Development Trends in Optical Subscriber Networks — Situation Today and Expected Evolution

Abstract

As early as 1980 DBP started its first fibre-to-the-home field trial, the BIGFON trial, to clarify the techniques and introduction conditions for this new subscriber line concept. Islands of pure fibre-to-the-home, services integrated, star-type networks designed for future public introduction were installed which provided the connected customers with all available telecommunication services including high bitrate picturephone. As the main outcome of this trial, which technically was a full success, the DBP realized that there was really no need for such a highly sophisticated subscriber line technique for public purposes since the conventional copper networks are well developed and well suited for the services available.

A more promising way for installing fibre-to-the-home connections has been tried by introducing the “Vorläufer-Breitband-Netz (VBN)” which has gone in operation in 1989. This overlay network offers to the connected customers fully switched transparent 140 Mbit/s channels to be used for broadband switched services. The network size is limited to 1000 subscribers situated in 29 cities.

In order to clarify the present situation for public optical subscriber networks with repect to techniques and economics the DBP has recently started a concept competition “Wirtschaftlicher Einsatz der Glasfaser-Technik im Teilnehmeranschlußbereich” the outcome of which will be competing trial networks implemented by different companies. The economical requirements have led to concepts which minimize the costs per subscriber by sharing the expensive optical systems by several subscribers. The resulting fibre-to-the-curb networks are particularly well suited for providing the subscriber with distributed services. Switched services can, however, be handled with some additional electronical effort by using various multiplexing techniques. Thus, the fibre-to-the-curb networks seem optimized for public installation as long as the bitrates of the switched services are moderate.

With this restriction, however, they primarily offer hardly any advantage with respect to new services compared to the conventional copper networks. The economic figures will very much depend on questions like size of the curb and powering. As a consequence, they cannot be estimated at present. Whether these systems will really be introduced in a larger scale depends therefore to a large extend on the flexibility they offer with respect to new services (like HDTV or broadband-ISDN), new technical requirements (like ATM carriers) and to network development.

The paper summerizes some technical possibilities for evolving fibre-to-the-curb networks to fibre-to-the-home networks with improved services provision and matched to foreseeable technical improvements. The main idea is to utilize fibre splitting to connect several subscribers to one main fibre. Most fascinating is the idea to implement logic ring type networks for ATM purposes and logic star type networks for distribution signals on the same splitting arrangement by using wavelength multiplexing.