Abstract
The European Regulatory Framework requires National Regulatory Authorities (NRAs) to conduct market analysis for a predefined set of markets that used to be subject to ex ante regulation (due to Significant Market Power (SMP) of the incumbent network operator), or that are expected to be associated with SMP. The service under consideration in this article—Bitstream Access— is considered in Market 12 (see ERG, 2003). Depending on the results of the market analysis, NRAs can impose remedies on the SMP operator, like cost accounting, long run incremental cost (LRIC), based ex ante regulation, or other requirements. Many European NRAs foresee price control of bitstream access service (BAS). This contribution provides a cost model for BAS, which takes into account the required bandwidth of a service and QoS parameters, mainly the average delay over the corresponding bitstream access configuration. The contribution shows in the second section the basic ideas of the FL-LRIC model and especially the so-called Total Element Long Run Increment Cost model (TELRIC) and the basic aspects of BAS network architecture. The third section deduces the proper TELRIC model for BAS under QoS differentiation, mainly considering delay limits. The section introduces two applications, one based on assuring QoS under the overengineering concept, and the other on traffic separation over different queues.
TopLric Cost Models For Bitstream Access Services
LRIC constitutes the dominant costing standard, in case of SMP and ex ante price control, recommended by the European Regulatory Framework (see BNA, 2005; Hackbarth, 2007). There are basically two methodologies to design LRIC cost models: TSLRIC (Total Service LRIC), and TELRIC (Total Element LRIC) (see Courcubetis & Weber, 2003). TSLRIC model is oriented to services and is used as basis for setting fixed network charges, but it doesn’t include common costs of joint production, as they are not incremental in providing a service.
TELRIC model is oriented to network elements. As the elements are dimensioned according to all services using it, TELRIC provides that the cost of a network element used by different services is shared by the services in relation to the intensity of use that each one does of the element. TELRIC can be designed from two different perspectives, Top-Down (Figure 1a) and Bottom-Up (Figure 1b).
Figure 1. (a) Top-down approach;(b) bottom-up approach
Under Top-Down modeling, historical accounting data are taken as a starting point. It relies on the actual network architectures and configurations of a specific carrier, and (implicitly) accounts for its efficiency.
Bottom-Up approach models the network of a hypothetical operator. This efficient operator employs the best current technology, and is not constrained by decisions of the past. Therefore, it reflects an efficient cost structure relevant to the market and regulatory decisions. Hence, for regulation purposes, TELRIC model with bottom-up approach is mainly used (BNA, 2005; Brinkmann, Hackbarth, Ilic, Neu, Neumann, & Portilla, 2007; Hackbarth, Portilla, & Diaz, 2005).
The TELRIC Bottom-Up cost models require knowledge on the traffic on all network elements. Since the traffic information is required for network dimensioning, it must reflect the demand in the high load period (HLP). Furthermore, this information on annual demand is necessary for costing.
Key Terms in this Chapter
DSL (Digital Subscriber Line): Telephone line improved by equipment, making it capable of broadband transmission. DSL comes in many flavors, known collectively as xDSL.
Cost Driver: Technical parameter that is acting as a restrictive factor for a network element. Therefore, the network element is dimensioned according to this factor.
Incremental Cost: Cost of providing a specific service over a common network structure.
Current Cost: Reflects the cost of the network investment over time, considering issues like amortization.
Quality of Service (QoS): It is an objective measure of the satisfaction level of the user. It refers to control mechanisms that guarantee a certain level of performance to a data flow in accordance with requests from the application program. QoS guarantees are important when network capacity is limited, especially for real-time and streaming applications—for example, voice over IP and IP-TV, since these often require fixed bit rate and may be delay sensitive.
Bitstream Access: Whole sale product consisting of the DSL part (access link) and backhaul services of backbone network (ATM, IP) (see Scott, 2003).
Historical Cost: Reflects the cost of the network elements at the time of acquisition.