. The link management function uses the LSSU to notify adjacent signaling points of link status, while the route management function uses the MSU to notify adjacent signaling points of problems with a route to a destination.

 

The LSSU consists of the same components as the FISU, with the addition of the status field (SF). The status field carries the link status information for the link on which it is carried. The LSSU is not transmitted on parallel links and does not carry information about other links. The status field indicates the status of the link on which the LSSU is carried.

		Again, this implies that the link did not have a hard failure. A hard failure is one in which no traffic can be carried by the link, such as in the case of a backhoe digging up a facility with SS7 links. The LSSU relies on level two and level three still being functional on the link.

		As with the FISU, when an error occurs within the signal unit, the LSSU is not retransmitted. An errored LSSU is discarded, and the error is counted as an error on the link.

		The value of the length indicator in an LSSU is either a 1 or a 2. Currently, the LSSU status field is always one octet in length. Until further definitions are made for additional status indications, this rule will probably not change.

		Message Signal Unit (MSU)

		The MSU as shown in Figure 4.3 provides the structure for transmitting all other protocol types. This includes the ISDN User Part (ISUP), the Transaction Capabilities Application Part (TCAP), and the Mobile Application Part (MAP). The difference between the MSU and the previous two signal units is the addition of the Service Indicator Octet (SIO) and the service information field (SIF).

		The SIO is used by level-three message discrimination to determine the type of protocol being presented in the MSU. This allows message discrimination to identify who the user will be at level four. For example, if the SIO indicates the protocol to be ISUP, then ISUP will be the user at level four.

		The SIO also identifies the version of protocol being presented: international or national. International applies only to the ITU-TS standard compliant protocols. This is used when connecting to the international plane of the SS7 network. The national protocol applies to all other standards, including the ANSI standard used in the U.S. It should be noted, however, that national does not imply ANSI. There are many national standards used throughout the world. For example, in Germany, the national standard would be 1TR7, in Hong Kong it would be the Hong Kong standard, and in New Zealand it would be the New Zealand standard.

		Figure 4.3 The MSU is used to deliver level-four information to its destination. Level-four information is found in the variable Service Information Field (SIF).

 

		The use of two planes in the network allows all nations to internetwork on the international level, using a gateway signaling point (usually a Signal Transfer Point (STP)) to gain access into the international network from the national side, and vice versa.

		Individual countries can then use their own individual flavors of the SS7 protocols, depending on the requirements of their own unique networks, without impacting the entire SS7 network worldwide.

		All nations must comply with the ITU-TS standards at the international plane and use some method of interworking between the two planes. Interworking will almost always require protocol conversion. The SIO can be used to determine when this will be necessary.

		Another reason this parameter is important is because of the difference between the point codes used between international and national. International point codes are formatted as a three-bit zone identification, an eight-bit area or network identification, and a three-bit signaling point identification. National point codes can be any variation, as long as the total field length remains the same. The ANSI standard is an exception to this rule, where the point code is a 24-bit point code, eight bits for a network identification, eight bits for a cluster identification, and eight bits for a member identification.

		The MSU provides a signaling information field with a capacity of up to 272 octets of user data. In the case of SS7, user data consists of any data from an upper layer (such as ISUP or TCAP). The SIF does not necessarily have to be used for level-four information. Network management is also a user of the signaling information field. Network management is a function of level three.

		The length indicator of the MSU can be any value over 2 and up to 64. The length indicator is a six-bit field, which limits it to the range it can represent. Yet the SIF can be up to 272 octets long. In TCP/IP-based signaling networks, this limitation no longer applies.

		When the SIF exceeds 64 octets in length, the length indicator of the MSU remains at 63. This is not an issue with the protocol, because the only purpose of this field is to allow level-three message discrimination to be able to determine the type of signal unit being received. There is no other use for this field.

 

Based on this fact, it is safe to say that any value in the length indicator over 2 is always an MSU, and any value over 2 is really insignificant. There is no reason to expand this field, because the exact length is not important to level three.

		Primitives

		In order for the various levels to interface with one another, some method of standard interfacing must be implemented. The use of primitives is not unique to SS7, although the particular primitive types used in this protocol are unique.

		Communications between levels two and three and between levels three and four are all software controlled. We do not see any communications over the network, although we will see the results over the network. A primitive is the method used by software to pass information to the next level, in either direction.

		The important thing to understand is that a primitive is pure software. There is nothing for us to see or examine, unless we are looking at the source code of a signaling point itself. Primitives are discussed here for those who are actively writing software for SS7 network products and need to understand the full picture of what is taking place.

		As seen in Figure 4.4, the primitive provides four fields. The first field, marked by the ''X," indicates the originator of the primitive. If the MTP is passing information up to the ISUP, then the first field would indicate "MTP."

		The next field is the generic name. The generic name identifies the type of information being provided. For example, if information regarding the address of the originator (such as the calling party address) is being sent from ISUP to MTP, the generic name would be "unitdata." The generic name will differ, depending on the level. For example, the Signaling Connection Control Part (SCCP) will have different generic names than, say, the ISUP. The functionality remains the same.

		Figure 4.4 Primitives are used to communicate with the various levels of the protocol stack within a network entity. Primitives are not seen in the network, but reside in software at each signaling point. This figure depicts the structure of a primitive.

 

The field after the generic name is the specific name. The specific name describes the action that is to take place. The specific name can be any one of the following:

		Request

		Indication

		Response

		Confirmation

		A request is used to invoke some type of service from another level. For example, in the case of network management, there may be the need to start a procedure. The request would be used to invoke that procedure at a higher level.

		An indication is used to inform the requesting level that the requested service has been invoked. This is like an acknowledgment between levels. Using the preceding case, when a user part invokes a management procedure, it will send an indication to MTP to inform MTP of the invocation.

		A response is sent to complete a particular transaction between a service element and a user. A user, in this sense, is the protocol, while the service element is something such as SCCP or the upper user parts. The response is used only when a service has been previously invoked and an indication has been sent.

		The confirmation is sent to inform the user part that a connection has been established or a requested service has been invoked. Confirmation, in some aspects, is similar to an acknowledgment.

		There are many procedures that surround the various primitives, depending on the level they are communicating with and which user part they are interfacing to. The purpose of the primitive, once again, is to provide a means of communication between the various protocol levels within a signaling point.

		Overview of SS7 Protocols

		As we have discussed in the previous chapters, the SS7 network uses many different protocols. Each protocol is used for a specific purpose and provides the necessary functionality to accomplish specific tasks.

		In this section, we will look at the various protocols used in SS7 networks, and discuss their usage and applications. This section will only provide an overview of the various protocols. For a more specific explanation of these protocols, refer to their respective individual chapters.

 

The SS7 network provides some basic services to the PSTN. The impetus behind deploying this network was to remove all signaling information away from the voice network. In the early days of common channel signaling, this certainly seemed enough to justify the usage of another network. However, the SS7 network slowly evolved into much more than just a signaling network. It has also evolved into a control network.

		The word "control" implies a lot of different things. From the voice network perspective, control refers directly to the ability to control features and tasks in a remote telephone switch or centralized computer. The user of this remote control capability is usually another telephone switch or another computer system.

		This network obviously forms the basis for the Intelligent Network (IN). Without the mechanisms for supporting remote control or other network entities, the IN would not be possible.

		In today's convergent network, call control has evolved. Next generation switches are computer platforms controlled by software. The media gateway (MG) provides access to the subscriber and transforms analog voice into digitized voice for transmission over the TCP/IP network (or ATM network in some cases). Call control is distributed throughout the network to lower the cost of these network switches. The call control is now more centralized in media gateway controllers (MGCs), eliminating the need for expensive processing resources in the MGs. A control protocol such as Megaco is used to communicate call control information between the MG and the MGC.

		However, the initial purpose of signaling cannot be ignored either, especially since today this remains as the principal function of this network. As the IN evolves, this will gradually change.

		As we discussed in the first chapter, the SS7 network evolved from the earlier CCS6 network, which was more limited, yet of similar technology. The primary difference between the two technologies is in the protocols used and their structures. CCS6 used a very stringent structure, with fixed-length signal units. This did not allow for variable-length signal units, and limited the protocol as far as the type of information that could be provided.

		It was for this reason that SS7 was structured the way it is today. By providing a basic structure, which various protocols can depend on for transport, it allows the upper level protocols to be more dynamic. This means they can grow and evolve with the network without affecting the transport mechanism. This, of course, was the main limitation in CCS6 networks. The structure was too constrained and did not allow for sufficient growth of any kind in the upper layers, due largely to the absence of an independent transport function.

		Message Transfer Part (MTP)

		The MTP is the transport protocol used by all other SS7 protocols in the TDM SS7 network. This protocol is actually divided into three different levels. In comparison with the OSI model, MTP provides the same functionality as layers one, two, and three.

		The physical level of MTP allows for the use of any digital-type interface supporting the data rate required by most networks. Common interfaces in most SS7 networks today include DS0A and V.35.

		The physical level, or level one, works independently of all other levels. This allows the upper levels to evolve to meet the everchanging demands of the network without affecting the interface.

		There is one exception to this rule, and that applies to the new broadband networks. There is a question as to whether or not existing interfaces will be sufficient for supporting the signaling in broadband networks. Telcordia has published standards for the usage of a full Digital Signal 1 (DS1) facility, at 1.544 Mbps, as a signaling link.

		The DS1 is commonly found in ISDN networks and is used by T-1 trunks as the basic carrier. Usage as a signaling link will reduce the number of multiplexers used in the network, since the use of DS0A requires a DS3 between two exchanges to be demultiplexed at two levels before the DS0A can be derived. Use of a DS1 will eliminate one multiplexer in these cases.

		The level-two function of MTP provides the functions necessary to provide basic error detection and correction for all signal units. This protocol is concerned only with the delivery of signal units between two exchanges or signaling points. There is no consideration outside of the signaling link.

		This implies that level two has no knowledge of the final destination. This is a fair assumption. Level two does not need to concern itself with this information. In the true spirit of the OSI model, this is left up to upper levels. Level two provides reliable transfer of information over a signaling link to the adjacent signaling point. Once the information reaches the adjacent signaling point, it is up to level three to determine how to route the message any further.

		Level two is maintained at the signaling link level. Each circuit card in an SS7 device must be able to provide and support this functionality independently of the rest of the system. For example, if there are several links connecting to the same signaling point, each link runs independently and does not concern itself with the activities of the other links.

		Sequence numbering is a function found at this level. Now that we understand that this level works independently of all other levels and all other links, we can assume that the sequence numbering is significant only on each particular link. In other words, if one link is transmitting messages using sequence numbers one, two, and three, there is no synchronization of sequence numbers on the other links. They may be using a completely different range of numbers, as long as they are all sequential. Each link maintains its own sequence numbering.

		This is also true for the adjacent signaling point on the same link. One signaling point can be sending sequence number 10, while the adjacent signaling point is sending sequence number 121. This is due to the fact that these links have independent processing that is not synchronized, allowing links to be much more efficient.

		Another function of level two is error checking. There are two methods of error checking: basic and preventive cyclic redundancy (PCR). Basic error detection/correction is used with all terrestrial signaling links. This is by far the most favorable, because it is much more efficient than PCR.

		PCR is used only with satellite signaling links, and it uses constant retransmission rather than error checking. With basic error detection and correction, when an error is detected, a retransmission is requested. The sequence number is provided for the last received signal unit that was good, allowing the originator of the bad signal unit to be able to determine which signal units to retransmit.

		In PCR, all transmitted signal units are retransmitted automatically during idle periods, until they are acknowledged. Once they have been acknowledged, they are dropped from the transmission buffer. They are continually retransmitted until the distant end acknowledges them. This, of course, is not efficient use of the network and creates a lot of overhead.

		The reason for this method lies in the propagation delay introduced when using satellite signaling links. If a signal unit is sent, a retransmission may cross an acknowledgment because of the delay encountered. This would lead to some confusing situations in which a signal unit is retransmitted (due to a time-out, for example) and, at the same time, an acknowledgment is received. The receiving end would also find itself somewhat confused if it sent an acknowledgment, only to find the same signal unit being retransmitted.

		Procedures to alleviate this are implemented whenever satellite is used. The general rule is not to use satellite for signaling links whenever possible, but when there are no other alternatives, the protocol will support the use of satellite and microwave as well.

		Level two also detects the presence of an opening flag for the delineation of an incoming signal unit. The flag is always a fixed pattern of 01111110, and is located in the first octet of the signal unit. As mentioned earlier, the opening flag is also the closing flag of the previous signal unit.

		MTP level three provides four functions: message routing, message discrimination, message distribution, and network management. Network management is probably the most important. Network management maintains the integrity of individual signaling links by continuously monitoring them and counting the number of errors which occur on any single link.

		When excessive errors have been counted, the link is removed from service (messages are blocked from the link) and the link is reinitialized. Since most errors are the result of clock signal degeneration and other related factors, resynchronization of the link usually resolves any problems that may occur.

		When a link is said to be functioning properly and messages are of the correct length, the link is in alignment. When messages are received that are not the correct length or if there is a ones-density violation, the link is said to be out of alignment.

		Network management can rectify this problem. There are several functions within network management. Each function looks after a specific area of the network. They are

		Link management

		Traffic management

		Route management

		Link management is concerned with the integrity of an individual link. While this is a level-three function, it relies on the service of level two to indicate when there is a problem on a link. The types of problems are typically errors, such as signal unit length and synchronization.

		Link management is capable of blocking messages from a particular link and notifying the adjacent signaling point to do the same. Once again, level two is utilized to alert the adjacent signaling point of a problem. The Link Status Signal Unit (LSSU) is used to inform an adjacent signaling point of the status of a link.

		Link management does not inform other signaling points in the network of its problems. Only adjacent signaling points need to be concerned about link troubles. Therefore, link management is a local function, and does not directly affect the performance of the overall network.

		There is a subtle impact, however, on the rest of the network when links fail. Link failures can cause traffic to reroute to another link, causing that link to become congested. If too much traffic is directed to another link and the processor cannot keep up with it, the signaling point can be considered congested.

		When a signaling point becomes congested, the adjacent signaling points are notified to reroute all traffic around the congested signaling point. This can result in delays in the network, and can even create congestion in other signaling points.

		Link management is also responsible for activating and deactivating links and, in some cases, even automatic allocation. Automatic allocation is a feature offered in some SSPs that provide both voice circuits and SS7 links. Automatic allocation removes voice circuits from service and automatically places them in service as SS7 signaling links. The circuits must be preconfigured for this capability. Not all systems offer this capability, but when it is offered, it can be valuable in handling sudden bursts in link demand.

		Traffic management provides the mechanisms for routing traffic around failed links within a linkset. Traffic management uses the MSU to send changeover and changeback messages to an adjacent node, informing the adjacent node of the failed links.

		This is not to be confused with link management, which is responsible for turning links up and taking links out of service. Link management is what controls the status of a link and informs the adjacent signaling point of the link status. The difference lies in the mechanism used to inform the adjacent signaling point.

		Link management uses the LSSU, which is carried on the link that is affected. Traffic management uses another link within the same linkset, and is used when a link fails for any reason to advise the adjacent signaling point to use another link within the same linkset. This mechanism is necessary when a link is unable to carry any level of traffic, such as when a backhoe digs up a link facility.

		Route management is used to advise other signaling points in the network about the inability of one signaling point to reach another signaling point. For example, if a signaling point becomes inaccessible by an adjacent signaling point, the adjacent signaling point will send a route management message to its adjacent signaling points to advise them that it can no longer reach the specified point code.

 

Transfer-restricted and transfer-prohibited messages are two of the most commonly used route management messages. In the event that a link becomes unavailable and link management sends a link management message to an adjacent signaling point, it is possible that, in time, if congestion occurs, route management will be implemented to alert other signaling points in the network (only those adjacent to the originating signaling point) that the destination (or affected signaling point) can no longer be reached.

		Besides the network management procedures just described, there are three other major functions within level three:

		Message discrimination

		Message distribution

		Message routing

		Message discrimination uses the routing label of the MSU to determine, first, whom a message is addressed to. If the routing label contains the address of the local signaling point, then the message is handed off to message distribution. If the address is of another signaling point, the message is handed off to message routing.

		Message distribution uses the service indicator octet (SIO) to determine who the user of a message is. If the SIO indicates that the user part is ISUP, the message is handed off to the ISUP. If the service indicator octet indicates the user part is the TCAP, the message is handed off to TCAP.

		Message routing attaches a new routing label to an outgoing message and determines which signaling link should be used to route the call. The signaling points routing table works with this function in determining the destination point code and the linkset that should be used to reach the destination.

		MTP2 User Adaptation Layer (M2UA)

		Developed by the IETF, this protocol is used to provide the services of MTP level two in TCP/IP networks used for SS7 transport. This includes providing link states and changing link states. M2UA also provides mapping from TDM SS7 links to IP links at the port level.

		One exception in M2UA is the absence of FSNs and BSNs. Sequence numbering is not maintained by this protocol. Sequencing is provided by the signaling control transmission protocol (SCTP).

		This protocol is used in IP-based networks to transport SS7 ISUP/TCAP messages between media gateway controllers (MGCs) and signaling gateways (SGs).

 

	MTP3 User Adaptation Layer (M3UA)

		This protocol provides the functions of MTP level three in TCP/IP networks used for SS7 transport. Address mapping from SS7 point codes to IP addresses is provided by this protocol. The actual mapping takes place at the signaling gateway (SG). Routing is based on several parameters depending on whether or not the message is ISUP or TCAP.

		For ISUP messages, routing is based on destination point code (DPC), origination point code (OPC), service indicator octet (SIO), and circuit identification code (CIC). These parameters are used to determine the IP address associated with media gateway controllers in the IP network.

		For SCCP/TCAP messages, routing is based on the destination point code (DPC), origination point code (OPC), service indicator octet (SIO), and subsystem number (SSN). These parameters are used to determine the IP address associated with service control points (SCPs) in the IP network.

		One significant difference between MTP3 and M3UA is the absence of protocol length limitations. MTP3 is limited to 256 bytes, whereas M3UA supports longer lengths. The BISUP and BICC (ISUP+) protocols allow these longer lengths for broadband applications.

		M3UA uses the services of the signaling control transmission protocol (SCTP).

		Signaling Control Transmission Protocol (SCTP)

		Also developed by the IETF, this protocol is used as a transport in IP networks and is used in association with M2UA and M3UA protocols. SCTP is used in place of TCP or UDP and has been developed specifically for use in SS7 networks where IP is the transport.

		Signaling Connection Control Part (SCCP)

		The SCCP is used only with the TCAP, although the standards indicate its use with the SUP.

		The purpose of SCCP is to provide the means for end-to-end routing. The MTP is only capable of point-to-point routing. This means that a message can be routed based only on the physical links available from a signaling point.

		SCCP provides the addressing to route a message through the entire network. This information is used at each signaling point by MTP level-three routing to determine which linkset to use.

 

The difference between MTP and SCCP is the way the information is used, and the nature of the addresses. The MTP provides both the origination point code (OPC) and the destination point code (DPC). In both cases, the point code is from a node-to-node perspective.

		In the case of SCCP, the address consists of three parts: called/calling party, point code, and subsystem number. The routing can be based on any of the three, although when routing by point code, the address is a combination of the point code and the subsystem number.

		When routing a TCAP message, the signaling point must be able to identify the destination, which is almost always a computer database or a specific signaling point. In many cases, there may not be any dialed digits associated with the transaction (although in today's applications this is not the case). SCCP provides the addressing needed by MTP to route a TCAP message through the network.

		The address information in SCCP remains fairly static, unless the point code and subsystem number are unknown to the originator. In this case, a STP will have to provide translation. This is usually the case when a number is dialed that cannot be routed by the network.

		The digits provided in the called party address are called global title digits. When the signaling point originating the message does not know the point code or the subsystem number of the database that will be providing a routing number for the requesting exchange, the global title digits have to be used by MTP level three for routing. At some point, the point code and subsystem number have to be provided so the message can reach its final destination. This function is known as global title translation and is usually provided by the STP adjacent to the destination database.

		When a number is dialed, such as an 800 or 900 number, the network cannot route the call based on conventional routing methods. This is because the numbering plan uses the area code of a number to determine which area in the nation's network the call should be routed to (the area being handled by a specific toll office). Likewise, the prefix usually denotes a specific central office that can route this call to the subscriber. In the case of 800 and 900 numbers, these do not have area codes that denote a toll office.

		The SS7 network will provide a routing number by which the end exchange can route the call. This requires the services of the TCAP and the SCCP. The called party address of SCCP will provide the dialed digits, although not all the digits are necessary. Only the area code (800) and the prefix are necessary.

		The number is compared in a database, which provides the routing number to TCAP. The routing number is then returned to the requesting exchange via TCAP and SCCP, so that a connection can be established for the call.

		This is just one example of how SCCP can be used. The called party address does not have to be dialed digits. In the cellular network, the MIN is placed in the called party address for roaming information.

		When used by the ISUP, SCCP will allow ISUP messages associated with an already established connection to be routed using end-to-end routing, the same as TCAP messages. This functionality has not yet been implemented in SS7 networks; however, with new services and the evolution of the IN, this may become necessary.

		ISDN User Part (ISUP)

		The ISUP is a circuit-related protocol, used for establishing circuit connections and maintaining the connections throughout a call. ISUP is associated only with voice and data calls and does not presently support broadband technologies such as Frame Relay and ATM. These new technologies will be addressed by a new version of ISUP called Broadband ISUP (BISUP).

		BISUP was developed by the ITU-TS. BISUP provides the mechanisms and parameters necessary to support the bandwidth and QoS requirements of these services.

		ISUP supports both analog and digital voice circuits, and was adopted by ANSI to replace the TUP. The TUP does not support data transmission or digital circuits. ISUP added the parameters necessary to support digital circuits and data transmission.

		ISUP is compatible with the ISDN protocol, which was developed as an extension of SS7 to the subscriber. There is direct mapping of ISDN Q.931 message types to ISUP message types, even though the message types are not the same. The purpose of the ISDN compatibility is to allow subscriber switches to send signaling information to remote subscriber switches during the call connection phase. After the connection has been established, ISUP supports communications between the two end-point subscriber switches. This feature may be necessary to support caller-invoked features, such as conference calling or automatic callback. The ability to invoke features and share information between two subscriber switches and/or networks is the unique capability of ISUP and the purpose for its development.

 

Broadband ISDN User Part (BISUP)

		To support the new broadband ISDN (BISDN) and ATM architectures, the ISUP protocol was modified. This new version of ISUP provides additional message types and parameters, which provide the support necessary for ATM and broadband networks.

		The most significant difference between the ISUP and the BISUP protocols is in the circuit assignment procedures and the type of circuits supported. ATM and broadband ISDN circuits are virtual circuits rather then physical circuits. This places new demands on the SS7 network, because it must be capable of assigning these virtual circuits and maintaining them. Because of the number of circuits available in broadband networks, a new circuit-numbering convention was adopted.

		In addition to the new circuit requirements, broadband networks also support the dynamic allocation of bandwidth, on a per-call basis. Now, when a call connection is established, the available bandwidth for that call is negotiated between the originating exchange and the destination exchange.

		There are a few other, more subtle, changes that appear in this newer version of the ISUP protocol. They have been described in more detail in the ISUP chapter (see Chapter 9). In addition to the procedure descriptions, there is also a section which explains the various message types and their parameters.

		Telephone User Part (TUP)

		The TUP is used in international networks. This protocol is compatible with the ISUP, with differences between the two mainly in the message type and parameters. Regardless of these differences, the two protocols can be mapped to one another successfully, even if a one-to-one mapping relationship does not exist. TUP is being replaced by ISUP at the international level as well.

		The U.S. and ANSI decided early on to replace this protocol with the ISUP protocol, in order to support the evolving services provided in many U.S. networks. The international market is just now learning of the possibilities that ISUP provides and is slowly evolving to this protocol.

		Before ISUP support for data services and digital facilities was provided through a now-obsolete protocol called the Data User Part (DUP). DUP is no longer used in U.S. networks and has been omitted from this book. Because of the migration towards ISUP at the international level, TUP has been omitted from this book as well.

		Probably the most noticeable difference between the structures of ISUP and TUP is in the header field. The ISUP protocol uses message types, whereas the TUP protocol uses an H0/H1 header. The H0/H1 header was originated from the CCS6 protocols. Messages are grouped into classes, which are represented by the H0 field. The H1 field denotes the specific message within that class. ISUP uses message types without any classification. This provides more flexibility in the protocol, and more room for growth.

		Transaction Capabilities Application Part (TCAP)

		The TCAP is probably the most versatile of all the SS7 protocols. TCAP is used for two purposes: accessing remote databases and invoking features in remote network entities.

		A network entity does not have to be a switch. Any network device, provided it is equipped with the proper interfaces and can provide all four levels of SS7 support, can be accessed by this protocol.

		In today's networks, TCAP is limited to database access, although more and more networks are providing new advanced services which require the use of TCAP to invoke those services. Custom calling features provided by the IN will most certainly require the support of TCAP to invoke features and services in remote switches.

		The TCAP protocol is not being used to its fullest potential today. The mechanisms currently provided in this protocol reach far beyond database access. As the IN evolves, the traffic mix in all SS7 networks will rapidly change to consist of mostly TCAP traffic.

		The TCAP protocol has been designed to provide for remote control of other network entities, which, in itself, holds many possibilities. For example, a subscriber wishes to change his or her telephone service. Normally, this would require a telephone call to the telephone company, which could remotely access the subscriber database and add the new service to the customer record. A service order would be generated and the new services programmed into the switch serving the subscriber. With TCAP capability, subscribers could enter the order entry system themselves. With an interface between the order entry system and the switch, subscribers could then change the program within the switch serving their telephone numbers.

		Of course, no one would expect a subscriber to understand the intricacies of a telephone office switch, so front-end interfaces using icons would be required to facilitate the subscriber. Sound far-fetched? Not really, since this is what the IN will do when it is complete.

		TCAP is an integral part of the IN, the cellular network, and, soon, the broadband services network. It is probably somewhat of a clich , but it is accurate to say that TCAP is ahead of its time.