Energy consumption of telecommunication access networks

5.2 HFC-Modelling As already mentioned, HFC networks exist in the model region. The locations of the individual network’s elements could not be obtained. However, this is not an obstacle. Unlike the DSL technologies, the HFC networks with DOCSIS 3.0 (864 MHz) do not suffer from length restrictions so that the exact spatial location of the individual network elements is of minor importance. Therefore, a green field approach was chosen, based on the service group size. A service group is a group of subscribers that are fed by a single fibre node. This also means that all subscribers in a service group share the available transmission capacity – a HFC network is a shared medium. Therefore, by varying the service group size, the exclusively usable data rate of each assigned subscriber is co-determined. At a service group size of 32 connections for example, 50 Mbit/s can be delivered to every subscriber exclusively. A larger service group size also allows to provide 50 Mbit/s, but it cannot be guaranteed for each subscriber at any given time (shared medium). A smaller service group size on the other hand allows higher exclusive data rates for each subscriber. As mentioned above, every service group is fed by a fibre node. Broadly speaking, a fibre node combines or separates the data signal and the TV/Radio signal and it is typically part of a bigger amplifier point which also contains coax line amplifiers (A, B, C lines) for the transmission on the coaxial infrastructure. Each fibre node is again assigned to a cable modem termination system (CMTS), which feeds the fibre nodes via a fibre optic cable, terminates the subscriber’s modem connection and provides access to the core network. Different service group sizes lead to different numbers of fibre nodes and CMTS. A typical service group consists of about 250 subscribers in active HFC networks in Germany today. Due to the above-mentioned characteristics, the number of network elements can be estimated by linear calculation. 5.3 FTTH-Modelling At the start of this project, there were no FTTH/B networks in the chosen model region. Hence, a green field approach is assumed in order to utilize the technological advantages of a FTTH network. Especially the large cable lengths of a multiple of ten kilometres allow high cable connection densities in a single point of presence (PoP). This will reduce the amount of properties and will help to further raise efficiency gains, like decreasing power losses due to inefficient power supplies for the active components. In a FTTH network, each subscriber is connected to a specific point of presence (PoP). The subscriber’s modem connection is terminated at the PoP and the access into the core network is provided. Between the subscriber and the PoP there are no active components. As a result, the number of the PoPs and their efficiency will be significant for the power consumption of the whole network. This applies both to Point-to-Point and to GPON networks. Point-to-Point In a Point-to-Point network, each subscriber is directly connected to a specific PoP. There are no active or passive network elements between the PoP and the subscriber, which separates or merges signals. According to information from different network operators, the PoP size varies significantly. There are small PoPs with only a few cable connections (<50) and very large PoPs with up to several thousand cable connections in use (the largest reported PoP size was around 4,000 cable connections). Due to the obtained information on the PoP power consumption, the maximum PoP size in this model is limited to 1,000 subscribers. Furthermore, the PoP size is clustered into six classes: 1. 10,000 subscribers max. 4. 250 subscribers max. 2. 750 subscribers max. 5. 100 subscribers max. 3. 500 subscribers max. 6. 50 subscribers max. Figure 11: Used PoP size classes for FTTH – Point-to-Point. 16