Intelligent Networks

Dr. Eng. Amr T. Abdel-Hamid

NETW 903

Winter 2007

Netw

orks &

Services

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Intelligent Networks

a SSP communicating with an SCP to retrieve information about processing a phone call.

triggered in different ways, but most often occurs in response to dialing phone numbers that have special significance; such as:1-800, 19000,…

The communication between the SSP and the SCP takes place over the SS7 network using the TCAP layer of SS7.

does not happen for every call but only for those that require IN services.

SSP

SCP

Response

Query

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Intelligent Networks

The early implementations of IN were based on a database performing number translation

IN implementations cover a more extensive set of services from time of day routing plans, follow-me services, pre-paid mobile services (wireless intelligent

networks), calling card services, to advanced network-based call centre.

The basic aim of IN is to decouple the service logic from the control of the switch fabric. Defined in Q.1201 as ‘‘integrated service creation and implementation by means of the modularized reusable network functions’’.

The business aim of IN is the removal of a dependency on switch manufacturers for the provision of new services.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Service Data and Logic

Service data is the information needed to process a call or a requested feature. Information such as Called Party Number, Routing Number, and Carrier are examples of service data.

Service logic is the decision-making algorithms implemented in software that determine how a service is processed. The service logic acts on service data in making these decisions and directing call processing to create the proper connections, perform billing, provide interaction to the subscriber, and so forth.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Service Data and Logic Until IN capabilities were introduced in the 1980s, the

service data for the PSTN resided within the telephone switches throughout the network.

The expansion of telecom services created several issues with this architecture, including the following: Increased storage demands Maintaining synchronization of replicated data Administrative overhead

One of the benefits of the IN is centralizing service data in a small number of nodes.

This alleviates the overhead of administering data at each switching node and reduces the problem of data synchronization to a much smaller number of nodes.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Service Distribution and Centralization

SSP

SCP

Response

Query

Adjunct

Response

Query

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Service Distribution and Centralization

IN redistributes the service data and logic to other platforms outside of the switch, leaving the switch to perform basic call processing. The SCP and Adjunct are two new nodes that IN has introduced for hosting service data and logic.

The SCP usually serves a large number of SSPs and maintains a large amount of data. It is typically implemented on larger-scale hardware to meet these needs.

The Adjunct is a much smaller platform that normally serves one or possibly a few local offices and is often co-located with the switch.

Adjuncts characteristically use generic hardware platforms, such as a network server or even personal computers equipped with an Ethernet interface card or SS7 interface cards.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

IN Services There have been two primary

drivers for IN services: regulatory mandates and revenue-generating features.

LNP is an example of regulatory mandates that have greatly expanded the use of IN.

Time Of Day (TOD) Routing, and Private Virtual (PVN) Networking provide solutions for everyday business needs are revenue generating services providers.

In Europe, Intelligent Network Application Part (INAP), developed by the ETSI standards body, interfaces with ITU TCAP for delivering IN information between nodes.

In North America, IN/1 and AIN, developed by Telcordia, interface with ANSI TCAP to deliver the equivalent information.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

AIN

A part of the evolution of the original IN concept. AIN is a term that is primarily used in North America to describe the evolution of the IN beyond the IN/1 phase.

AIN defines a Basic Call State Model (BCSM), which identifies the various states of call processing and the points at which IN processing can occur, Points In Call (PIC) and Detection Points (DP), respectively.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Detection Point (DP)

DPs between the various PICs represent points at which IN processing can occur. The DP detects that the call has reached a particular state,

DP is a generic term that identifies the insertion point for IN processing. More specifically, each DP is either a Trigger Detection Point (TDP) or an Event Detection Point (EDP).

Trigger Detection Point (TDP): TDP is a point at which the SSP can set triggers that execute when the TDP is encountered. The trigger represents an invocation point for an IN service. When a trigger has been subscribed for a particular TDP and the TDP is encountered, the SSP software launches a query to the SCP.

Event Detection Point (EDP): An EDP is a point at which the SCP "arms" an event at the SSP. The event is armed to request that the SCP be notified when the particular EDP is reached during call processing. The SCP can then determine how the call should be further directed. For example, the SCP might want to be notified before a user is connected to a "busy" treatment so that a call attempt can be made to another number without the phone user being aware that a busy signal has been encountered.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Detection Point (DP)

PIC is defined as call processing state. A set of entry events define the transitional actions that

constitute entering into a PIC. Exit events mark the completion of processing by the current

PIC. Within each PIC, the switch software performs call processing

for that stage of the call in the same processing procedure that existed before the introduction of IN.

PIC

DP

PIC

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Detection Point (DP)

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Detection Point (DP)

Wireline networks have agreed on the IN/AIN triggers for querying databases.

wireless networks do not necessarily support IN/AIN. The industry is looking at IS-41 and GSM protocols for

querying the LNP database. Both the IS-41 and GSM protocols are being modified

to support additional parameters for LNP. LNP has required new parameters to the ISDN User

Part (ISUP).

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Network Architecture

Service Switching Point (SSP): The SSP performs basic call processing and provides trigger and event detection points for IN processing.

Service Control Point (SCP)/ Adjunct: The SCP stores service data and executes service logic for incoming messages.

Intelligent Peripheral (IP): The Intelligent Peripheral (IP) provides specialized functions for call processing, including speech recognition, prompting for user information, and playing custom announcements.

Service Management System (SMS): Most of the IN services require the management of a significant amount of data. The SMS generally consists of databases that can communicate with IN nodes to provide initial data loading and updates.

Service Creation Environment (SCE): The SCE allows service providers and third-party vendors to create IN services.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Network Architecture

SSP

SCP

Adjunct

Intelligent Peripheral

Service Creation Environment

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Intelligent Network Conceptual Model

The ITU Intelligent Network Conceptual Model (INCM) divides the network into different "planes." Each plane shows a particular view of the components that make up the IN. The model is an abstract representation that provides a common framework for vendors and service providers, thereby giving IN architects and implementers a common terminology base for discussion and allowing the development of modular network components.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Network Architecture

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Intelligent Network Conceptual Model Service Plane: Represents a view of the network strictly

from the view of the service. The underlying implementation is not visible.

Global Functional Plane: A view of the common building blocks across the network that comprise service functions and how they interact with Basic Call Processing.

Distributed Functional Plane: A view of the Functional Entities (FE) that compose the IN network structure. The DFP is where the collection of SIB implementations represent real actions in the course of processing actual service functions. The formal term used to describe these functions is Functional Entity Actions (FEA). For example, this plane describes BCSM within the CCF.

Physical Plane: Represents the physical view of the equipment and protocols that implement the FE that are described in the DFP.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Intelligent Network Conceptual Model SSP

Call Control Function (CCF): Provides call processing and switch-based feature control. This includes the setup, maintenance, and takedown of calls in the switching matrix and the local features that are associated with those calls.

Call Control Agent Function (CCAF): Provides users with access to the network.

Service Switching Function (SSF): Provides cross-functional processing between the CCF and SCF, such as the detection of trigger points for IN processing.

SCP Service Control Function (SCF): Directs call processing

based on Service Logic Programs. Service Data Function (SDF): Provides service-related

customer and network data for access by the SCF during the execution of service logic.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Intelligent Network Conceptual Model SMS

Service Management Function (SMF): Manages the provisioning and deployment of IN services and service-related data.

Service Management Access Function (SMAF): Provides the interface for accessing the SMF.

SCEService Creation Environment Function (SCEF):

Provides for the creation and validation of new services. Generates the logic used by the SCF.

IPSpecialized Resource Function (SRF): Provides

resources for end-user interactions, such as recorded announcements and user input via keypads, voice recognition, and so forth.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

SS7 in the Converged World

The "Converged World" of Next Generation Networks (NGNs) brings with it the promise of voice, video, and data over a single broadband network.

This transition from the traditional circuit-switched networks to packet-switched networks has been underway for many years, and Voice over IP (VoIP) is now leading the transition.

The immediate benefits of NGNs are decreased cost of infrastructure and improved ease of management. Longer-term benefits include the ability to rapidly deploy new services.

Switched Circuit Network (SCN)

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

NGN Architecture

Media Gateway (MG) handles the media, or bearer, interface. It converts media from the format used in one network to the format required in another network. For example, it can terminate the TDM trunks from the PSTN, packetize and optionally compress the audio signals, and then deliver the packets to the IP network using the Real Time Protocol (RTP).

Media Gateway controller (MGC) (also known as a Call Agent) contains the call processing. In addition, it manages the resources of the MGs that it controls.

Signaling Gateway (SG) sits at the edge of an IP network and terminates circuit-switched network signaling, such as SS7 or ISDN, from the circuit-switched network. It transports, or backhauls, this signaling to the MGC or other IP-based application endpoint.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

NGN Architecture

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Signaling Transport (SigTran) protocol Signaling Transport (SigTran)

protocol defined by RFC 2719 in the 90’s

The protocol framework identified three necessary components for the SigTran protocol stack: A set of adaptation layers that

support the primitives of telephony signaling protocols

A common signaling transport protocol that meets the requirements of transporting telephony signaling

IP network protocol

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

SCTP The Working Group began evaluating the two commonly

used transport protocols, User Datagram Protocol (UDP) and Transport Control Protocol (TCP) against these requirements.

UDP was quickly ruled out because it did not meet the basic requirements for reliable, in-order transport.

TCP met the basic requirements, it was found to have several limitations, such as: Head-of-line blocking: Because TCP delivery is strictly

sequential, a single packet loss can cause subsequent packets to also be delayed. The analysis showed that a 1% packet loss would cause 9% of the packets being delayed greater than the one-way delay time.

Timer granularity: The retransmission timer is often large (typically one second) and is not tunable.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

SCTP

TCP further limitations: Also, because of a timer granularity issue and the lack

of a built-in heartbeat mechanism, it takes a long time to detect failure (such as a network failure) in a TCP connection.

A new transport protocol, Stream Control Transmission Protocol (SCTP) was developed for transporting SCN signaling. Note that SCTP is a generic transport that can be used for other applications equally well.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

SCTP RFC 3309 is the most updated version that describes this

protocol. SCTP provides the following features:

Acknowledged error-free, nonduplicated transfer of user dataData segmentation to conform to path MTU size (dynamically

assigned)Ordered (sequential) delivery of user messages on a per "stream"

basisOption for unordered delivery of user messagesNetwork-level fault tolerance through the support of multihomingExplicit indications of application protocol in the user messageCongestion avoidance behavior, similar to TCPBundling and fragmenting of user dataProtection against blind denial of service and blind masquerade

attacksGraceful termination of associationHeartbeat mechanism, which provides continuous monitoring of

reachability

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

SCTP SCTP is a connection-oriented protocol. Each end of the connection is a SCTP endpoint. An

endpoint is defined by the SCTP transport address, which consists of one or more IP addresses and an SCTP port. The two endpoints pass state information in an initialization procedure to create an SCTP association.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

SCTP SCTP uses streams as a means of decreasing the impact of

head-of-line blocking. Streams provide the ability to send separate sequences of

ordered messages that are independent of one another.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

SCTP SCTP provides the ability to have multiple streams

within an association. Each stream provides reliable delivery of ordered messages that are independent of other streams.

Packet 2 is dropped again. However, because packets 3, 4, and 5 belong to a different stream, they can be delivered to the application without delay.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

User Adaptation (UA) Layers The User Adaptation (UA) layers encapsulate different SCN

signaling protocols for transport over an IP network using SCTP.

UA layer is unique in terms of the encapsulation because of the differences of the signaling protocols themselves, following are some common features among all UA layers: Support for seamless operation of the UA layer peers over an

IP network. Support for the primitive interface boundary of the SCN lower

layer, which the UA layer replaces. For example, M2UA supports the primitive interface boundary that MTP Level 2 supports.

Support for the management of SCTP associations. Support for asynchronous reporting of status changes to layer

management.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

User Adaptation (UA) Layers

MTP Level 3 User Adaptation (M3UA) layer is defined for the transport of SS7 User Part messages (such as ISUP, SCCP, and TUP).

SCCP User Adaptation (SUA) layer is defined for the transport of SCCP User Part messages (such as TCAP)

MTP Level 2 User Adaptation (M2UA) layer is defined for the transport of MTP Level 3 messages.

MTP Level 2 Peer Adaptation (M2PA) layer is defined for the transport of MTP Level 3 data messages over SCTP. M2PA effectively replaces MTP Level 2. It provides the ability to create an IP-based SS7 link.

The ISDN User Adaptation (IUA) layer is defined for the transport of Q.931 between an ISDN SG and a MGC.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

UA Common Terminology The UAs introduce some new terminology that did not

exist in the SS7 world.: Application Server (AS): A logical entity that serves

a specific Routing Key. An example of an Application Server is a virtual switch element that handles all call processing for a unique range of PSTN trunks. Another example is a virtual database element, handling all HLR transactions for a particular SS7 combination. The AS contains a set of one or more unique ASPs, of which one or more is normally actively processing traffic.

Application Server Process (ASP): A process instance of an Application Server. An ASP serves as an active or backup process of an Application Server (for example, part of a distributed virtual switch or database).

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

UA Example

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

MTP Level 3 User Adaptation

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

SCCP User Adaptation

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

MTP Level 2 User Adaptation

The M2UA protocol defines the layer split between MTP Level 2 and MTP Level 3. M2UA is defined by RFC 3331.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

MTP Level 2 Peer Adaptation

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Switches and Softswitches

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Decomposing SSL

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Call Flow

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

COMPUTER TELEPHONY INTEGRATION

Computer telephony integration (CTI) is technology that allows interactions on a telephone and a computer to be integrated or coordinated CTI provides the ability to integrate multiple call centers in different physical locations into a single entity

• Call information display (caller's number (ANI), number dialed (DNIS), and Screen population on answer, with or without using calling line data • Automatic dialing and computer controlled dialing (fast dial, preview, and predictive dial) • Phone control. (answer, hang up, hold, conference, etc.) • Coordinated phone and data transfers between two parties (i.e pass on the Screen pop with the call) • Call center phone control. (logging on; after-call work notification) • Advanced functions such as call routing, reporting functions, automation of desktop activities, and multi-channel blending of phone, e-mail, and web requests

Functions provided:

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

COMPUTER TELEPHONY INTEGRATION First-Party call control : • First party call control operates as if there is a direct connection between the user's computer and the phone set. • An example of this would be a modem card in a desktop computer, or a phone plugged directly into the computer. • Only the computer associated with the phone can control the all functions, by sending command directly to the phone.

Third-Party call control ( Used in call centers) : • requires a dedicated telephony server to interface between the telephone network and the computer network. • works by sending commands from a user's computer to a telephony server, which in turn controls the phone centrally. • Information about a phone call can be displayed on the corresponding computer workstation's screen while instructions to control the phone can be sent from the computer to the telephone network• The user's computer has no direct connection to the phone set and controlled by an external device.

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Softswitches in call centers

The combination of CTI and intelligent network services has created the ability for call centers to route calls across national and international boundaries transparent to the caller

In Call centers, CTI and Database Servers are used in addition to PBX/ACD

ACDS communicate with softswtiches to enhance call routing from callers outside the IP network

The use of CTI(Computer Telephone Integration) will allow the IVR system to look up the CLI (Calling Line ID) on a network database and identify the caller and provide the required service

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Presence center architecture

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Interactive voice response, or IVR, is a phone technology that allows a computer to detect voice and touch tones using a normal phone call.• Call centers use IVR systems to identify and segment callers. • The ability to identify customers allows the ability to tailor services according to the customer profile. • It also allows the option of choosing automated services.• Information can be fed to the caller allowing choices such as: wait in the queue, choose an automated service, or request a callback. • The use of CTI(Computer Telephone Integration) will allow the IVR system to look up the CLI (Calling Line ID) on a network database and identify the caller.

interactive voice response

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

Dr. A

mr

Talaat

Netw 903 Lecture 2

Serv

ices &

N

etw

ork

s

CTI message exchange