A Management for the Deployment of Presence Service Using A Dynamic Routing Algoritm in the

IMS nodes

Jae-Hyoung Cho Information Telecommunication Lab,

Dept. of Electrical and Electronics Engineering Korea University of Technology and Education

Cheon Ahn, Korea tlsdl2@kut.ac.kr

Jae-Oh Lee Information Telecommunication Lab,

Dept. of Electrical and Electronics Engineering Korea University of Technology and Education

Cheon Ahn, Korea jolee@kut.ac.kr

Abstract— According to increasing the network traffic in the IMS, the role of Network Management System (NMS) is very important because of limited network resources. It can perform two kinds of routing ways with the capability of static or dynamic routing. A dynamic routing way is more efficient than static routing one because it can make the flow of traffic changeful among nodes in the IMS. Therefore, in this paper, we suggest a management function of NMS, applying a dynamic routing algorithm for managing the CSCFs in the IMS.

Keywords-CSCF; IMS;Routing Algorithm; Presence Service;

I. INTRODUCTION The IP Multimedia Subsystem (IMS) is an architectural

framework for delivering Internet Protocol (IP) multimedia to mobile users. As it is such a framework that provides access to the content of Internet and mobile services anywhere and anytime with guaranteed Quality of Service (QoS) and manageability. The IMS is used to unify the network in order to control session. IMS is not intended to standardize applications but rather to aid the access of multimedia and voice applications from wireless/wired terminals. This is done by having a horizontal control layer that isolates the access network from the service layer. From a logical architecture perspective, services need not have their own control functions, as the control layer is a common horizontal layer. According to increasing use of the IMS, traffic of the IMS increases steadily, so the IMS need to manage network traffic related the IMS nodes. The IMS nodes are called Call Session Control Function (CSCF) that controls session establishment. If a CSCF is overloaded, it might have some problems such as traffic jam and system down. In order to solve these problems, the IMS needs to apply the routing algorithm that is able to control CSCFs. Because the traffic of the IMS is very fickle, it is desirable to make the NMS control the traffic flow among IMS nodes. Therefore we study a dynamic routing algorithm which is much more efficient than a static routing algorithm to manage the IMS nodes. In this paper, we suggest a dynamic routing algorithm to manage traffic among CSCFs. Also, we show the scenario that uses presence service, one of application services, by applying the proposed algorithm. In section 2, we

explain related works such as the structure of the IMS, presence service and routing way in the IMS. And in section 3, we show the process of dynamic routing algorithm by deploying the service of presence service.

II. RELATED WORKS

A. The Structure of the IMS The IMS has been developed to realize real-time or non-real-

time multimedia service in the mobile environment. The service object of the IMS provides synthetic multimedia services using IP protocol such as voice, video and data services. Additionally, the IMS has advantages that are easy in service developments and modifications. The IMS can also is easy to interlock with various 3rd party applications that are based on efficient session management. It enables mobile operator to support the structure based on Session Initiation Protocol (SIP) such as session management, security, mobility, QoS and charging. In the IMS, protocols for the definition of the IMS interfaces are Diameter and SIP. In the IMS network, CSCFs process their related signal. The CSCFs based on SIP implement basic functions to control multimedia session with infra-system, and are divided into P(Proxy)-CSCF, I(Interrogating)-CSCF, S(Serving)-CSCF by depending on their role. The P-CSCF is the first point of contact between the UE (User Equipment) and the IMS network. From the SIP point of view the P-CSCF is acting as an outbound/inbound SIP Proxy server. An IMS network usually includes a number of CSCFs for the sake of scalability and redundancy. The P-CSCF performs the role of either Proxy or User Agents that is defined in the IETF RFC 3261. The I-CSCF is a SIP proxy located at the edge of an administrative domain. The address of the I-CSCF is listed in the Domain Name System (DNS) records of the domain. When the P-CSCF follows SIP procedures to find the next SIP hop for a particular message the CSCF obtains the address of an I-CSCF of the destination domain. In order to discover P-CSCF, 3GPP is presenting method using Dynamic Host Configuration Protocol (DHCP) or gets address through PDP context. The I-CSCF retrieves user location information and routes the SIP request to the

appropriate S-CSCF. The S-CSCF is the central node of the signaling plane. The S-CSCF is essentially a SIP server, but it performs session control as well. In addition to the S-CSCF also acts as a SIP registrar. Also, one of the S-CSCF’s main functions is to provide SIP routing services. And profile information, authentication and related location information data of a subscriber are stored in Home Subscriber Server (HSS). NMS is used to monitor and administer network resources. Effective planning for a NMS requires that a number of network management tasks be performed. NMS should discover the network inventory, monitor the health and status of devices and provide alerts to conditions that impact system performance. NMS systems make use of various protocols for the purpose they serve. SNMP protocol allows them to simply gather the information from the various devices down the network hierarchy. NMS software is responsible for identification of a problem, the exact source(s) of the problem, and solving them. NMS systems not only are responsible for the detection of faults, but also for collecting device statistics over a period of time. A NMS may include a library of previous network statistics along with problems and solutions that were successful in the past—useful if faults recur. NMS software can then search its library for the best possible method to resolve a particular problem. Network elements or devices are managed by the NMS, so these devices are used to be called as managed devices. Device management includes Faults, Accounting, Configuration, Performance, and Security (FCAPS) management. Each of these five functions is specific to an organization, but the basic idea to manage these devices is FCAPS.

B. Presence service and Dynamic Routing Algorithm in the IMS

Presence is one of basic services that are likely to become omnipresent. On the one hand, the presence service is able to provide an extensive customized amount of information to a set of users. On the other hand, third-party services are able to read and understand presence information, so that the service provided to the user is modified (customized) according to the user’s needs and preference expressed in the presence information. Presence is the service that allows a user to be informed about the reach-ability, availability, and willingness of communication of another user. The presence service is able to indicate whether other users are online or not and if they are online, whether they are idle or busy (e.g., attending a meeting or engaged in a phone call). Additionally, the presence service allows users to give details of their communication means and capabilities (e.g., whether they have audio, video, instant messaging, etc. capabilities and in which terminal those capabilities are present).

In the presence service, each Presence User Agent (PUA) has a piece of information about the user, the presentity. The laptop knows whether the user is logged on or not, as does the desktop computer. The IMS terminal knows the user’s registration status and whether he is engaged or not in any type of communication. They can have even richer presence information, such as what time the user will be back from lunch, whether the user is available for videoconferences, or whether he only wants to receive voice calls right now. All

PUAs send their pieces of information to a Presence Agent (PA). The PA gathers all the information received and obtains a complete picture of the user’s presence. A PA can be an integral part of a Presence Server (PS). A PS is a functional entity that acts as either a PA or as a proxy server for SUBSCRIBE requests.

In typical Simple Network Management Protocol (SNMP) use, one or more administrative computers called managers have the task of monitoring or managing a group of hosts or devices on a computer network. Each managed system executes, at all times, a software component called an agent which reports information via SNMP to the manager. Essentially, SNMP agents expose management data on the managed systems as variables. The protocol also permits active management tasks, such as modifying and applying a new configuration through remote modification of these variables. The variables accessible via SNMP are organized in hierarchies. These hierarchies, and other metadata, are described by Management Information Base (MIB).

Figure 1 shows dynamic SIP message flow using NMS. When there are many S-CSCF (a, b, c) in the NMS, NMS notifies routing address to I-CSCF and selects suitable S-CSCF (a or b or c). The NMS includes information such as performance of CSCF, routing algorithm and so on. There are many kinds of the way about information of the performance but in this paper, we consider the following information:

-System down

-CPU Utilization (Tc)

-Memory Utilization (Tm)

-Throughput: coefficient of utilization of process

-Response Time

Using the above mentioned information, the manager can select more efficient and better CSCF using the dynamic routing algorithm.

Figure 1. The Structure of the NMS in the IMS

Following are results of a study on the “Human Resource Development Center for Economic region Leading Industry” Project, Supported by the Ministry of Education, Science & Technology(MEST) and the National Research Foundation of Korea(NRF)

III. IMPLEMENTATION

A. Presence Service and CSCF Algorithm in the IMS Figure 2 represents the structure of the Presence Service in detail. When SIP message accesses from the IMS, the roles of Presence Server (PS) accomplish to register presence information and upload registering information to ResourceList.

Pidfparser defines related to information (e.g., presence, location and status of device, user information) of user and entities using PIDF. ResouceList is implemented possible to store resource information of user and entities. Later on, it will be extended to interact with HSS for AAA and QoS. If the user that sends SUBSCRIBE message or PUBLISH message for registration in ResouceList is existence in ResourceList, PM re-uploads partly modified presence information or creates PIDF document about presentity. Presence Manager (PM) makes suitable SIP messages to process SIP messages from PS.

Figure 2. The Structure of the Presence Service

Presence Enabler (PE) is able to process SIP message

(PUBLISH, SUBSCRIBE, NOTIFY). The HSS processes related to authentication, authorization and services via Location Information Request (LIR) and Location Information Answer (LIA). PUBLISH or SUBSCRIBE message is used to upload application services or user information (e.g., status, location) to PE. It doesn’t ingenerate an event message but upload its changed information to a server. SUBSCRIBE message is ingenerated specific events, and receives NOTIFY message as a response message. It contains URIs of the servers and users, authentication information, connecting information, expires, created time and home directory of receiving user. NOTIFY message is used to response about specific events (SUBSCRIBE). It includes the URI of the receiving user and contact information, related to authentication information (black list or white list), expires, created time and home directory information. Authentication information of UA can know through verifying initial Filter Criteria (iFC) that indicates whether users are able to use services or not.

Figure 3 shows dynamic routing algorithm, using information of MIB such as Tc, Tm and response time and so on.

Figure 3. Dynamic Routing Algorithm of CSCF

A NMS periodically gets the information of performance

from CSCF agent. Basically SNMP based manager can get the performance related information from the agent periodically to perform the monitoring function. So the period value for monitoring can depend on the performance of CSCF’s agent. So, NMS administrator should condignly adjust monitoring period value. When there are numerous CSCFs, it is possible to change the routing information in CSCF’s agent using SNMP Set message by analyzing the monitoring information related to CSCFs. Additionally, when the utilization(CPU and memory) is exceeded to limited value (Tm, Tc), CSCF’s agent must notify manager of Trap message. And then the manager can store the received Trap message in Log_DB. In order to analyze the performance information of CSCFs, the manager must perform the following four steps:

• Step 1: The manager obtains a list of CSCFs by using the management information from each CSCF and initializes/updates the agent information after performing SNMP Get-Operation to agents. The information includes CSCF ID, Type, IP address and state information of CSCF which are shown in Table 1. The CSCF_ID field indicates the ID of CSCF and the CSCF_Class field denotes the type of CSCF (e.g., P-CSCF, I-CSCF, and S-CSCF). The CSCF_Name field is the name of CSCF and the IP field indicates the IP address of CSCF. The Port_Num field means the port number of CSCF and the Status field represents the operation status of CSCF. The SysUpTime field is represented as the time that goes by an activated state of CSCF. The Description field means the characteristics of CSCF.

Table 1 MIB of the CSCF

• Step2: The manager extracts the performance information from the agent. In this processing, the manager should go through authentication. If the manager successes with authentication, it can get the performance information from agent. But if the manager fails to authentication, it receives the Trap message from the agent. The manager looks up the information about a present running process by using

hrSWRunPerfCPU and hrSWRunPerfMem field value. However, the manager needs whole of CPU utilization rate and memory utilization rate of CSCFs for the dynamic routing of SIP message. Therefore, the structure of performance information in the database of manager is shown in Table 2. The PerfCPU field is the information about hrSWRunPerfCPU and the PerfMem field indicates the information about hrSWRunPerfMemThe CPU_Util field means a CPU utilization rate of CSCF and the Mem_Util field denotes a memory utilization rate of CSCF. The Throughput field indicates the information about a processing rate and the Delaytime field means the information about transmission delay time.

Table 2 Performance Information

• Step3: In this step, the manager analyzes the performance

of each CSCF. After finishing the step 2, the manager can preserve the performance information about CSCFs in the management database. The manager decides the priority order for SIP routing among CSCFs. And then, it renews and modifies the priority information in the management database which is depicted in Table 3. The Priority field indicates the information about routing priority order and the OperStatus field means state information (e.g., active, standby).

Table 3 Priority-based Routing Information

• Step4: The manager selects the best CSCF for performing

a dynamic routing among CSCFs. After finishing the step 3, the manager decides a CSCF with the highest priority field and configures the best CSCF using Set-Request message. Additionally, if the CSCF exceeds the defined value of agent's threshold (Tm, Tc), it notifies the manager of the Trap message that is stored into Log_Db. The information structure of Log_Db is shown in Table 4. The Log_ID field means ID information of the Trap message and the TrapType field indicates the type information of Trap message. The TrapInfo field denotes the detail information about Trap message and Timestamp field means the information about the creation date and the creation time of Trap message.

Table 4 Trap Information

B. The Implementation of CSCF routing algorithm using Presence Service in the IMS

Figure 4 presents the GUI of NMS manager. To get the

performance information, the administrator should add the IP address of the CSCF’s agent. SNMP OID is the unique identifier for collecting performance information. CPU Utilization shows a coefficient of utilization about CPU, and Memory Utilization shows a coefficient of utilization about memory.

Figure 4. The GUI of NMS

The information of CSCF includes various information such

as CSCF Agent’s IP, sysName, sysDesc (H/W and S/W information), sysUpDown(server operation), hrSystemUptime(recent initiation time), CPUUtilization, Memory_Utilization and so on.

IV. CONCLUSIONS In this paper, we suggest the dynamic routing algorithm to manage network traffic among CSCFs using SNMP. We show the process of dynamic routing algorithm by deploying the service of presence service. Later, we will create the enhanced algorithm which is combined in the management functions such as Accounting, Configuration, and Security. Additionally, we should evaluate the performance in order to optimize the dynamic routing algorithm.

REFERENCES

[1] Gonzalo Camarillo, Miguel A. Garcia-Martin, "The 3G IP Multimedia Subsystem(IMS) merging the internet and the cellular worlds", Wiley, 2006.

[2] 3GPP, "IP Multimedia (IM) Subsystem Cx and Dx Interfaces; Signaling flows and message contents. TS 29.228, 3rd Generation Partnership Project (3GPP)", June 2005.

[3] RFC 3859, “A Presence Event Package for Session Initiation Protocol (SIP)”, August, 2004.

[4] Jae-Oh Lee, et al, "JNMWare: Java-based Network Management Platform", APNOMS, December 2000.

[5] Ji-Hyun Hwang, Jae-Hyoung Cho, Jae-Oh Lee, "A Routing Management among CSCFs Using Management Technology ", LNCS 5787 ,September 2009.

[6] Jain SIP API 1.2 specification, 2006-12. [7] Jae-Hyoung Cho, Jae-Oh Lee, "An Implementation of the SDP Using

Common Service Enablers", LNCS 5787, September 2009. [8] OMA, ”Presence_SIMPLE-V1_1”, January, 2008. [9] Jae-Hyoung Cho, Jae-Oh Lee, “A Dynamic Routing Algorithm for

Management of the IMS Nodes UIsing SNMP”, KICS Vol.36, 2011.03