BROADBAND CONVERGENCE NETWORKING

ABSTRACT
A broadband convergence network (BcN), which enables the convergence of communications and broadcasting services, the convergence of voice, video, and data services, and the integration of wired and wireless technology in order to provide high-quality broadband multimedia services, is an implementation example of a next generation network (NGN). To rapidly provide various new application services, a simple but feature-rich control network that performs all demanded control functions required for the provision of service is indispensable. However, there are few studies on control networks for BcN or NGN, and therefore we need to carefully consider the architecture and functions of the control network. The main purpose of this paper is therefore to present a BcN control network architecture and its design methodology. The network design process includes capacity and expense calculation process of control systems and control links, which enables the estimation of the overall network building cost. A reliability analysis for the control network is performed in detail based on various system redundancy policies. The proposed design methodology is expected to be utilized in the design of BcN control networks prior to field deployment.
[attachment=4162]
[attachment=4163]


1. INTRODUCTION
Convergence is a keyword into day’s telecommunications world .The concept of convergence encompasses the convergence of voice, video, and data services, the integration of wired and wireless communications networks, and the integration of broadcasting and communication technologies.
This trend is part of the efforts made by telecommunications companies to satisfy customer demand for high-quality multi media services any where, at any time, and with any device. Next generation networks (NGN) have been developed and standardized to satisfy the demands of network operators, who plan to offer range of new services over common transport networks. An NGN is defined as packet-based networkable to provide Telecommunication services, and to make use of multiple broadband and QoS-enabled transport technologies, in which service-related functions are independent from the underlying transport-related technologies. It also enables the unfettered access of users to the networks and services of their choice, and supports generalized mobility, thereby allowing ubiquitous and consistent provision of services.
Broadband convergence network (BcN)is an implementation of this type of feature-rich NGN in Korea. Currently in Korea, network operators have been implementing BcN, which enables IP multi media subsystem (IMS)-based wired/wireless integration, the migration of public switched telephone network (PSTN) to IP networks, and broad casting/telecommunications convergence. IMS was originally developed as an architecture frame work for the delivery of IP-based multimedia service to mobile users, but is now regarded as a core system that integrates wired and wireless networks. Traditional telephone networks (PSTN) are now being migrated to IP networks. Digital switching equipment is being replaced by IP packet switching capable gate ways, such as access gateways and trunk gateways, and the SS7 system is being replaced by signaling gateways. In addition, internet protocol television (IPTV) is one of the potential broadcasting and communications-converged solutions provided through an NGN. IPTV provides digital television service, either via real-time streaming or video on demand, using Internet protocol over a packet-based network infrastructure.

To provide such wide range of new applications, a flexible and simple control network for BcN is required. The BcN control network performs all required control functions for service provisioning, including call or session control, QoS provisioning, network resource management, and access authentication. Since BcN/NGN is a brand-new concept of network architecture and its standardization has only recently begun, few studies on BcN control networks exist. For this reason, a study on BcN control network architecture and its design methodology is required. Network design and planning is multi-faceted and requires the consideration of a range of activities taking place in a network operator. A network designer should take a holistic perspective, and consider the level of system detail insight, service provisioning scenarios, and network planning strategy. The design process includes market and demand forecasting, tele traffic-related methods, economic engineering, operational research and optimization, financial and personnel limitations, a historical analysis including trends, and soon. Since legacy communications services have been provided through unique network infrastructures, different network design methods were applied to their own service networks. However, BcN provides several application services over a common transport network, and enable the easy addition of new services. Therefore, anew design methodology for a BcN environment is solicited, that aims at reducing the complexity of the design process and calculating the network resources that are additionally required to provide new services, without drastic changes in the design process.
To achieve this aim, this paper introduces a BcN control network architecture and service provisioning scenarios in Section2. A design methodology and detailed processes for the control network are presented in Section3. InSection4,we introduce an automated design system for a BcN control network that we have developed. Finally, Section5gives several implementation examples according to network building scenarios, and shows the irrespective building costs and availabilities.
2. BcN CONTROL NETWORK ARCHITECTURE

2.1 BcN CONTROL NETWORK ARCHITECTURE AND COMPONENT CONTROL SYSTEMS.

BcN is one type of realization of NGN, suitably modified for real businesses and communications environment in order to provide high-quality broadband/multimedia services converged with communication / broadcasting / Internet technology any where, at anytime, and with any device. Based on the concept of NGN, BcN has several functional platforms that are independently operated among others in order to provide legacy communications services as well as emerging new services, simply by manipulating network resources in a transport network. As shown in Fig. 1, a BcN mainly consists of an application service platform, a control network, an OAM function module, a transport network, and a module to inter-work with other networks. The BcN control network is responsible for call or session control, QoS control, network access and authentication, subscriber and service profile management, and more. The application service platform contains several application services for various application services, and alsointerfaces with third-party service providers. The OAM function module provides operation, administration, and maintenance related functions, and it also includes a charging function. The transport network, which consists of many transport systems, such as switches, IP routers, and gateways, performs service provisioning and bearer traffic delivery for service requests.
Now, let us observe the control network in detail. The control network consists of three platforms: session control platform, QoS control platform, and access control platform. The session control platform controls call or session based on E.164 or session initiation protocol-uniform resource identifier (SIP-URI) manages service level profiles, and provides media resources for communication services. This platform consists of a service information provisioning system (SIPS) a number resolution server (NRS), a media server (MS), and IP multimedia subsystem (IMS). The SIPS manages terminal mobility and the profiles of subscribers and services. The NRS is responsible for number translation between E.164andSIP-URI. The MS provides tone generation and announcement broad casting for the subscribers of communication services. The IMS controls calls or sessions between IP terminals for communication services.




Fig. 1. BcN control network architecture

The QoS control platform contains network resource management systems for providing QoS and call admission control (CAC). This includes a QoS manager (QM), an access QoS manager for wired network (AQM), an access QoS manager for wireless network (WAQM), and an access QoS manager for leased line on-demand service (LAQM). The QM is a global coordinator between IMS and the AQMs that directly control the under lying transport systems, and controls network resources in the core transport network. The AQM manages resource information in wire line access networks, and controls access systems for resource reservation. The WAQM performs the same operations as AQM for wireless access networks. The LAQM delivers are quest of resource management for leased line on-demand (LLoD) service to QM. This system also decides service priority related to route control and delivers this information to the edge control system (ECS).




Fig. 2 . Service control scenarios through BcN control network.


The access control platform mainly performs authentication and authorization-related functions to allow user terminal s to access transport networks. This platform includes an access authentication and authorization system (AAAS), an access control system (ACS), an ECS, and a terminal information provisioning system (TIPS). The AAAS controls authentication and authorization for user terminals to access transport networks. The ACS delivers resource control commands from AQM to L2/L3 switching equipment. The ECS delivers network access authentication and resource control commands from AQM to an edge router in core transport network. The TIPS manages the profiles of terminal, subscriber, and service. Note that we simply introduced the most representative systems in each platform, and that more systems may be required. In addition, we compare these control systems with the ITU-T standard in Appendix A.

2.2. SERVICE CONTROL SCENARIOS

This section introduces several representative application services and their provisioning scenarios. First, unified communication (UC) is an integrated multimedia communication service that can provide voice / video / multi-party communication service, messaging service, presence service, data sharing service, and soon. Such UC service is provided regardless of user mobility (whether the user’s service is fixed or mobile). IPTV Is a broadcasting service that can provide digital television services as real-time streaming video or video-on-demand. LLoD service enables a third-party application service provider to supply a high-quality leased line delivery service.
Now let us observe how these services can be provided through a BcN control network. Fig. 2
illustrates service control scenarios in a simplified model of a BcN control network with 11 control systems. Note that this does not show the full-set of control systems and control signaling among them, but uses simplified examples. We also do not consider the application service platform, but focus only on session control, network resource control, and network access and authentication function. For the UC service shown in Fig. 2(a), an end user (UT: user terminal) attempts to make a call to another user. When IMS charging of session control receives a call request, it asks NRS for the corresponding user’s address number and asks SIPS for the subscriber’s profile. Following this, the IMS requests QM to provide the required resources for call delivery. The AQM that receives the request for the provision of resources from QM asks AAAS, TIPS, and ACS to allow the user terminal’s access and to allocate network resources from the calling party to the called party by manipulating transport network systems. If the corresponding user is located in another network or is being controlled by another IMS, the IMS that receives the call request should deliver it to the neighboring IMS where the corresponding user is connected. For UC service requested from a mobile user or to a mobile user in a wireless network, the IMS delivers this request to WAQM, which performs wireless resource control, through QM. We assume that other functions related to session control and resource control in the wireless network are performed in control systems in the wireless network. This service control scenarios shown in Fig. 2(b). As shown in Fig. 2©, IPTV service does not require session control to be performed by the IMS, but can simply be provided by the provisioning of net worker sources manipulated by QM and AQM. The ECS functions to deliver resource control commands to an edge router located at the border of the transport network. Network access control and authentication are performed as the same operation of Fig. 2(a). Finally, LLoD is provided by the LAQM under the control of QM as shown in Fig. 2(d)

3. DESIGN METHODOLOGY OF A BcN CONTROL NETWORK

Based on the study on BcN control network architecture and service control scenarios, this section introduces a design methodology for a BcN control network. The fundamental difference in the design of a BcN control network when compared to legacy communication networks is that in legacy networks, switching systems, as the objects of design, provide a single service and have a horizontal relationship with other systems, but in a BcN control network, one control system may support one or more services, and may have certain relationship with other systems. Hence, the design of a BcN control network should take into account the relationship between the control system and the services supported by the system, and between different control systems. There might be no optimal solution for a design goal, so we need to take an engineering approach toward design. The objectives of control network design are as follows:
(i)To configure common control platforms and systems in order to support emerging application services.
(ii)To calculate the required amounts and expenses of control systems and control links based on the estimated service demands and service coverage.
(iii) To achieve target network performance and reliability while maintaining a marginal budget.

3.1.OVER ALL DESIGN PROCESS OF BcN CONTROL NETWORK

Fig. 3 illustrates an over all design process for a BcN control network. Based on BcN design strategy and service scenarios, we first estimate the amount of subscribers per service, region, and year. Since the estimation is used for the service provisioning strategy as well as for network planning, its accuracy should be maintained within permissible bounds. Estimated service demands are also closely related to the decision regarding the amount of control systems and the capacity of control links. We then determine the specifications of control systems, including supporting services, unit capacity, unit cost, availability, and soon. Based on this information, we can compute the amount and expense of the control systems, which will be explained in detail in the following section. For control page link design, we model control traffic between two control systems, since the control protocols between them are different. Calculation of the amount of control traffic and control links follows, and then the total control page link expense is obtained.
We also take network performance into account in network design. To this end, we need to specify performance metrics of the control network and their target values. Performance metrics may include end-to-end delay time, failure rates for service requests, network reliability, and soon. As an example, this paper considers network reliability. For improving network reliability, we apply a protection scheme, which implies control systems and control links duplication. Since the protection scheme results in an increase in the overall network building cost, network designers should balance
between allowable budget and network performance. Before finishing the whole design process, we need to verify whether the result reaches the target design goal. If not, we need to modify aspects of the design policy, such as service coverage, placement, and target performance value. The following section will touch on each processing detail.

3.2. CAPACITY AND EXPENSE CALCULATION PROCESS OF CONTROL SYSTEMS

This section introduces the capacity and expense calculation process of control systems. Let us first classify control systems into transaction processing systems and database processing systems. The capacity of database processing systems, such as SIPS and TIPS, is estimated based on the number of subscribers. The capacity of transaction processing systems is estimated based on the capacity to process transactions for service requests.

3.2.1.Capacity calculation process of control systems

Based on the estimation of the number of subscribers per service and region, the capacity calculation process of control systems is performed. Since control systems may support one or more services, the processing capacity (PC) of a control system is calculated by multiplying the estimated number of subscribers of services that the system supports and the busy hour call attempt(BHCA) of the services. The following equation summarizes this calculation:

where SV is the number of services processed through the considered control system and SBi is the number of subscribers for the service i. Instead of BHCA, we may use the number of transactions, which is obtained from characteristics and BHCA for services.
3.2.2. Quantity calculation process of control systems according to service coverage

As opposed to legacy networks, BcN control systems may cover a specific small region or an entire nation depending on their characteristics and processing capacity. To calculate the quantity of a control system, we need to consider the service coverage of the control system. Fig. 4 exemplifies a system quantity calculation for three levels of service coverage :nation-wide, regional, and branch. This capacity can be obtained by using Eq.(1),and is normalized by the unit capacity of the system. Due to the modular loss of system quantity, the amount of control systems can be varied according to the level of service coverage. This means that if the service coverage is divided in to many small segments, the resulting quantity of systems might be more than the actual demand. Therefore, it is required to carefully determine service coverage, which should take in to account both the demanded system quantity and network survivability, because the former might demand larger coverage but the latter might prefer many small segments. Let us calculate the quantity of control systems from the obtained capacity by considering the different levels of service coverage. First, the total quantity of control systems for branch-level service
Coverage can be calculated with the following equation:
QB
where Qrb =[PCrb/PCU], PCrb is the required processing capacity of a control system at a branch, PCU is the unit capacity of the system. R is the number of regions in a country and B is the number of branches in a region. dx e implies the smallest integer, but cannot be less than x. That is, even if the system capacity is less than one, at least one system should exist at the branch.

Fig. 4. Example of control system quantity calculation for three levels of service coverage.(a)Control system capacity (normalized by the unit capacity of control system).(b)Control system quantity calculation from(a).



3.2.3. Positioning process of control systems

After obtaining the total quantity of control systems, systems are then positioned based on the placement strategy. Placement policy is determined inconsideration of network management and operation, operational expense, and network survivability. From the viewpoint of network management and operation, and operational expenses, a centralized placement is preferred. However, in viewpoint of network survivability, a distributed placement is preferred. Therefore, it is necessary to harmonize both issues to optimize network design. Note that the placement policy does not affect the total expense of control systems (that is, capital expenditure).

3.2.4. Expense calculation process of control systems

With the total quantity of control systems, we then calculate total system expense. The basic ideais to multiply the quantity by the unit system cost for all types of systems as follows:

where ST is the number of control system types in the control network, Qi is the quantity of the i
control system, and CUSi is the unit cost of the i control system.

3.3. CAPACITY AND EXPENSE CALCULATION PROCESS OF CONTROL LINKS

This section introduces the capacity and expense calculation process of control links in the BcN Control network. A control page link in this paper implies a logical connection between them, where
Control traffic is transmitted. Thus, the capacity of a control page link implies the amount of control traffic
Flowing between control systems.

3.3.1. Capacity calculation of control links

Control links contain various types of control traffic, which varies depending on the two neighboring systems. Therefore, the page link capacity (in bps) between two control systems A and B is calculated by multiplying the number of subscribers to the service, the number of service requests per hour (BHCA), the number of requested transactions (TR) for the provision of service, and the length of the control message (ML) used to provision a service through the two systems.
The following equation summarizes this process:


The obtained page link capacity is the total amount of control traffic between two different control systems. Due to the limited capacity of control systems, there may need to be one or more systems in order to support the provision of services to subscribers. Therefore, two different systems may have several control links, rather than a single link. In addition, depending on the relationship between two systems, there may be several different page link configurations. For example, an IMS should connect with other IMS in full-mesh to provide communication services. Thus, the number of control links (NL) among IMS is , and each page link capacity (LCE) is , where W is the number of IMS in the network. In the case of AQM–AAAS, AQM–ECS, AQM–ACS, one control system in system group1 should be connected to only one system in the other system group. Thus, the number of control links between the two groups will be max[WA,WB], and a single page link capacity will be LCT/max[WA,WB]. Finally, among the other groups, two control systems should be connected in full-mesh for the provision of service so that the number of control links is WAWB, and the each page link capacity is LCT=WA/WB.

3.3.2. Expense calculation process of control links

The expense of control page link is now calculated by considering the number of control page link types(L) (that is, pairs of control system groups) in a network, the capacity for a specific page link (LCE), and the cost per unit band width (CUB). The following equation shows the total control page link cost in a control network:
CL [ ]
ARTICLEINPRESS
where LCe = LCl/Nl
L. l implies a page link group between two control system groups. LCl
T is the total capacity of page link l, NL the number of control page link l, and BU the unit band width.

3.4. CAPACITY AND EXPENSE CALCULATION CONSIDERING PROTECTION SCHEME

As we will see in detail in the following section, maintaining and improving there liability of a network is one of the most essential tasks in network design. In order to protect a network from unexpected accidents, such as fire, earthquake, catastrophic accidents, or even human error, systems and links should be duplicated. This section thus calculates the quantity of control systems and the capacity of control links in consideration of system redundancy to improve network reliability. Fig. 5(a) shows a case where by each IMS has its own dedicated backup system (1:1protection). Since IMS takes charge of call or session control for communication services, all IMS are assumed to be connected full-mesh with others, for better connectivity and fast session control. Thus, the total number of IMS is 2W, where W is the number of working IMS. When Wi system goes down, other working IMS can still communicate with the Bi. The resulting page link capacity, therefore, is
LCT = [ + + ] LCE = 2 LCE

Now, let us calculate control page link capacity among IMS with N:1 protection, where all IMS in a Network share one backup system as shown in Fig. 5(b). The total number of IMS comes to W þ 1. As all IMS share only one backup system, each IMS should have one page link to the backup system. The resulting page link capacity can be calculated using the following equation:

LCT = [ ] LCE


Another N:1 protection scheme can be applied in the IMS, where N is smaller than the number of IMS in the network, as shown in Fig. 5©. The result control page link capacity can be summarized by

LCT = [ ] LCE

where B =W/N, and the total number of IMS is W + B.
Fig. 5(d) shows the example where by two different control system groups are connected in full mesh. Each control system group may have different sharing ratios, such as A system group applying an N:1 protection scheme and B system group applying M:1. N or M may be less than or equal to the total number of A or B systems. The total page link capacity can be summarized by

LCT = (WAWB + WBBA + WABB + BABB)LCE
= [WAWB( 1+ BA/WA + BB/WB + BABB/WAWB ]LCE
where BA = WA/N and BB = WB/M, and the total control systems are WA + BA and WB + BB,
respectively. Note that in the case of IMS to non-IMS, control links among IMS should be added to the above equation, as obtained from the third case (Fig. 5©). Finally, Fig. 5(e) shows the case where by two different control system groups are connected as required. By assuming that the minimum number of links are demanded between the two groups, we can calculate the total page link capacity as follows:

LCT = (max[WA,WB] + max[WB, BA] + max[WA, BB] + max[BA, BB]) LCE.
The total number of control systems is the same as in the example of Fig. 5(d).

3.5 PERFORMANCE REQUIREMENTS OF BcN CONTROL NETWORK

Network performance metrics and QoS criterion have been mostly taken in to account in transport networks. However, there are few studies on control networks and its related performance metrics. ITU-T has defined performance metrics for call processing for voice service in terms of speed, accuracy, and dependability. Speed implies the required time interval used to perform control functions, such as connection setup or release. Accuracy implies the degree of correctness with which control functions are performed. Dependability means the degree of certainty with which control functions are performed, regardless of speed or accuracy, but with in a given observation interval.


Fig. 5. Configuration of control systems and links considering system redundancy.(a)IMS–IMSwith1:1protection.(b) IMS–IMS with N:1 protection (N = W). © IMS–IMS with N:1 protection (1<N<W). (d) Between two control systems connected by full-mesh with N:1 protection: IMS–non-IMS,QM–other systems, AQM–TIPS.(e)Between two control systems connected by only one page link with N:1 protection: AQM–AAAS,AQM–ECS,AQM–ACS.

Such performance metrics have been defined for call processing in best-effort IP networks and legacy networks, so new performance metrics should be defined for BcN control networks. BcN provides bidirectional communication services, such as UC, and unidirectional services, such as IPTV and LLoD, and therefore performance metrics and their target values should be carefully determined to satisfy service characteristics. For example, since communication services demand real-time service, speed is the most important metric, and thus connection setup delay time should be minimized, to the point where it is below the bounds of human perception. On the other hand, a managed delivery service requires an accurate and reliable connection rather than a fast connection. Nevertheless, there has been no real service provision through BcN control networks as of yet. Thus, a practical study on performance metrics that takes service characteristics in to account should be carefully performed.

3.6. RELIABILITY ANALYSIS OF BcN CONTROL NETWORK

Since a BcN control network is responsible for control functions that provide a range of application services, high reliability is required. There are two key qualities required for network reliability, which are availability and survivability. Availability refers to the probability that a system
is working at a specific point of observation, and it is represented by the ratio of working time over
total observation time. On the other hand, survivability refers to the probability that service can be provided even after a system has failed. The former is a performance metric prior to an event, and the
latter is a performance metric post-event. This section presents the network reliability issue in terms
of availability.
Availability of control system is calculated by
A =
where mean time between failures (MTBF) is the average time between system failures and mean
time to repair (MTTR) is the average amount of time following a failure for a system to be repaired.
Since service is provided through all related control systems and links between two systems, service
Availability is calculated by multiplying availabilities of all related control systems and links on the
service route. The resulting availability of a service is
AS =
where Ai is the availability of each component, such as control systems and control links.
On the other hand, if control systems and control links have backup systems and links, system availability should be re-calculated. The availability applied by protection scheme can be calculated by using the M/M/1/N+1/N+1 queuing system, where N is the number of working systems per backup system. From state transition diagram, each state probability of the system can be obtained. The system is available only when one system or none has failed. The resulting system availability is then calculated as follows:

Ai =
where (=1/MTBF) is the average failure occurrence rate and (= 1/MTTR) is the average failure
completion rate.

4. AUTOMATED DESIGN SYSTEM OF BcN CONTROL NETWORK

Based on the aforementioned design methodology for BcN control network, we have developed an automated BcN control network design system. This section briefly introduces the design system. As shown in Fig. 6, the design system consists of three main parts: data input part, design execution part, and result output part. In the data input part, the provided services are defined and their estimated subscriber numbers are input. The specifications of the control systems, including unit cost, unit capacity, supporting services, and soon is also input. Control links are also specified. Following this, the execution part runs the design system with design input parameters ,such as provided services, service coverage and positioning of each control system, and protection scheme. From the result output part, we can obtain the reliability, investment expense of control networks, and the geographical position of control systems in a map. Note that by simply adding new service information, including related control systems, this design system can produce a control network design result that provides newly deployed service.
After performing the design with input parameters, if the result does not reach the performance target , including the metric for estimated investment expense, we then change input parameters and retry to test different alternatives until we achieve the design goal.

5. EXAMPLES OF BcN CONTROL NETWORK IMPLEMENTATION

This section presents implementation examples of a BcN control network, according to various design scenarios. In the scenario, the placement of control systems is assumed to be identical to their respective service coverage. Each design alternative for the network is evaluated in terms of building cost and service availability. The design results are obtained from the automated design system of the BcN control network.

5.1 COST ANALYSIS OF BcN CONTROL NETWORK

Fig. 7 shows the total building cost of the BcN control network according to different service
coverages : case1(nation-wide),case 2 (two regions),case 3 (four regions),and case 4 (11 branches).
The figure also shows the effect of a protection scheme on the cost: no protection(1:0),dedicated protection(1:1),and shared duplication(3:1and10:1).The total costs are normalized to that of

Fig. 6. Configuration of the automated design system of a BcN control network.

and the unit cost of control links is set based on the leased line service. This analysis is based on the
estimated number of subscribers from the survey for service demand. As the service coverage is smaller, the building cost increases due to the modular loss of system capacity. System redundancy to improve network reliability also results in a higher building cost. From the perspective of building cost alone, nation-wide service coverage would be recommended. However, network reliability should be taken in to account, which may affect service coverage as well as system placement. shows how much backup systems affect the total control network building cost. The Required amount of backup systems is similar to the sharing ratio, compared with that of working systems. On the other hand, the additional control links for backup require more than the sharing ratio. This is due to the fact that control links should be connected not only between backup systems, but also between working and backup systems. Nevertheless, since the control page link cost is much smaller than that of control systems, its effect on the total network-building cost is trivial.

5.2. AVAILABILITY ANALYSIS OF BcN CONTROL NETWORK

This section presents the service and the network availability based on various protection schemes. We assume that the each system’s availability is set to 0.999, and the average repair time is
set to 2h. We only consider control systems on the route for the provisioning of service, but not control links. Table1 shows the availability per service and control network. Without system redundancy, all services are vulnerable. On the other hand, dedicated protection (1:1) provides highly reliable services. For example, the availability of IPTV service implies that it may fail for only 5min per year. If each system’s availability is improved, a policy of greater sharing can provide more reliable services.

6. CONCLUSION

This paper presented a BcN control network architecture and various services that could be provided over the network. We also presented a design methodology for a BcN control network that
is reliable and economical. Based on the automated design tool for BcN control networks, we could
obtain the building cost and availability of the network, according to alternative scenarios for network implementation. The tool is expected to provide accurate and fast network design guidelines prior to field deployment.


REFERENCES

1. “POF & Home Network Technology”, Korea POF Communication Forum.2006
2. J.-W. Lee, “BcN, Broadband Home Network & POF”, ICPOF 2008
3. ITU-T Recommendation Y.2261, PSTN/ISDN evolution to NGN; 2006.
4. Lee C-S, Knight D. Realization of the next-generation network; 2007.
5. Ministry of Information and Communication (MIC), Korea, IT 839 Strategy, 2006.
6. Robert Cohen, “The Impact of Grids on Broadband Demand”, 2007.