This paper is concerned with long-term (20+years)\nforecasting of broadband traffic in next-generation networks.\nSuch long-term approach requires going beyond extrapolations\nof past traffic data while facing high uncertainty in\npredicting the future developments and facing the fact that,\nin 20 years, the current network technologies and architectures\nwill be obsolete. Thus, ââ?¬Å?order of magnitudeââ?¬Â upper bounds of\nupstream and downstream traffic are deemed to be good\nenough to facilitate such long-term forecasting. These bounds\ncan be obtained by evaluating the limits of human sighting\nand assuming that these limits will be achieved by future\nservices or, alternatively, by considering the contents transferred\nby bandwidth-demanding applications such as those\nusing embedded interactive 3D video streaming. The traffic\nupper bounds are a good indication of the peak values and,\nsubsequently, also of the future network capacity demands.\nFurthermore, the main drivers of traffic growth including\nmultimedia as well as non-multimedia applications are identified.\nNew disruptive applications and services are explored\nthat can make good use of the large bandwidth provided by\nnext-generation networks. The results can be used to identify\nmonetization opportunities of future services and to map\npotential revenues for network operators.
Loading....