New books available: Giappone per caso and L’altro Iran

The problem of broadband in Italy

The rational use of ADSL technology in relation to the needs of today's users

The International Network
Icon: Wikimedia Commons
The issue of digital divide between Italy and other industrialized countries is well known. While in Italy are regularly sold lines with a speed of 7 Mbps (but there are many who still use 640 kbps and even those whose speed remains up to 56 kbps), in other countries like the United States, Germany and Japan is not uncommon to take out a subscription of 100 Mbps.

The Italian situation is further aggravated by the massive use of a technology such as ADSL, one of the few possible ways to introduce the broadband widespread in this territory, excluding the still relatively limited phenomena such as wireless and satellite connections.
This situation derives from the characteristics of the Italian telephone network, mostly dating to the 50s if not to prior periods, and still having the technological features of the time, mainly consisting of copper cabling.
The recent progress in infrastructures has been largely focused on the backbone network with the replacement of outdated wiring with fiber optic connections between the main nodes and the local exchanges.

This however did not provide an answer to the private customers, still served by the old cooper pair, and has highlighted two key issues: Therefore, in the 90s, once dismantled the old national telephone company and left the field open to private corporations, the network's unique owner decided, unlike other countries, not to do extensive fiber optic wiring, a choice that would have clearly been costly but that would have helped to permanently solve the problem of the link between the local telephone exchange and the final user (the "last mile").

The solution that was widely adopted was instead a compromise between cost and performance, with the fiber-optic wiring (or radio link) up to the local exchanges, and installing there the DSLAMs, a series of modulators and demodulators of DSL signal, able to route data traffic in a range of frequencies not used by the telephone cooper pair, independently from the voice signal.

They preferred, however, focus not on a symmetrical distribution of transmission's and reception's bandwidth (SDSL), but on a non-symmetrical bandwidth's distribution (ADSL), which implies a strong restriction on the ability to transmit (upload) in favor of the receive (download).
The situation has, after a dozen of years, virtually unchanged. The needs of Internet users are instead radically changed, with the dynamism of their new social role and the interdependence among them, a factor which led to an inevitable decentralization and relocation of worldwide bandwidth resources.

Interactivity is expressed in social networks and virtual communities, where the users take an active role uploading movies and entire photo albums, and to do that they require an adequate upload capacity.

The issue of relocation of resources is represented by peer to peer connections (direct data exchange between users), the real problem of all obsolete networks and their providers.

From this point of view it is clear why a currently sold ADSL line with a 1:20 ratio between the transmission and the reception carrier, is no longer appropriate to the needs of today's users.

The inspiration for a more rational use of ADSL technology could derive from the exploitation of its inherent characteristics, analyzed here in relation to the most common standard, known as the Recommendation G.992.1 or G.DMT, established in 1999 by the International Telecommunication Union, for which it has been provided the theoretical maximum speed of 12 Mbps on downstream and 1.3 Mbps on upstream.

However, it is necessary to consider two preliminary factors: the first concerns the amount of bandwidth and the speed made available by a provider: for example, a subscription called "7 mega" could mean that you have a downstream carrier of 7 megabits per second (not Byte... between them there is a difference of almost an order of magnitude!), which is however often only guaranteed to 640 kbps or not guaranteed at all.

The second factor concerns the quality of the line and the resulting performance, due to the need of a limited length of the link (no more than a few kilometers) from the local exchange to the user because of the rapid decay of the DSL signal.

Considering the above concerns, we might think of a general rationalization, possibly based on the one hand in the rise of the single upstream carrier to the theoretical maximum of 1.3 Mbps and, only if it is necessary to connect a series of users reached by poor quality lines and far from the exchange, by reducing the downstream carrier up to the level to which the signal can be sufficiently strong within 5 km from the exchange.

This technique could be done due to the independence of the two channels of reception and transmission, which both have their independent parameters that can be detected and evaluated separately (for example the noise margin and attenuation of the line). This is also the reason why, in an ADSL, the transmission channel has a greater "strength" than the reception due to the difference in terms of bandwidth.

In any case, in Italy the available speed is already lower than normal, not only compared to the theoretical possibility of the G.DMT standard, but also to the carrier made available by the same providers because of the inadequate infrastructure.
Therefore we can observe the practice of overbooking on the DSLAMs, reason for which the providers statistically stimate, or more simply hope, that all the users linked to a particular exchange will not connect at the same time.
Part of this discourse is the hostility among the providers to peer to peer applications because they, by providing a continuous flow of data, increase the probability of overload the Net.
A high ADSL's ratio imbalance in favor of downstream is therefore welcome by the service providers because of its inherent limiter role, allowing them to postpone to a brighter economic future some possible infrastructural upgrades.

Published in 2010