SIGNALING TRANSFER POINT - Electrical and Computer Engineering

bunchlearnedNetworking and Communications

Oct 30, 2013 (3 years and 10 months ago)

213 views


PERFORMANCE TESTING OF

SIGNALING TRANSFER POINTS USED IN SIGNALING
SYSTEM 7 (SS7) NETWORKS



THESIS



Submitted in Partial Fulfillment

of the REQUIREMENTS for the



Degree of


MASTER OF SCIENCE (Telecommunications Networks)


at the


POLYTECHNIC UNIVERSIT
Y


by


Rimma Iontel


June 2001


_________________

Advisor

_________________

Date

_________________

Department Head

_________________

Date

Copy No._______



ii

AN ABSTRACT

PERFORMANCE TESTING OF SIGNALING TRANSFER

POINTS USED IN SIGNALING SYSTEM 7 (SS7) NETWOR
KS


by

Rimma Iontel

Advisor: Malathi Veeraraghavan


Submitted in Partial Fulfillment of the Requirements for the Degree of Master of Science
(Telecommunications Networks)

January 2001

The goals of this project are to understand how to test STPs for protoco
l conformance,
interoperability and performance. Detailed technical reports have been written for
protocol conformance and interoperability tests, and listed in the references of this thesis.
Hence the focus of this thesis is exclusively on performance t
esting. A rapid increase in
the volume of SS7 traffic has led to higher demands on Signaling Transfer Point (STP)
performance, specifically impacting processing delays. Taking required values as a
starting point, STPs should subject to a series of tests

to determine whether they exhibit
performance compatible with that demanded by the extensively deployed Advanced
Intelligent Network (AIN) services. By applying traffic from simulated nodes to a real
STP, processing delays are measured using monitoring e
quipment. After analyzing
captured data, we concluded that the Tekelec


Eagle


STP (Release 26) is able to
perform according to specifications.



iii

Table of Contents


List of Figures…………...………………………………………………….
iv

List of Tables
…………...…………….…………………………………….
v

1.0

Introduction…………...…………………………………………….
1

2.0

SS7 Overview………………………………………….……………
4

2.1

SS7 Network Architecture……….…………………………

4

2.1.1

Signaling Switching Point………………………….

5

2.1.2

Signaling Transfer Point……………...…………….

5

2.1.3

Service Control Point……………………………….

5

2.2

SS7 Protocol………………………………………….……..
9

2.2.1

Message Transfer Part……………………...……….
10

2.2.1.1

Primitives……………………………..…….

10

2.2.1.2

Signal Units…………...…………………….
13

2.2.1.3

Detailed View of Level 2 Functions………..

19

2.2.1.4

Detailed View of Level 3 Functions………..

22

2.2.2

Signaling Connection Control Part………………....

32

2.2.3

ISDN User Part……….….…………………………

37

2.2.4

Transacti
on Capabilities Application Part……...…..

38

3.0

SS7 over ATM……………………………………………………...

40

3.1

SS7 over High Speed Links Protocol Details……………….
40

3.1.1

AAL5 Common Part…………...…………..……
….

41

3.1.2

Service
-
Specific Connection Oriented Protocol……

42

3.1.3

Service
-
Specific Coordination Function……………

45

3.1.4

Layer Management...……………………………….

46

4.0

Testing Methods…………...……………………………………….

48

4.1

Conformance Testing…………...………………….……….

48

4.2

Interoperability Testing…………...………….…….……….

49

4.3

Performance Testing…………...……………
…….…….…..
50

5.0

Test Cases for STP Performance Testing…………...………………
53

5.1

Test Case 1…………...………………….………………….

56

5.2

Test Case 2…………...………………….………………….
61

6.0

Conclusion…………...………………….…………………………..
64

List of Acronyms…………...………………….……………………..…….

66

Appendix A: Captured MSU…………...………………….……………….

69

Appendix B: MGTS Traffi
c Reports for Test Case 2……………………....
71

References…………...……………………….……………………………..

75





iv

List of Figures


2.1

SS7 Network Topology…………………………………………….

6

2.2

STP Quad Configur
ation……………………………………………
8

2.3

SS7 Protocol Stack vs. OSI Model…………………………………

11

2.4

Format of the Primitive……………………………………………..

12

2.5

SCCP Connection Establishment………………
…………………...

12

2.6

Signal Unit Formats

a.

Fill
-
in Signal Unit…………………………………………..

17

b.

Link Status Signal Unit……………………………………..

17

c.

Message Signal Unit………………………………………..

17

2.7

Routing Label Format………………………………………………

18

2.8

Load Sharing…………………………………………….………….

19

a.

Linkset…………………………………………….………...

19

b.

Combined Linkset
…………………………………………..

19

2.9

User Part Unavailable Message……………………………………..
23

2.10

Message Handling…………………………………………………..

24

2.11

Selected Network Management Messages Form
ats

a.

Changeover Signal…………………………………………..
29

b.

Changeback Signal………………………………………….

29

c.

Transfer Prohibit/Restricted and Allowed Signals………….
29

d.

Transfer Controlled Signal…………
……………………….

29

2.12

Changeover/Changeback Procedure Flow……………………….....

29

2.13

Receive Buffer Congestion Thresholds…………………………….

32

2.14

General SCCP Message Format…………
………………………....

34

2.15

SCCP Connectionless Service Message Formats

a.

Unitdata Message…………………………………………...

35

b.

Unitdata Service Message…………………………………..

35

2.16

GTT Translati
on Example……………………………………….....

36

2.17

Basic ISUP Call Setup……………………………………………..

37

2.18

TCAP Call Setup……………………………………………………
39

3.1

High Speed Links SS7 Protocol Stack……………
………………...

41

3.2

AAL5 Common Part Data Units Formats…………………………..

42

3.3

SD PDU…………………………………………………………….

43

3.4

Signal, Primitive, and PDU Exchange for SSCOP Connection…….

47

5.1

Network Map for Test Case 1………………………………………

56

5.2

10% Initial Load Delay……………………………………………..

60

5.3

20% Initial Load Delay……………………………………………..

60

5.4

Netwo
rk Map for Test Case 2………………………………………

61

5.5

Cross
-
STP Delay for SSP 1 and SSP 2……………………………..

62

5.6

Cross
-
STP Delay for SSP 3 and SSP 4……………………………..

63

5.7

Cross
-
STP Delay for S
SP 5 and SSP 6……………………………..

63



v

List of Tables


2.1

Status Indication Values……………………………………………

15

2.2

Signaling Information Octet Values

a.

Sub
-
Service Field Values..……………….…………………

16

b.

Service Indicator..……………….………………………….

16

2.3

Heading Codes H0 and H1…………………………………………

30

5.1

STP Node Processing Time...………………………………………

53

5.2

Link Output an
d cross
-
STP Delays………………………………..

55

5.3

Performance………….……………………………………………..

59



1

1.0

Introduction


In the last hundred years telephones have become a standard item in any home,
office and on the st
reet. No one gives a second thought to the action of picking up a
receiver, dialing a number, and hearing a voice at the other end, with all the actions
occurring within less than a minute. Very few people realize the flurry of activities
occurring in th
at minute. Consider this; raising the receiver produces a dial tone;
punching in a few numbers connects you to that one unique phone, perhaps a continent
away from you. What makes this possible?


In the early days of telephony you would first call an o
perator sitting at the
switchboard, give the operator the telephone number of the called party and upon
determining that the other end is free and available the operator would complete the
connection, allowing you to talk. Sounds good enough if there are
a few thousand
possible connections, but what happens when the number grows to millions? When the
phone networks spans around the globe and you want to talk not only to your
grandmother in Cleveland but also to your business partner in Tokyo? The job of
a
person at the switchboard becomes harder and less efficient. You would not be able to
talk to Tokyo immediately. If you are lucky, you might be able to do it in an hour,
perhaps longer. Waiting hours to complete one phone call in this information era?

This
era would not have arrived if we still had people sitting at switchboard.


The process of establishing connections became automated, and in the place of a
switchboard, we now have electronic switches. Switches communicate with each other
using si
gnaling messages traveling through the network. There are two types of
signaling: in
-
band and out
-
of
-
band. Until 1986, the telephone network used in
-
band


2

signaling, carrying signaling data over the same voice trunks that would later carry the
correspondi
ng conversation. This method gave a reasonable assurance that trunks would
be available and voice data would go through once the connection was established. The
problem was the trunks that between switches would be in use until released even if the
desti
nation turned out to be unavailable. No other call would be able to utilize those
trunks even though they do not carry any actual conversation. In
-
band signaling for
telephone networks proved to be inadequate and out
-
of
-
band signaling was adopted
instead
. Out
-
of
-
band signaling is signaling that does not take place over the same path as
the conversation.


Signaling System 7 (SS7) is the currently used standard for the telephone
signaling network. It uses common channel signaling, which is a “
signalling

method in
which a single channel conveys, by means of labeled messages, signalling information
relating to a multiplicity of circuits or calls and other information, such as that used for
network management [1].”
It means that a separate network with its

own nodes and links
was built to provide support to the conventional voice network. SS7 is a digital packet
switched network that can carry information about a number of calls over the same link
simultaneously. It is responsible for connection set up, c
ontrol and tear down, as well as
routing and network maintenance. With SS7, signaling can take place during the
conversation instead of only at the beginning.


An added benefit of the SS7 network and protocol is that when combined with
Intelligent Ne
twork (IN) or Advanced Intelligent Network (AIN) SS7 can support a
number of different services that can be implemented without any modification to the
network structure.



3


Each added service, however, generates increased message load, and it is vital for

the reliable and timely network performance that nodes can handle traffic offered to
them. To ensure this, SS7 is slowly migrating to higher speed links that can carry greater
loads at the same time, increasing the processing power of the network compone
nts.
Analysis of performance and development of the methods to conduct that analysis is one
of the factors that can help the network grow in a stable and reliable manner, preventing
both over engineering and under engineering.


In the following chapt
ers, I provide a detailed overview of the SS7 architecture
and protocol as well as the test cases for performance testing with the corresponding
analysis and conclusions. Chapter 2 contains an overview of SS7 over low speed links.
Chapter 3 reviews SS7 pr
otocol used over high speed links. Chapter 4 covers different
testing methods that can be applied to SS7. Chapter 5 describes test cases performed in
the course of the research, and finally, Chapter 6 states or conclusions reached after the
completion of
research and testing.





4

2.0

Signaling System 7 Overview

Signaling System 7 (SS7) is a connectionless packet switched network made up of
a number of different Signaling Points (SP) interconnected with 56 kbps, 64 kbps, or
1.544 Mbps transmission links. The ne
twork is used to set up and tear down calls made
by user of the Public Switched Telephone Network (PSTN). In addition, it provides
access to various databases (800 numbers, for example) and Advanced Intelligent
Network (AIN) services, such as caller ID, c
onference calling, call forwarding and so on.

One of the most important features of the SS7 network is its reliability. Telephone
companies can ill afford to have network failures that would leave their customers
without telephone service. Since the PSTN
cannot function without its backbone SS7
network, failure in providing signaling results in failure in the phone network’s ability to
support a call. To assure consistent service, most nodes in signaling network are
duplicated and loaded in such a way tha
t in case of a failure, the full load of one of the
nodes can be picked up by the remaining node. The same is true for signaling links
connecting the nodes.


2.1

SS7 Network Architecture

An SS7 network (example shown in Figure 2.1) is made up of different

types of
nodes, each of which serves specific signaling functions. Each node has one or more
addresses, called point codes (PCs), associated with it. Nodes communicate with each
other over the connecting links using datagrams, the headers of which carry
destination
PCs.



5

2.1.1

Service Switching Point

A Service Switching Point (SSP) is a local exchange switch that is used to convert
signaling received from the voice switch into SS7 signaling messages.

2.1.2

Signaling Transfer Point

A Signaling Transfer Po
int (STP) provides SS7 with the functionality of a router,
it routes messages through the network to their appropriate destinations. There are three
types of STPs:



National



International



Gateway



Used to convert between protocols

Besides routing data, STP
s gather measurements that can be used to monitor
traffic and network use. Examples of data acquired at the STP are peg counts, statistical
and maintenance information as well as message types.


Because of the high reliability requirements specified for

the signaling network,
STPs are usually deployed in pairs called mated pairs. The two mates are fully
redundant, and each one is expected to be able to assume the full load of its mate in case
of failure.

2.1.3

Service Control Point

A Service Control Point (
SCP) is an interface into the phone company’s
databases. It is a computer that stores information related to one or more intelligent
network based services. Among the most common databases used by telephone
companies are:



6



































Figure 2.1 SS7 Network Topology












STP

STP

STP

SSP

SSP

SSP

SSP

SSP

SSP

SSP

SSP

SCP

SCP

SCP

SCP

SSP

SSP

SSP

D

D

C

A

A

B

F

E

STP

STP



7



800 Number Translations Database



Call Management Services Database (CMSDB)



Local Number Portability (LNP)



Line Information Database (LIDB)




Business Services Database (BSDB)



Home Location Register (HLR)



Visitor

Location Register (VLR)

As the Advanced Intelligent Network (AIN) services become more widespread, databases
play an increasingly important role in the telephone network. They are a focus point of
the AIN services functionality.

2.1.4

Signaling Data Li
nks

SS7 networks use bi
-
directional high speed (ATM, 1.5 Mbps) or low speed (DS0,
56 kbps/64 kbps) links to interconnect signaling points. Two signaling points connected
by one link are said to be adjacent. A collection of links that connect together the

same
signaling points is called a link set and each link set can contain up to 16 separate links.
A group of link sets used to reach a particular destination is called a route and a
combination of routes that can be used to reach the same destination is
called a route set.

While each link should be able to handle a full load, entities are set up to load
each link only up to 40% when operating under normal conditions. Even though the link
utilization suffers from such a setup, it allows for a greater re
liability. In case of a link
failure the full load from a failed link can be safely transferred to the remaining links in a
linkset with no degradation of service as long as the load on each link does not exceed
80%.



8

Links employed in the SS7 network are

divided into 6 different types depending
on which two network nodes are connected by the link. Each type has a maximum
allowed and a minimum required number of links in the linkset.



A
-
links

Access links connect together an SSP and an STP, or an SCP and a
n STP. They
provide access into the network and to the databases. Each linkset has at least one
and at most 16 A links supporting it.



B
-
links

Bridge links connect one mated STP pair to another mated STP pair at the same
hierarchical level (two pairs of r
egional STPs for example). B links are always
deployed in a quad configuration shown in Figure 2.2. There can be a maximum
of eight B links connecting the pairs.









Figure 2.2 STP Quad Configuration



C
-
links

Cross links connect STP mates together.
They only carry user traffic in case of
congestion or network failure but usually are reserved for network management
messages. They are always deployed in pairs for redundancy, with a maximum of
eight C links between two STPs.



D
-
links

Mated pair



9

Diagonal links conn
ect mated STP pairs at one hierarchical level to mated STPs at
another hierarchical level. They are deployed in the same fashion as B links.



E
-
links

Extended links are used to connect an SSP to a remote STP. They are used when
there is a significant am
ount of traffic going between the nodes to avoid
congestion.



F
-
links

Fully associated links connect two SSPs when there is either a large amount of
traffic between the two SSPs or when the SSPs cannot be connected through an
STP.

2.2

SS7 Protocol

The SS7

Protocol provides a set of rules controlling “the way data is transmitted
and received over the data communication (SS7) network [2].” The SS7 protocol stack
approximately maps to the OSI model, as demonstrated in Figure 2.3.


SS7 protocol is divided i
nto 4 separate layers: physical, data link, network and
user parts. The first three layers together make up Message Transfer Part (MTP) that is
responsible for transmitting messages between signaling nodes. All SSPs and STPs
terminate MTP. For database
transactions STP also terminates Signaling Connection
Control Part (SCCP). MTP and SCCP combined are referred to as Network Service Part
(NSP), which serves functions similar to those of the first three layer of the OSI. SSPs
also contain User Part layer
s that correspond to Application layer of the OSI model.
There are different user parts defined and implemented but not all of them have to be
present in a particular switch. For example, Broadband ISDN User Part (BISUP) is


10

defined to be used with ATM fa
cilities and hence, it does not have to be implemented in
the switch that is only used with low speed links.

Similarly, Telephone User Part (TUP) is only used in international networks and
does not have to be included in switches inside the national networ
k, which instead use
the ISDN User Part (ISUP).

2.2.1

Message Transfer Part

MTP combines in itself functions of the first three layers of the SS7 protocol
stack. It provides message handling and traffic management functions.



Layer 1 converts digital d
ata into a bit stream transmission of the stream over the
network. It is defined for use with various interfaces, such as DS0 (64 kbps, bipolar
non
-
return
-
to
-
zero format) or V.35.



Layer 2 provides error detection/correction and sequenced delivery of signa
ling
messages on a link
-
by
-
link basis.



Layer 3 is responsible for message routing, discrimination, and distribution in
addition to performing network management functions, including link, route and
traffic management.

2.2.1.1

Primitives

Primitives are us
ed to pass information between protocol layers. The format is
illustrated in Figure 2.4.



X marks the originator of the primitive, it can be either MTP or N (SCCP)



Generic Name defines the type of information provided by the layer. Each layer has a
numbe
r of different primitives associated with it.



11



Specific Name describes to the receiver the action that is to take place. There are four
types of primitives:



Request

is used to invoke a service from the recipient of the primitive




































Figure 2.3 SS7 Protocol Stack vs. OSI Model


Phy
sical

1

Data Link

2

Network

3

Transport

4

Session

5

Presentation

6

Transport

7

Level 1

1

Level 2

2

Level 3

3

SCCP

TCAP

OMAP

I

S

U

P

MTP

User
Parts

NSP



12



Indication

is returned by a peer entity to advice that a service had been invoked
by the user or service provider



Response

is used to complete a transaction between layers



Confirmation

informs the req
uest
-
generating layer that the request has been
completely processed [3].






Example:




Figure 2.4 Format of the Primitive

A detailed example of the use of all four types of primitives is shown in Figure 2.5,
where two entities are trying

to establish an SCCP connection, used by a connection
-
oriented class of service.










Figure 2.5 SCCP Connection Establishment


1.

Setup is initiated by a calling user of Entity 1 with N
-
CONNECT.request, which
it sends to SCCP.

2.

SCCP reviews the reques
t and after attaching protocol control information (PCI)
to it passes the primitive on to its peer in Entity 2 using underlying MTP.

User

SCCP

MTP

User

SCCP

MTP

1

2

3

4 pending

4 pending

5

6

7

Entity 1

Entity 2

X

Generic Name

Specific Name

Parameters

X

MTP
-
STATUS

Indication

Affecte
d DPC, Cause



13

3.

SCCP of Entity 2 recognizes the request and sends N
-
CONNECT.indication to
the appropriate user letting it know that a rem
ote point is trying to establish a
connection.

4.

At this point the connection is pending at both ends.

5.

User issues an N
-
CONNECT.response telling SCCP that it approves the attempt.

6.

SCCP passes the response along through MTP to its peer in the other entity a
fter
attaching PCI field.

7.

After receiving the primitive SCCP generates an N
-
CONNECT.confirmation to
its user, which at last completes the connection.

2.2.1.2

Signal Units

SS7 networks are packet switched networks, and information is passed between
nodes
in packets called Signal Units (SUs). There are three types of SUs: Fill
-
in Signal
Unit (FISU), Link Status Signal Unit (LSSU) and Message Signal Unit (MSU). They
have similar formats, presented in Figure 2.6.



FISU (Figure 2.6a)

FISUs are used for conti
nuous error checks on the link. Since user data usually
comes in bursts, if FISUs were not used, there would be periods of silence on the links.
In an event of a malfunction or a failure the condition would not be detected until a node
attempts to transf
er some data. Use of FISUs guarantees early problem detection. They
are only sent when there are no other messages waiting for transmission, which means
that no resources are taken away from users.

Fields present in FISUs that are also used by LSSUs an
d MSUs are:



Frame Check Sequence (FCS)



14


SS7 uses CRC
-
16 for error detection. Upon receiving an SU, MTP compares
the calculated CRC against the one in the FCS field and if there is a
discrepancy, it is counted against the link.



Length Indicator (LI)


LI
can be 0, 1, 2 or greater. It refers to the length in octets of the information
field. FISUs don’t contain information fields, so their LI is always set to 0.
Value of the LI is used to determine the type of the SU.




Forward Indicator Bit (FIB)/Backward

Indicator Bit (BIB)


These two fields are used for acknowledgment purposes. There are set to the
same value, unless a negative acknowledgement is being sent. Then, the
value of BIB is toggled and message is sent back indicating request for
retransmissio
n from the sending MTP Level 2.



Forward Sequence Number (FSN)/Backward Sequence Number (BSN)


To acknowledge a successfully received SU, BSN of the message being
transmitted is set equal to the FSN of the received SU. When negative
acknowledgement is bei
ng sent, BSN indicates the last good SU received and
all the messages at the other end with sequence numbers greater than BSN
will be retransmitted.



Flag


Flag is a fixed bit pattern, 1 octet long, used by Level 2 to determine the
boundaries of Signal Unit
s. The opening flag of an SU serves as the closing
flag of the previous SU.



LSSU



15

LSSUs are used to indicate the status of a link between two adjacent signaling
points. Information is transferred in LSSUs using a Status Field (SF), which carries the
stat
us of the link on which it is being transmitted. An SF can be 8 or 16 bits long,
producing LI value of 1 or 2, but only the first 3 bits of the first octet are currently being
used, the rest are set to zeros. The way the SF is divided is shown in Figure
2.6b. Table
2.1 contains the values and corresponding link status values that the Status Indication (SI)
portion of the SF can take.

SIO warns about the loss of alignment when the received Signal Unit violates
ones density or its Signaling Information Fi
eld (SIF) exceeds 272 bytes. SIN and SIE are
only used during the alignment procedure to indicate the length of proving period. SIOS
means that SP cannot receive or transmit any MSUs but there is no processor outage. It
is also used at the start of the
alignment procedure. SIPO indicates that level two cannot
reach level 3 or 4. This condition stops the SP from transmitting any MSUs, which
means that only FISUs will be sent across the links between affected point and its

CBA

Status Indication

000

“O”:

潵t 潦 alig湭ent

〰M

“N”: normal alignment

〱M

“E”: emergency alignment

〱M

“OS”: out of service

㄰N

“PO”: processor outage

㄰N

“B”: busy




Table 2.1 Status Indication Values


adjacent points. If the condition persists, realignment will occur. Fi
nally, SIB is used to
indicate congestion at level two. An SP, when it receives an SIB, stops transmitting
MSUs and waits until the congestion has abated. After 3
-
6, seconds if the congestion did
not subside, level two notifies level three of a link fail
ure. To prevent excessive delay of


16

acknowledgment timer (T7) from expiring and level three from initiating realignment,
SIBs are sent every T5 (80
-
120 ms) to reset T7. Still, if T6 expires indicating a long
congestion period alignment will be activated.


If a received LSSU contains an error, it is not retransmitted but the error is
counted against the link.



MSU

MSUs are used to transmit user data and all the information other than what is
transmitted in the LSSUs. There are two information fields in t
he MSU, Service
Indicator Octet (SIO) and Signaling Information Field (SIF). SIO is used to determine
the protocol at the User Part level the message is destined to and the version of the
protocol used, national or international. Two portions of the SIO
are illustrated in Figure
2.6c with the values listed in Table 2.2.

DC BA


00 xx

International Network

01 xx

Spare

10 xx

National Network

11 xx

Reserved for national use


a.

Sub
-
Service Field


DCBA

Protocol

0 0 0 0

Signaling Network Management

0 0 0 1

Signaling Network Testing (SNT)

0 0 1 1

SCCP

0 1 0 0

Telephone User Part (TUP)

0 1 0 1

ISUP

0 1 1 0

Data User Part(DUP), call and circuit related

0 1 1 1

DUP, facility registration and cancellations


b.

Service Indicator

Table 2.2 Signaling In
formation Octet Values



17






a. Fill
-
in Signal Unit












b. Link Status Signal Unit


















c.

Message Signal Unit

Figure 2.6 Signal Unit Formats

Direction of
transmission

FCS


LI

F

I

B

FSN

B

I

B

BSN

Opening

Flag

8

16

2

6

1

7

1

7

Level 2

Direct ion of
t ransmission

FCS


LI

F

I

B

FSN

B

I

B

BSN

Opening

Flag

8

16

2

6

1

7

1

7

SF

8 or16

Level 2


Level 3

Level 2

Spare

Status Indications

Direction of
transmission

FCS


LI

F

I

B

FSN

B

I

B

BSN

Opening

Flag

8

16

2

6

1

7

1

7

SIF

8n, n>2

SIO

8

Lev
el 2

Level 2


Level 3 and above

Sub
-
Service
Field

Service Indicat or

Management or User Information

Rout ing
Label



18


The SIF field contains upper layer data, including a Routing Label. The SIF can
be

of variable length, between 3 and 252 octets. The value of LI greater than two means
that the SU being transmitted is an MSU. The Routing label (Figure 2.7) is 56
-
bits long
and consists of three fields: Destination Point Code (DPC), Origination Point Co
de
(OPC), and Signaling Link Selection (SLS).









Figure 2.7 Routing Label Format

DPC contains the point code of the destination signaling point, while OPC
contains the point code of the sender. They have the same format and each is 24 bits
long. Th
e fields that make up a point code are: Network Identifier (NI), Network Cluster
(NC), and Network Cluster Member (NCM). Network Identifier can either directly
identify the network to which the point code belongs or it can contain one of the escape
codes,

which were designed to increase the available number of different networks that
can be addressed. The 8
-
bit length of the NI field allows for a maximum of only 256
different values. When one of reserved values used for escape code is placed in the NI
fi
eld, NC is used to determine the network for which the message is intended. The
combination of NI, NC and NCM uniquely identifies a signaling point.


The SLS field is used for load balancing, i.e., to enable all the available links in a
link set to carry
equal loads. Load can be shared between links in the same link set or in
a combined link set, as illustrated in Figure 2.8.

SLS

OPC

DPC

Network
Identifier

Network Cluster

Network Cluster
Member

24

24

8

8

8

8



19







a. Linkset












b. Combined Linkset

Figure 2.8 Load Sharing

2.2.1.3

Detailed View of Level 2 Functions

Level 2,
being a Data Link Layer, makes sure that packets flow across a correctly
functioning link between two adjacent points in an orderly, sequential manner, with no
loss of data. It is the level charged with monitoring the performance of a link and taking
appr
opriate actions in case of malfunction or failure.



The assigned duties of this layer are:



Signal Unit delimitation

This function is performed using a unique 8
-
bit pattern, flag (01111110), to
indicate the boundaries of a signal unit. Most of the time

each SU has only one
opening flag, which acts as a closing flag for the previous SU. Nevertheless, an
SP should be able to handle the case when it receives two or more consecutive
flags. To prevent the flag pattern from appearing in the information field

of a
signal unit, bit stuffing (inserting 0 after 5 consecutive 1s) is used before flags are
SLS = XXXXXXX0

SLS = XXXXXXX1

Link set

Link set

SLS = XXXXXXX0

SLS = XXXXXXX1



20

attached to the SU. Upon receiving an SU, MTP level 2 strips the flags and
removes 0s following five consecutive ones.



Signal unit alignment acceptance

Detects
events indicating a possible loss of alignment that occurs when ones
density violation is detected (more than 7 consecutive ones received), the length
of a signal unit exceeds 272 bytes indicating the loss of a flag, or the signal unit’s
length is not a mu
ltiple of 8. Any of these events prompts a change in the error
monitoring function.



Signal unit error detection

Errors are indicated either by a discrepancy in the FCS field with the CRC
-
16
computation performed by the recipient or by sequence number erro
rs.



Signal unit error correction

There are two error correction methods: “basic method” (for terrestrial
transmissions), which is more efficient, and retransmits the errored SU and all
SUs with sequence numbers following it, and “preventive cyclic retra
nsmission
method” (for satellite transmissions), which keeps retransmitting all SUs in the
buffer until it receives an acknowledgement.



Signaling link error monitoring

Separate monitors are used when a link is in service, and during the link proving
period
:



Signal Unit Error Rate Monitor (SUERM) is used when a link is in service.



21

It keeps a counter, which is incremented with every error, and decremented by
one every 256
th

signal unit received without an error. When the counter
reaches 64, the link is tak
en out of service and realigned.



Alignment Error Rate Monitor (AERM)

This monitor is used during the link proving period. A counter is incremented
with every error detected, and alignment is restarted if too many errors are
detected.



Signaling link al
ignment

Alignment can be performed both when the link is first brought into service and
when the link is being restored after a failure. It follows the set procedure [4]:



State 00


idle

The procedure is suspended



State 01


not aligned

Time
-
out timer T2
is associated with this state. The link is not aligned and the
SP sends SIO to the adjacent node.



State 02


aligned

The link is aligned and SINs or SIEs are sent but the signaling point is not
receiving SINs, SIEs, or SIOS. Decision whether to perform n
ormal or
emergency alignment depends solely on MTP Level 3.



State 03


proving

Validates link’s ability to carry traffic by sending FISUs over the link for T4
period. Expiration of T4 signals successful ending of proving, except when
proving had fail
ed up to four times before.



22



Aligned/ready state

Lasts for T1 period to allow the remote end to perform four alignment
attempts.



In Service

Whenever the alignment is aborted, the signaling point returns into the idle
state and restarts the procedure.



Flow c
ontrol

Flow control function is used to notify the sender of the congestion status at the
receiving end using the SIB LSSU. The sender should stop transmitting MSUs
until congestion has abated or, in case of an excessive congestion period, until
after suc
cessful alignment.

2.2.1.4

Detailed View of Level 3 Functions

Level 3 functions are divided into Signaling Message Handling and Signaling
Network Management functions, whose functions are in turn divided as follows:



Signaling Message Handling (SMH)

SMH is
responsible for making sure messages originating at a User Part in one
entity are delivered to the corresponding User Part in the entity the message is
intended for (as indicated by the DPC), regardless of whether the nodes are
directly connected.



Message
Discrimination

Determines whether the message is destined to the User Part in this entity, in
which case, the message is passed on to Message Distribution; if not, and the


23

SP has message routing capability, the message is transferred to Message
Routing.



Me
ssage Distribution

Determines which User Part the message should be handed to based on the
information in the SIO field of an incoming MSU. If the User Part is
unavailable a “user part unavailable (UPU)” (Figure 2.9) message is generated
to the originator

of the message indicating that the message was discarded.
UPU contains a CAUSE field that lists the reason the message could not be
delivered to the destination:



User Part function was not equipped at the destination



User Part function was not accessible



User Part function could not be reached for an unknown reason




Figure 2.9 User Part Unavailable Message



Message Routing

Message Routing determines which link an incoming message should be
routed toward in order to reach the destination. Routing is don
e based on the
routing label and can either be full point code routing or partial point code
routing, which includes cluster and network routing. The route is determined
by a lookup of the routing table, which is set up by the network administrator.
The
routing table indicates outgoing links or link sets corresponding to
various destinations. The routes listed can include a primary route and a
FCS

CAUSE

User ID

Destination

H1

H0

Routing
Label

SIO


LI

FIB

FSN

BIB

BSN

Flag



24

number of alternate routes arranged in order of priority. The routing function
should pick the best route marke
d as available from the list of multiple routes.

The flow of traffic between message handling functions is illustrated in Figure 2.10.



















Figure 2.10 Message Handling



Signaling Network Management (SNM)

In case of link failure or congestio
n, SNM ensures that the network adapts to
handle the changes. There are a number of procedures defined with each SNM
function to make sure the data keeps flowing even in the presence of
malfunctions.



Signaling Traffic Management

Diverts traffic from fail
ed or congested links/routes/SPs to links/routes/SPs
that are still available. It employs the following procedures:

1.

Changeover

The Changeover procedure is initiated when a link (or a link set) becomes
unavailable due to a failure, blocking or inhibition.

Traffic is rerouted to an available
User Parts or

Network Management

MTP Level 2

Discrimination

Distribution

Routing

incoming

outgoing

internal

ex
ternal

MTP3

SMH



25

link (picked from the routing table) in the same link set, in a combined link set or to a
completely different route, which might not be a preferred route. The goal of this
procedure is to ensure the traffic keeps flo
wing with messages arriving at the destination
in proper order without any losses and also without any negative effect on the traffic
already flowing on the link used for rerouting. The procedure also retrieves messages not
yet received by the other end o
f the failed link, and retransmits these messages on the
new route.


Changeover
-
order (COO) and changeover
-
acknowledgement (COA) signals are
used during the changeover procedure. They are transferred as a part of an MSU with
information included in the SI
F field, as shown in Figure 2.
11a
. Spare bits are set to 0s
for low speed links and are used with some messages on 1.536 Mbps links, i.e. they are
used to accommodate a longer FSN in COO [5]. Heading codes are listed in the Table
2.3.

2.

Changeback

The c
hangeback procedure is used when a previously unavailable link becomes
available, and traffic can be return to its normal route. The changeback signal (CBD),
containing a changeback code as shown in Figure 2.11b, is used to perform the
procedure. The rem
ote end confirms its acceptance of the procedure via a changeback
-
acknowledgement signal (CBA). The flow and timing of the messages exchanged during
changeover/changeback procedures are shown in Figure 2.12.

3.

Forced rerouting



26

Forced rerouting takes place

when an SP receives a transfer
-
prohibited (TFP)
message indicating that a certain destination has become unavailable on a given route.
Traffic will be rerouted to a new route picked from a routing table.

4.

Controlled rerouting


Controlled rerouting is init
iated upon receipt of a transfer
-
allowed (TFA) or a
transfer
-
restricted (TFR) message by a signaling point. A TFA is generated when a
previously unavailable route to a destination is recovered and, a TFR message is
generated when there is a route still av
ailable to a destination but it is not a preferred
route.

5.

MTP restart

MTP restart is used when a node has been isolated from the network for a period
of time and is then restored. The procedure allows the node to bring up some of its links,
sufficient to

support some traffic load, before actually starting to transfer any traffic.
This protects the node from being overwhelmed by user data and going back down again
because not enough links were made available.

6.

Management inhibiting

This procedure is used
by the administrator during maintenance or testing and
does not take the link out of service. It remains available for network management
messages. The node can refuse the link to be inhibited if it affects the availability of the
destinations or during
congestion.

7.

Signaling traffic flow control

Signaling traffic flow control ensures that a node does not generate traffic that the
network cannot handle in the presence of some network failures or congestion.



27



Signaling Link Management (SLM)

Manages the s
tatus of links, activating idle links and restoring failed links by
initiating the alignment procedure, and deactivating aligned links when
needed. For example, if the number of available links exceeds the allowed
limit (a link set should contain a maximu
m of 16 links) extra links should be
deactivated. The SLM procedures are:

1.

Signaling link activation, restoration, and deactivation

Both the link activation of a previously inactive link, and the restoration of a
failed link, go through two stages: initial

link alignment and link test. In both cases,
measures are taken to prevent the link from oscillating between in
-
service and out
-
of
-
service states by preventing the link from being brought into service upon failure if the
failure occurred before a corresp
onding timer expired.


2.

Link set activation

The link set activation procedure is used when a link set contains no functioning
links in it. Either “normal” or “emergency restart” activation can take place. Emergency
restart is used when the normal p
rocedure is deemed to be too slow; for example, the case
when activation of the link set will make a previously unavailable signaling point
accessible.

3.

Automatic allocation of signaling terminals and signaling data links



Signaling Route Management

Distri
butes information about the status of the network and the availability of
routes using the following procedures:

1.

Transfer
-
prohibited procedure



28

A signaling node initiates the transfer
-
prohibited procedure to notify its neighbors
nodes about its inability t
o route messages to a certain signaling point or cluster. The
format of transfer
-
prohibit (TFP) message is shown in Figure 2.11c. The reception of a
TFP message may set off forced rerouting procedure, and, in turn, generate additional
TFP messages.

2.

Tran
sfer
-
allowed procedure

When a route through a signaling node to an affected SP becomes available,
transfer
-
allowed (TFA) messages are issued to adjacent nodes, which might return
previously diverted traffic to its normal route and notify other nodes of th
e route
availability through TFA messages. The TFA has the same format as a TFP.

3.

Transfer
-
restricted procedure

With the help of transfer
-
restricted (TFR) messages, a node notifies its adjacent
nodes that, if possible, they should use a different route to

reach a specified destination.
The procedure is invoked if a node has to use a lower priority alternative route to reach
the destination or when the employed route experiences congestion. In the first case
upon switching from the preferred to an alterna
te route, the node starts timer T11, and
when the timer expires, broadcasts

TFR messages to all adjacent routes. Also, if
the
route is in the danger of being congested, the node issues TFR messages in response to
receiving a message directed to the concer
ned destination. The format of TFR messages
is illustrated in Figure 2.11c.

4.

Signaling
-
route
-
set
-
test procedure

The procedure is used to test the current availability of the destination and a node
invokes it upon receiving a TFP or TFR message from the ad
jacent node. Route
-
set
-
test



29






a.

Changeover Signal






b.

Changeback Signal







c.

Transfer Prohibit/Restricted and Allowed Signals






d. Transfer Controlled Signal


Figure 2.11 Selected Network Management Messages Formats















Figure 2.12 Changeover/Changeback Procedure Flow



FSN of last
accepted MSU

SLC

H1

H0

Routing Label

56

4

4

4

5

7

Changeback
Code

SLC

H1

H0

Routing Label

56

4

4

4

4

8

St at us

Dest inat ion

H1

H0

Routing Label

56

4

4

24

6

2

Dest inat ion

H1

H0

Routing Label

56

4

4

24

SP A

SP B

COO (link2)

COA

St art T2

Traffic will be diverted either
when COA is received or
when T2 expires.

Traffic diverted
t o link 2

Link 1 fails

Traffic flows over link 2

Link 1
rest ored

CBD

CBA

Traffic flows over link 1

Tr
af f i c di vert ed
back t o l i nk 1

St ar t T 4

If T4 expires before CBA is
received CBD is t rans mit t ed
again and T5 s t art s. Traffic
will not be divert ed unt il CBA
is received.



30

DCBA

Signal Codes

DCBA

Signal Codes

H0

H1

0 0 0 1

Changeover and
changeback

0 0 0 1

Changeover order

0 0 1 0

Changeover acknowledgment

0 0 1 1

Extended changeover order

0 1 0 0

Extended
changeover acknowledgment

0 1 0 1

Changeback declaration

0 1 1 0

Changeback acknowledgment

0 0 1 0

Emergency changeover

0 0 0 1

Emergency changeover order

0 0 1 0

Emergency changeover acknowledgment

0 0 1 1

TFC and RSCM

0 0 1 0

Transfer
-
contro
lled

0 0 0 1

Signaling
-
route
-
set
-
congestion
-
test

0 1 0 0

TFP, TFA, and TFR

0 0 0 1

Transfer
-
prohibited

0 0 1 0

Transfer
-
cluster
-
prohibited

0 1 0 1

Transfer
-
allowed

0 1 1 0

Transfer
-
cluster
-
allowed

0 0 1 1

Transfer
-
restricted

0 1 0 0

Transfer
-
cluster
-
restricted

0 1 0 1

Signaling
-
route
-
set
-
test

0 0 0 1

Signaling
-
route
-
set
-
test for prohibited
destination

0 0 1 0

Signaling
-
route
-
set
-
test for restricted
destination

0 0 1 1

Signaling
-
route
-
set
-
test for prohibited cluster

0 1 0 0

Signaling
-
route
-
set
-
test for restricted cluster

0 1 1 0

Management inhibiting

0 0 0 1

Link inhibit

0 0 1 0

Link uninhibit

0 0 1 1

Link inhibit acknowledgment

0 1 0 0

Link uninhibit acknowledgment

0 1 0 1

Link inhibit denied

0 1 1 0

Li
nk force uninhibit

0 1 1 1

Link local inhibit test

1 0 0 0

Link remote inhibit test

0 1 1 1

Traffic restart

0 0 0 1

Traffic restart allowed

0 0 1 0

Traffic restart waiting

1 0 0 0

Signaling
-
data
-
link
-
connection

0 0 0 1

Signaling
-
data
-
link
-
connection
-
order

0 0 1 0

Connection
-
successful

0 0 1 1

Connection
-
not
-
successful

0 1 0 0

Connection
-
not
-
possible

1 0 1 0

MTP user flow control

0 0 0 1

User part unavailable


Table 2.3 Heading Codes H0 and H1



31

(RSM) messages are sent every

T10 and upon receiving one, signaling node
compares the status in the message to the current status of the route. It they are the same,
no action is taken, otherwise TFA, TFP, or TFR may be sent as appropriate.

5.

Transfer
-
controlled procedure

Transfer
-
con
trolled (TFC) messages (Figure 2.11d) are used to prevent the adjacent
nodes from sending traffic below or equal to a specified priority in order to prevent
message discard. The priority set in the TFC message refers to the congestion status. It
ranges be
tween 0 and 3 with zero indicating no congestion. Transmit buffers at the nodes
are designed in such a way that they go through different levels of congestion as they are
filled with messages (Figure 2.13). Congestion status is set to zero if the buffer
occupancy is below engineered normal level. As the buffer fills up, congestion status
reflects the value of the highest congestion onset level crossed (i.e., if the current buffer
occupancy were between
n

and
n + 1

congestion onset thresholds, congestion
status
would be
n
.) Congestion abatement thresholds associated with each level are set below
the corresponding onset level, and, for
n
= 1, above normal occupancy level. Message
discard
n

thresholds are located before the
n + 1

congestion onset threshold
, to achieve
greater congestion control. As congestion abates and buffer use decreases, congestion
status changes in the opposite direction. With each
n

congestion abatement threshold
crossed, status value becomes
n
-
1
. Actual threshold values are implem
entation
dependant [Ibid.].

6.

Signaling
-
route
-
set
-
congestion
-
test procedure



32


Signaling points use signaling
-
route
-
set
-
congestion
-
test messages to determine
the current congestion status of the link in order to decide whether messages with a given
priori
ty will be delivered to the specified destination.




Figure 2.13 Receive Buffer Congestion Thresholds

2.2.2

Signaling Connection Control Part

Signaling Connection Control Part (SCCP) enables signaling points to perform
database transactions. Each SP em
ployed in the national SS7 network, including STPs
that do not support any Level 4 capabilities, should terminate this part. It provides SPs
with ability to perform end
-
to
-
end signaling and allows STPs to do Global Title
Translation (GTT).

SCCP is define
d to perform both connection
-
oriented and connectionless services
divided into four protocol classes:


Class 0: basic connectionless

Class 1: sequenced (MTP) connectionless

Class 2: basic connection oriented

Class 3: flow
-
controlled connection oriented


Cu
rrently only classes 0 and 1 (connectionless services) have been implemented
by most equipment vendors.

Connection
-
oriented services, if implemented, would
require set up and tear down of a connection. Parts of this process had been
demonstrated during t
he discussion of primitives (Section 2.2.1.1).

L1 congestion
onset

L1 message
discard

L2 congestion
onset

L3 congestion
onset

L2 message
discard

L3 message
discard

input

L3 congestion
abatement

L2 congestion
abatement

L1 congestion
abatement

Normal

occupancy



33


Class 0 is used for pure connectionless transfer of messages when the originator
does not care whether messages are delivered in sequence or not, since each message is
independent of its predecessors. Class
1 offers an option of requesting that messages
coming from the same source be delivered in sequence, in which case all messages
corresponding to the same stream would be assigned the same SLS field value. When
SCCP messages of class 1 are routed through t
he network, every node should ensure that
the sequence of messages is maintained, barring the unexpected occurrences of errors,
network failures, or congestion.


A number of different messages are defined for use in various SCCP procedures.
The general

format for all SCCP messages is shown in Figure 2.14. Messages are
divided into groups based on which class of service (connection
-
oriented or
connectionless) uses the messages. For example Connection Request (CR) or Release
Complete (RLC) are only used

with classes 2 and 3. To transfer data between users in
class 0 and 1 services, SCCP uses Unitdata (UDT), Extended Unitdata (XUDT), and
Long Unitdata (LUDT) messages. Since SCCP uses underlying MTP layers to transfer
the data, and MTP may discard messag
es under certain conditions, if the user wants to be
notified about messages getting dropped with the reason for the discard, it would have to
set “Return Option” in the primitive for SCCP. In this case Unitdata Service (UDTS) (or
Extended or Long Unitdat
a Service) messages would be returned when discard occurs.



UDT and UDTS messages contain the routing label, three pointers and parameters
as shown in Figure 2.15. Message type is “0000 1001” for UDT and “0000 1010” for
UDTS. Protocol class in UDT can
be 0 or 1. Called Party Address (CPA) and Calling
Party Address (CgPA) both have the same general format except that CgPA may contain



34














































Figure 2.14 General SCCP Message Format [6]

Routing Label

Message Type Code

Mandatory Parameter A

Mandatory Parameter F

Pointer to Parameter M

Length Indic
ator of Parameter M



Parameter M



Length Indicator of Parameter P



Parameter P

Parameter Name = X


Length Indicator of Parameter X



Parameter X



Parameter Name = Z


Length Indicator of Parameter Z



Parameter Z

End of Optional Parameters

Option
al

part

Mandatory
variable

part

Mandatory
fixed

part

1

2

3

4

5

6

0

7



Pointer to Parameter P

Pointer to Optional Part





35





















a. Unit
data Message






b.

Unitdata Service Message


Figure 2.15 SCCP Connectionless Service Message Formats














Data

Calling Party
Address

Called Party
Address


Message
Type

Return
Cause

1

1

3 or greater

2 or greater

2
-
252

Data

Calling Party
Address

Called Party
Address


Message
Type

Protocol
Class

1

1

3 or greater

2 or greater

2
-
2
52

Standard
Indicator

Routing
Indicator

Global Title Indicator

Point Code
Indicator

SSN
Indicator

1

1

1

1

4

Subsystem Number


Signaling Point Code

0

1

2

7

6
1

5

4

3
1

Global Title

Address Indicator


Address

0

1

2

7

6
1

5

4

3
1

O
ctet 1

Octet 2

Octet n



octets



36

fewer fields. Address Indication Octet contains directions on which addressing method
should be used for routing:



Set Subsystem Number (S
SN) Indicator means SSN is included in Address Field



Set PC Indicator means PC is included in Address Field



Global Title Indicator can assume three states

0000 no Global Title


0001 GT includes translation type, numbering plan and encoding scheme


0010
GT includes translation type

When GT is used SSN is set to “0000 0000” before the translation.



Routing Indicator 0 means routing should be done using GTT, 1 means routing
should be done using DPC and SSN



Standard indicates whether national or international

addressing format is used

An example of messages, with corresponding parameters, exchanged between entities in
the GTT process is shown in Figure 2.16.







Figure 2.16 GTT Translation Example

Transaction Capabilities Application Part (TCAP) primarily u
ses SCCP as described in
section 2.2.4.

SSP

STP

SCP

X

Y

Z

OPC = X

DPC = Y

GT routing

AIO: 1000 1001

GT = 800
-
Nxx
-
Nxxx

SSN = 0

CgPA

= X

TCAP: 800 translation


OPC = Y

DPC = Z

SSN routing

AIO: 1100 1011

GT = 800
-
Nxx
-
Nxxx

SSN = 254

CgPA = X

TCAP: 800 translation


GTT translation

OPC = Z

DPC = X

No GTT required

TCAP: NPA
-
Nxx
-
Nxxx




37

2.2.3

ISDN User Part

ISDN User Part (ISUP) is used for call setup/teardown services. An example is
procedure illustrated in Figure 2.17.

Phone 1 is trying to establish a voice connection with Phone 2 at (516)555
-
9
721. Using
Dual Tone Multifrequency (DTMF) signaling Phone 1 supplies the address of the
destination (Phone 2) to its local office. SSP uses ISUP messages sent over the signaling

links to reserve the trunks for the call. The final destination is resolve
d step
-
by
-
step and
routed according to the dialed digits coded based on location of the phone.























Figure 2.17 Basic ISUP Call Setup

1.

SSP1 recognizes that code 516 is not its local code and sends Initial Address Message
(IA
M) to the SSP2, whose code is 516.

SSP2

SSP4

SSP1
3

SSP3

DTMF
Signaling

IAM


1


2

IAM

ACM

ACM

ANM

ANM

ring

response


3

4

5

6

7

8

Voice Trunk

Signaling Link

Phone 1

301
-
555
-
3864

Phone 2

516
-
555
-
9721



38

2.

SSP2 accepts the IAM but determines that it does not terminate trunks to the 555
phones so it sends IAM message to SSP3. As IAMs pass between SSPs, trunks that
would be used during the voice call get reserved, but at t
his point they are still
unused.

3.

SSP3 looks at the last four digits and recognizes the address of the phone; it response
with an Address Complete Message (ACM) to SSP2.

4.

SSP2 informs SSP1 that address had been resolved successfully by sending it an
ACM mess
age.

5.

At this time SSP3 alerts Phone 2 to the fact that somebody is trying to establish a call
by sending it a RING signal.

6.

When the receiver of the phone is picked up the response is generated. SSP3 notes
that the loop was complete and current is f
lowing between TIP and RING, indicating
that the Phone 2 is ready to participate in the call.

7.

SSP3 generates an Answer Message (ANM) towards SSP2.

8.

SSP2 passes the ANM to SSP1 and SSP1 completes the call setup.

9.

At this point Phone 1 and Phone 2 are connecte
d and can communicate over the voice
trunks.

2.2.4

Transaction Capabilities Application Part

Transaction Capabilities Application Part (TCAP) messages are used to invoke
calls with services, such as 800 and 888 number, Local Number Portability and other
s
ervices. The procedure is illustrated in Figure 2.18.

A local SSP receives a dialed 800 number from a phone. It transfers the number
to the STP for GTT translation. After querying its local SCP, STP determines that the


39

number belongs to a long distance
carrier. The information is returned to the local SSP,
which then transfers the number to a long distance SSP. The procedure of translating the
address repeats with long distance Signaling Points until the number is resolved to be
(516)555
-
1278. From th
at point the call gets completed in the same manner as a basic
ISUP call.













Figure 2.18 TCAP Call Setup








SSP


STP

SCP


SSP

SSP


SSP


STP

SCP


STP

SCP


STP

Local Carrier

Local Carrier

Long Distance

800
-
555
-
4400

800
-
555
-
4400

800
-
555
-
4400

516
-
555
-
1278

ri
ng

Voice trunks

Signaling links

Long distance carrier

516
-
555
-
1278

800
-
555
-
3480



40


3.0

SS7 Over ATM

A boost in demand for SS7 network resources resulted from the addition of
various AIN services, and an increase i
n the number of customers. This prompted the
development of a high
-
speed interface that can handle the extra SS7 load. Asynchronous
Transfer Mode (ATM) was chosen to be this interface. SS7 was adapted to function in
the new ATM environment.

To allow

SS7 to be transmitted over ATM without losing any of the functions
built into the signaling system required adding an intermediate Signaling ATM
Adaptation Layer (SAAL) between ATM layer and MTP layer 3 to emulate MTP 2.

3.1

SS7 over High Speed Links Prot
ocol Details

Since ATM layers used to substitute MTP Levels 1 and 2 were designed for a
packet switched connection oriented network, in contrast to connectionless narrowband
SS7 network, provisions had to be made to adopt the signaling points to a new mode

of
operation without losing any of the functions stipulated by the requirements for SS7. The
new protocol stack (Figure 3.1) introduces layers and sublayers that take on the tasks
previously performed by the eliminated MTP levels.

The User Parts laye
r, independent of underlying layers of the protocol stack,
communicates with MTP 3. Since MTP3 did not undergo any major changes to
accommodate the high
-
speed links (HSLs), the migration from low speed links to HSLs
is transparent to the User Parts. Sinc
e MTP2/MTP1 were replaced by SAAL/ATM/T1,
the latter will be the main focus of discussion here.





41









Figure 3.1 High Speed Links SS7 Protocol Stack

3.1.1

AAL5 Common Part

ATM Adaptation Layer 5 (AAL5) Common Part (CP) is divided into two parts.
AAL5

Segmentation and Reassembly (SAR) sublayer is responsible for fragmenting the
Protocol Data Units (PDUs) received from upper layer (Figure 3.2) into ATM cells and
for reassembling received cells back into complete PDUs. An ATM cell contains 48
bytes of d
ata and 5 bytes of header information. When a PDU is delivered to the SAR
layer, it is divided into 48
-
byte
-
long chunks that are then placed in the payloads of the
cells. Next, the ATM layer attaches a header with routing information and the complete
cel
l is passed on to the Physical layer for transmission.

AAL5 Common Part Convergence Sublayer (AAL5 CPCS) is responsible for
error detection. Since the length of the data unit used with high speed links could be
greater than in LSL SS7, 32 bits are used
for error detection instead of 16. CPCS PDU
also contains information about the length of data field, User to User (UU) information
SSCF

SSCOP

AAL5 CPCS

AAL5 SAR

Layer Management

SSCS

AAL5 CP


User Parts


MTP Level 3


SAAL


ATM


T1



42

field, and Pad of variable length used to ensure that the PDU is a multiple of 48 bytes, for
ATM cell payload insertion.










Figure 3.2 AAL5 Common Part Data Units Formats

3.1.2

Service
-
Specific Connection Oriented Protocol


Service
-
Specific Connection Oriented Protocol (SSCOP) in addition to
performing regular functions of MTP Level 2, such as sequence integrity, error

correction, flow control and transfer of user data, performs connection control function
and reports errors to Layer Management (LM).

Sequence integrity is kept at the SSCOP sublayer by means of a sequence number
field N(S) in the Sequenced Data PDUs (F
igure 3.3) and POLL and STAT/USTAT
PDUs. Before being transmitted, each PDU is assigned a 24
-
bit
-
long sequence number,
which is then used by the destination to either acknowledge the receipt of a valid PDU or
to inform the sender that the received PDU con
tained an error. The source verifies the
delivery of PDUs by generating POLLs to the destination at regular intervals of
CRC
-

32

Length

Rsvd

UU

Pad

Data

Information
Field

Information
Field

Information
Field



Information
Field

Header

4

2

1

1

0
-
47

1
-
65,535

AAL5 CPCS PDU

AAL5 CPCS SDU

48

48

48

48

5

AAL5 CPCS

AAL5 SAR

ATM



43

Timer_POLL. When the destination receives a POLL, it issues a STAT in response.
STAT provides the sender with the sequence number of

the PDU the receiver is
expecting to receive next, VR(R). VR(R) indicates that every PDU with a sequence
number smaller than or equal to VR(R) has been received successfully. In case PDUs are
received out of order (one or more PDU in a sequence is missi
ng), the destination issues
an unsolicited STAT (USTAT) containing the sequence numbers of the missing PDUs.
This mechanism provides for error correction using selective retransmissions and unlike
in MTP2, only the missing PDUs will be retransmitted inste
ad of all the PDUs with
sequence numbers greater than that of the negatively acknowledged PDU.






Figure 3.3 Sequenced Data (SD) PDU [7]

SSCOP performs flow control on a peer
-
to
-
peer basis using a credit mechanism.
The receiver provides the transmitter

with a window size indicating the number of PDUs
that can be transmitted without waiting for an acknowledgement in the form of a range of
sequence numbers that will be accepted by the receiver. The mechanism works as
follows:

The receiver keeps three cou
nters [8]:

1.

VR(R) is next expected sequence number (all PDUs with smaller sequence
numbers have been received and acknowledged)

PL

Rsvd

N(S)

Pad (0
-
3 octets)

Dat a

PDU t ype

1

2

3

4

8 7 6 5
4 3 2 1
1



44

2.

VR(H) is the highest expected sequence number (the highest sequence number
received so far is VR(H)
-
1)

3.

VR(MR) is the highest sequ
ence number that will be accepted by the receiver

The granted credit thus is [VR(R), VR(MR)
-
1]; any PDU with a sequence number
outside of this range will be discarded.

The transmitter also keeps three counters [Ibid.]:

1.

VT(A), which is equal to the VR(R),

contained in a STAT/USTAT

2.

VT(S) is the next sequence number to be transmitted

3.

VT(MS) is the highest sequence number that can be transmitted based on the
granted credit

If VT(S) is equal to the VT(MS), then the credit is zero and transmitter cannot
send an
y PDUs to the destination. When this situation occurs, Timer_NO
-
CREDIT is
started at the transmitter, and if no credit is granted before the timer expires, SSCOP
informs LM of the event and the link is subsequently taken out of service.

An aspect of SSCOP

is that it is connection
-
oriented. Upon request from an
upper layer protocol, SSCOP exchanges PDUs with the peer entity to establish a
connection. PDUs used for this task are BGN, BGAK and BGREJ. The last one is used
if for some reason the peer is not
able to start the connection. In this case the originator
keeps transmitting BGN PDUs until it either receives BGAK or the timer T2, started with
initial connection establishment attempt, expires. BGNs are sent in groups of size set by
the MaxCC paramete
r; BGNs within a group are sent every Timer_CC; T1 intervals
separate each group. Since all these parameters are user configurable, SSCOP will




45

transmit groups of BGN PDUs before giving up. The connection gets established when
the

peer issues an acknowledgment, and it stays up until one of the entities request a
termination. When there is no actual exchange of data, POLL/STAT transfers are used as
“heart beats” to keep the connection alive. END and ENDAK PDUs are exchanged to
tea
r down the connection. These procedures involve additional interaction inside the
entity both between SAAL sublayers, and between SAAL and the higher layer, with
details provided in subsequent sections.

3.1.3

Service
-
Specific Coordination Function

S
ervice
-
Specific Coordination Function (SSCF) participates in procedures of two
types: with no peer
-
to
-
peer messages and with peer
-
to
-
peer messages. In the former case,
SSCF maps SSCOP signals into primitives that can be understood by MTP level 3,
provides

links status information to both the SSCOP and MTP3 layers, does flow control
within the node based on the four congestion levels defined in MTP 3 section. SSCF also
provides LM with information concerning the current status of SSCOP connection and
parti
cipates in the failure detection procedure by informing LM about the number of
errored PDUs. Peer
-
to
-
peer functions include emulation of the LSSU functionality of
MTP2. (For example: out of service, processor outage, normal/emergence alignment,
etc.)

A
nother procedure where SSCF plays a key role is alignment. SSCF participates
in the procedure by informing local LM of the detected errored PDUs and by providing
the peer with the status of the local entity. The message flow, including the details on
exc
hanged primitives and active timers, used during alignment is provided in Figure 3.4.



46


Alignment gets initiated by an MTP 3 in a request type primitive [9, 68]. In
response SSCF issues a PDU to its peer and then progresses through alignment states
until t
he “In Service” state is reached. In the process, SSCF cycles through the “Proving”
state that requires close cooperation with the LM. Similar to the LSL MTP 2 procedure,
SSCF generates a number of proving messages, and if the error monitor located at LM

does not indicate excessive number of detected errors, a primitive gets issued to LM
requesting the end of proving and a PDU is transmitted to the SSCF peer informing it of
the success of the procedure. A positive acknowledgment of this PDU places SSCF i
n
service.

3.1.4

Layer Management

SSCS relies on Layer Management to provide support for error monitoring,
measurements, processor outage, and link quality determination during alignment and
link operation [10]. LM makes decision about successful or

failed alignment basing it on
the applicable error
-
monitoring tools. LM processes information supplied by the SSCOP
and SSCF about the detection of errored PDUs and the status of SSCOP connection and
then signals of the appropriate action to be taken. I
n addition to participating in the
alignment procedure described in 3.1.2 and illustrated in Figure 3.4, LM has the authority
to initiate emergency alignment, in which case proving period does not take place. This
option exists in order to prevent link os
cillation. Upon bringing the link into service LM
starts a timer and if the same link goes out of service before the timer expires, emergency
alignment is used.

Layer Management keeps track of statistics pertaining to each link: time in
service, number
of failures, presence of congestion, and number of congestion threshold


47

crossings, and so on [9, 71]. This information is vital for OAM functions, and smooth
performance of SSCS.




































Figure 3.4 Signal, Primitive and

PDU Exchange for SSCOP Connection

BGN

BGAK

MTP

SSCF

SSCOP

SSCOP

SSCF

MTP

Start Alignment

Proving

Aligned Ready

Entity A

Entity B

AAL
-
START.req

AA
-
ESTABLISH.req

AA
-
ESTABLISH.i nd

AA
-
RELEASE.res

BGREJ

AA
-
RELEASE.i nd

MAAL
-
REPORT.ind

AA
-
ESTABLISH.req

T1 ex
pires

BGN

AA
-
ESTABLISH.i nd

AA
-
ESTABLISH.res

AA
-
ESTABLISH.conf

MAAL
-
REPORT.ind

MAAL
-
PROVING.ind

AA
-
DATA.req

SD

AA
-
DATA.i nd

T3 expires,
C1>0

POLL

AA
-
DATA.req

SD

AA
-
DATA.i nd

T3 expires,
C1>0

AA
-
DATA.req

SD

AA
-
DATA.i nd

T3 expires,
C1>0

Time
r_POLL

POLL

STAT

Restart

Timer_NO
-
RESPONCE



AA
-
DATA.req

SD

AA
-
DATA.i nd

T3 expires,
C1=0

AAL
-
IN_SERVICE.i nd

MAAL
-
REPORT.ind

MAAL
-
STOP_PROVING.ind

SD

AA
-
DATA.i nd

AAL
-
IN_SERVICE.i nd

MAAL
-
REPORT.ind

I n Servi ce

Data

MAAL
-
STOP.req

AA
-
RELEAS
E.req

AA
-
RELEASE.conf

AA
-
RELEASE.i nd

MAAL
-
REPORT.ind

END

ENDAK


LM


LM

Connection

Tear Down



48

4.0

Testing Methods


Most products require testing before they are made available to the customer. It is
especially true for any electronic equipment, above all one that supports vitally important
communication networks
. Before any telephony equipment is deployed into the field, it
has to be thoroughly tested, assuring both the ability of the given piece of equipment to
perform its task reliably and accurately and its ability to work with equipment already in
the field.



Each SS7 node, of the types described in details in the previous chapters, is
subject to rigorous testing before being deployed in a Central Office. The goal of testing