NETWORKS WITH METALLIC PAIRS

* * * * *

DESCRIPTION

The invention relates to the technical field of ultra-broadband transmissions (over 30 Mbit/s), with particular reference to fast access to the Internet and to other prestigious services and contents (unicast video, etc.).

More in detail, the invention relates to a method, and a system of electronic apparatuses that cooperate with one another to implement this method, designed to allow access multiplexing on networks with metallic pairs, through a division of the transmission frequencies into spectral resources destined for separate operators and individually subject to vectoring-type interference suppression techniques.

Recent evolution of access technologies on networks with metallic (copper) pairs enables transmission speeds (data-rates) in the order of tens/hundreds of Mbit/s to be reached, using system architectures of FttC (Fiber to the Cabinet) type and its variants or evolutions, such as FttB (Fiber to the Building) and FttDp (Fiber to the Distribution Point).

These architectural solutions are advantageous for DSL (Digital Subscriber Line) technologies, significant examples of which are VDSL2 (Very high bit-rate DSL type 2), also known as ITU-T G.993.2, and the more recent G.Fast (G-series Fast Access to Subscriber Terminals), also known as ITU-T G.9700/G.9701.

To ensure customers obtain the highest data-rate values, the multiplexing electronic apparatuses used by the aforesaid architectural solutions, named DSLAM (DSL Access Multiplexer), are positioned inside roadside cabinets (CAB), or small boxes, known as Distribution Points (DP), normally located close to homes.

For ultra-broadband uses on networks with metallic pairs and over large distances, the performances of the maximum data-rate that can be reached by the aforesaid architectural solutions can be limited by the occurrence of interference produced by the crosstalk phenomenon and in particular by Far-end crosstalk (FEXT).

To combat the effects of FEXT and, consequently, greatly improve the performances that can be obtained with the aforesaid architectural solutions, it is possible to use an interference suppression technique, called Vectoring, initially proposed by Ginis-Cioffi and subsequently internationally standardised as ITU-T G.993.5. Vectoring ideally eliminates FEXT by pre-compensating downstream (DS) disturbances, i.e. directed toward the customer, and simultaneously cancelling upstream (US) disturbances, i.e. coming from the customer.

However, vectoring has the noteworthy limitation of expressly requiring the presence of only one telecommunications operator on the cable of metallic pairs on which it is performed.

As the intensity of FEXT increases as a function of the transmission frequency used, on networks of metallic pairs the use of higher frequencies and in particular those higher than 17.66 MHz, is in fact excluded. Even for frequencies below 17.66 MHz, FEXT still causes a considerable reduction of the data-rate otherwise achievable on ideal transmission channels, i.e. in the presence of thermal noise alone.

Unfortunately, even a limited number of metallic pairs not controlled centrally in the same cable can cause a substantial deterioration of the transmission quality of the metallic pairs subject to vectoring (vectoring groups), deriving from the occurrence of interference produced by alien- FEXT. More in detail, when several telecommunications operators actuate vectoring on the same bandwidth and on the same cable of metallic pairs separately, i.e. without a centralised control body, each vectoring group acts on the others as alien-FEXT.

This makes vectoring unusable where the National Regulatory Authority (NRA) imposes a regulatory remedy of SLU (Sub Loop Unbundling) type, designed to enable passive physical access of Other Licensed Operators (OLO) to the local sub-network of a given incumbent operator, i.e. in multi-operator access configurations.

To combine vectoring and SLU, alternative architectural solutions (Multi Operator Vectoring - MOV) have been proposed, capable of enabling vectoring to be applied in the case in which several operators connected to the same CAB, or to several adjacent CABs, are closely coordinated and synchronised by a single centralised control body.

Considering the well-known difficulties of implementing MOV solutions, and it limitations, in order to combine vectoring and SLU there have been proposals to extend the ultra-broadband transmission frequencies, with the aim of obtaining an increased data-rate in the presence of vectoring. DSLAM apparatuses currently being developed and not yet available on the market are expected to operate at a bandwidth of over 17.66 MHz, adopting dual band solutions (35.32 MHz).

Although promising high data-rate values and extensions of range beyond the typical lengths of VDSL2 with vectoring, also due to the use of powerful channel codes, the aforesaid dual band apparatuses are nonetheless unable to solve the problem of the simultaneous presence of several operators co-located on the same cable of metallic pairs, as the strong dependence of alien-FEXT on the transmission frequencies inevitably tends to limit the benefit of broadening the spectrum of these frequencies, in the presence of separate vectoring groups.

The object of the invention is to propose a method and a system of reference that allow the problems and the limits described above to be solved, making access multiplexing on networks with metallic pairs possible through a hierarchical division of the transmission frequencies into spectral resources reserved for separate operators and individually subject to vectoring-type interference suppression techniques, so as to enable the implementation of ultra-broadband transmissions by several operators connected on the same cable of metallic pairs and not coordinated or synchronised with one another. The object of the invention is achieved with a method for access multiplexing on networks with metallic pairs according to the independent claim 1.

The invention also relates to a system for access multiplexing on networks with metallic pairs according to claim 7.

Further features of the method and of the system for access multiplexing on networks with metallic pairs according to the invention are contained in the respective subsequent dependent claims.

The method and the system for access multiplexing on networks with metallic pairs according to the invention, through a particular division of the transmission frequencies into spectral resources destined for separate operators and individually subject to interference suppression techniques, for example of vectoring type, produce the following important advantages:

- they allow implementation of ultra-broadband transmissions on access networks with metallic pairs;

- they allow co-location of several operators on the same cable of metallic pairs, without requiring the coordination and/or the synchronisation of these by a centralised control body;

- they allow co-location of several operators on the same cable of metallic pairs, avoiding deterioration of the data-rate produced by

FEXT and by aiien-FEXT;

- they allow co-location of several operators on the same cable of metallic pairs, placing said operators on equal terms in provision of the access service, through satisfaction of a criterion of fairness of performance between these operators, both in terms of total bandwidth allocated to the operators and of the bit rate obtainable per user when the distance from the CAB changes (hereinafter "dual fairness criterion");

- they allow the customers of different operators, co-located on the same cable of metallic pairs, at the same distance from a cabinet of reference, all to obtain, in practice, the same performances, and therefore to be served with the same data-rate.

- they allow the use of vectoring-type interference suppression techniques, even in the presence of several operators co-located on the same cable of metallic pairs;

- they allow the implementation of system architecture based on a single DSLAM apparatus accessible to several operators, obtaining frequency unbundling solutions, and of system solutions based on several DSLAM apparatuses also separated from one another, each accessible by a single operator, obtaining solutions of physical unbundling of the resources (example: two or more DSLAMs in the same Cabinet, or two or more DSLAMs in two adjacent Cabinets, or two or more DSLAMs, one or more of which at the Cabinet and one or more of which at the DP or at the building);

- they allow high data-rate values to be obtained ( 00 Mbit s) for a large number of users, relatively distant from the DSLAM apparatuses;

- they allow an operator to maintain full responsibility for the confidentiality of the user's information by avoiding having to delegate the management of its customers' data streams to a competitor.

Further features and advantages of the invention will be more apparent from the detailed description set forth below, with the aid of the figures, which show preferred embodiments thereof, illustrated by way of non- limiting example, wherein:

- Fig. 1(a) schematically shows a possible hierarchical division of an ultra-broadband transmission frequency into sub-channels, sub- bands and related frequency slots reserved for separate operators according to the method of the invention;

- Fig. 1(b) shows an example of hierarchical division of a sub-channel into sub-bands and of allocation of the related slots in the case of three operators; - Fig. 2 schematically shows an extended band DSLAM apparatus, designed to determine the division of an ultra-broadband transmission frequency into frequency sub-channels reserved for separate operators according to the method of the invention;

- Fig. 3 shows in a schematic block diagram, the operating logic of a control device for extended band DSLAM apparatuses, designed to enable access multiplexing on networks with metallic pairs according to the invention;

- Fig. 4 shows, in a schematic block diagram, an extended band DSLAM apparatus including the control device of Fig. 3;

- Fig. 5 schematically shows a complete system architecture, of multi-CAB type, comprising extended band DSLAM apparatuses, designed to enable access multiplexing on networks with metallic pairs according to the invention;

- Fig. 6 graphically shows the data-rate values achievable when the transmission bandwidth changes, in different conditions of alien- FEXT and of distance, with or without access multiplexing on networks with metallic pairs according to the invention;

- Fig. 7 graphically shows the gain attainable on the downstream data-rate subject to vectoring, as a function of the CAB-NT (Cabinet

- Network Termination) distance for different bandwidth values of a DSLAM apparatus of reference, with or without access multiplexing on networks with metallic pairs according to the invention.

Hereinafter in the description, purely by way of example, system architectures that can be produced according to the method and the system will be described, with reference to DSL FttC type system architectures.

With reference to Figs. 1(a), 1(b) and 2, a method for access multiplexing on networks with metallic pairs according to the invention substantially comprises the steps of:

- dividing ultra-broadband transmission frequencies into a hierarchical structure organised in bands, the last level of which comprises spectral resources obtained by applying a dual fairness criterion that substantially guarantees equality of the total spectral resources allocated to each operator and of the bit-rate obtainable per user for each distance;

- reserving said spectral resources for separate operators;

- applying vectoring-type interference suppression techniques to said spectral resources.

The hierarchical structure of dividing the ultra-broadband transmission frequencies comprises, as non-limiting example, a first level of dividing into sub-channels, a second level of dividing the sub-channels into sub-bands and a last level of dividing the sub-bands into slots, which represent the spectral resources to be allocated to each operator, as represented in Figs. 1(a) and 1(b).

More in detail, the first channel (/ = 0), i.e., the channel that will occupy the band 0 - 17.66 MHz in the cable of metallic pairs, is not subject to vectoring if operating in a scenario of backward compatibility. The baseband signals that correspond to the generic i-th channel (/≥1) are translated by the baseband to higher frequencies, for example using a single sideband modulation, through specific SSB (Single Side Band) processors, followed by an appropriate conversion, through specific up- converters.

With particular reference to Fig. 1(a), each of the two operators Op1- Op2, indicated by way of non-limiting example, also transmits on frequencies higher than 17.66 MHz and therefore on higher channels (i≥ 1 ), modulating only its DMT (Discrete Multi-Tone) carriers in the slots allocated present in each of the sub-bands Bi, i = 1 , 2,... ,K, of each subchannel.

The method of dividing into sub-bands Bi of the invention ensures compliance with a fairness criterion (dual fairness) between the customers distributed in the access network, regardless of the distance of customers from the cabinet (CAB) of reference, at least up to the maximum distance of coverage of the service. The technique based on the dual fairness criterion at sub-band level (blocks of 5 MHz, or similar) prevents customers of different operators, positioned on the same cable of metallic pairs, at the same distance from the cabinet of reference, from being served with different data-rates. The technique of dividing the sub-bands based on the dual fairness criterion is therefore an essential element that distinguishes the method of the invention from a generic frequency division operation, as it is aimed at placing operators on equal terms in the supply of the access service both in terms of bit rates per user and of total band allocated per operator.

The term dual fairness technique is intended as a method of planning the channelling into sub-bands Bi and related division into slots, two examples of which are illustrated by way of non-limiting example in Figs. 1(a) and 1(b), designed to ensure fairness of band allocated per operator and to produce practically overlapping trends of the data-rate curves as a function of the CAB-NT (Cabinet - Network Termination) distance for any operator that connects its DSLAM to the seat of the CAB, up to the maximum distance of interest for supply of the service to the customers.

The method for access multiplexing according to the invention ensures compliance with the aforesaid dual fairness condition taking account of the propagation characteristics on the cable of metallic pairs and regardless of the geometrical arrangement of the customers of different operators within the area covered by the service. Examples of hierarchical schemes of division of the band that ensure dual fairness are shown in Figs. 1(a) and 1(b), wherein the slots all have the same dimensions, regardless of the sub-bands, and are allocated to different operators (three in the example in Fig. 1(b)) according to an alternate and interleaved scheme.

Although the example of distribution of the bands indicated above refers implicitly to a fair and static division of the spectral resources available, the method of the invention can be applied both in the case of uneven division of the spectral resources and of dynamic division thereof.

With reference to Fig. 2, a DSL channel that operates using frequencies even higher than 17.66 MHz can be generated in baseband using existing VDSL2 modulators, designed and optimised for the transmission band 0 - 17.66 MHz, together with vector processors of standard type, also operating on the transmission band 17.66 MHz. Therefore, the possibility of re-utilising existing technology can be a reason for choosing the sub-channel band of the standard profile 17a, also for the purpose of backward compatibility. However, the method and the system of the invention will not be limited by the development of DSLAM apparatuses, dual band or higher, without fairness measures designed to enable them to be used by multiple operators.

As the method of multiplexing of the invention provides for the hierarchical division of the transmission frequencies into sub-bands and related slots that can be allocated to separate operators, a system designed to implement this method, simply and inexpensively, consists in the sharing between several operators even of a single DSLAM apparatus, positioned in a cabinet or in a Distribution point, thereby determining a new form of frequency unbundling. Although not optimised from the viewpoint of costs, the case of several DSLAM apparatuses, positioned in adjacent cabinets, each operating in different sub-bands with physical unbundling modes, is operationally very similar.

With reference to Fig. 3, a system for access multiplexing on networks with metallic pairs according to the invention substantially comprises a control device 1 , having the task of implementing a DSLAM apparatus designed to allow the objects and the advantages specified above to be achieved through a controlled and intelligent alteration of the operation of a DSLAM apparatus associated therewith.

Said device 1 comprises at least:

- a subsystem 2 comprising enabling means of the carriers;

- a unit 3 comprising resource negotiating means; - an interface 4 comprising remote control means;

- a subsystem 5 comprising means for monitoring the frequency spectrum and assessing the interferences thereon.

The subsystem 2 for enabling the carriers takes a central role, as it has the task of setting the parameters of all the extended band modem-routers (NT - Network Termination) interconnected with the DSLAM apparatus, constraining them to operate only on the carriers of interest, according to the specifications of the desired frequency division, both for downstream DS transmissions and for upstream US transmissions.

In the downstream DS direction, the aforesaid device 1 performs at least one of the following functions:

- it specifies the carrier values that the extended band modulator MOD DMT of the DSLAM apparatus must use to transmit on the downstream stretch (the carriers to use are defined based on the slots assigned to the operator or according to the spectral planning established for the operator);

- it specifies the carrier values to which the symbol pre-coding operation must be applied in order to implement the vectoring procedure (DS Vectoring - FEXT Precoder);

- it specifies the carrier values on which the matrices needed for pre- coding must be calculated, which are obtained on the basis of appropriate vectoring algorithms that may be modified to take into account any limitations on the number of carriers to use for downstream transmission;

- it processes the data containing the channel estimates performed both by apparatuses inside the DSLAM and by the modems active on the user's side in order to set up the calculation of the matrices needed for pre-coding and optimise the transmitting power at single carrier level;

- it sets the symbol coders (Symbol Cod.) of the DSLAM apparatus to take into account the number of bits per symbol and the carrier values to use for downstream transmission;

- it monitors the generation of the probing signals needed by the modems on the user's side to estimate the transmission channel (this information is then retransmitted by the user modems to the DSLAM apparatus in order to calculate the matrices needed for pre- coding);

- if necessary, it uses a signalling channel from the DSLAM apparatus to the modems on the user's side to indicate which carriers must be used to receive the symbols sent on the downstream stretch.

In the upstream US direction, the aforesaid device 1 performs at least following functions:

- it specifies the carrier values that must be used to receive signals by each extended band demodulator of the DSLAM apparatus associated with the signals sent by the various users;

- it specifies the carrier values to which the FEXT (US Vectoring - Fext canceller) cancelling operation must be applied for upstream transmission;

- it specifies the carrier values on which the matrices needed for the FEXT (US Vectoring - Fext canceller) cancelling operation must be calculated, which are obtained on the basis of appropriate vectoring algorithms, and may be modified to take into account any limitations on the number of carriers to use for upstream transmission;

- it processes the upstream signals received from the single users to obtain estimates of the transmission channels used to calculate the matrices needed for the FEXT (US Vectoring - Fext canceller) cancelling operation for upstream transmission;

- it sets the symbol decoders (Symbol Dec.) of the DSLAM apparatus to take into account the number of bits per symbol and the carrier values to use for receiving the upstream signals;

- if necessary, by means of an appropriate signalling channel from the DSLAM apparatus, it tells the transmission subsystem of the single modems on the user's side which carriers to use to transmit data on the upstream stretch.

The device 1 can implement functions of monitoring and controlling the status of transmissions, both upstream US and downstream DS, and assessing interference produced by operators on adjacent channels, with signalling of any faults, through a specific unit 5 that also uses the information relating to the signal-to-interference-plus-noise ratio (SINR) on each sub-carrier of each user and the estimates of the transmission channels.

The device 1 can also be controlled both locally and remotely, through a specific interface 4 that enables the carrier values to be set for both downstream DS transmissions and upstream US transmissions.

The device 1 can be used to implement dynamic division techniques of the frequency spectrum, designed to increase spectral efficiency based on bilateral (or multilateral) agreements between operators connected to the same CAB, or to the same DP.

Unlike MOVs with centralised control, in dynamic management of the frequency spectrum, performed through the device 1 , an operator does not have to delegate the management of its customers' data streams to a competitor, thus maintaining full responsibility for the confidentiality of the user's information, and total freedom to plan its future systems in the same cable area.

If applied, a dynamic management of the frequency spectrum enables operators to optimally serve a different number of customers and/or applications that require different data-rate values, with the result that the data traffic load is equalised in the sub-bands. In conditions of imbalance, temporary transfer of one or more slots in one or more sub-bands to the operator that supports the highest traffic load causes a data-rate benefit to its customers without penalising the customers of the operator that temporarily transfers or more slots. In this way, both operators benefit from dynamic management of the frequency spectrum available and the transmission quality is on average better for all the customers served by them.

The device 1 , in its configuration for dynamic use of the frequency band available, can include some measures that, however, do not require the exchange of private information between operators. At application level, a service channel for exchanging synthetic traffic data, and a management protocol of a shared algorithm to automatically negotiate the band and for any cross charging, are needed.

In the simplest case, if the operators grant one another a somewhat balanced average traffic potential, the principle of compensation with no exchange of money, namely "Bill & Keep", can be implemented.

Assuming synchronisation of the streams of several operators (although asynchronous behaviours would also be possible), the definition of a frame time, at the level of DSLAM apparatus (10 ms, or similar), useful to periodically mark resource negotiation with "band on demand" techniques, is particularly advantageous (in the case of a single DSLAM apparatus shared among them). For DSLAM apparatus of the same manufacturer, temporal synchronisation of several DSLAM apparatus can take place.

With reference to Fig. 3, to implement the above, the resource negotiation unit performs at least the following functions:

- it uses a service channel for exchanging information on the characteristics of the operators' traffic;

- it performs a shared algorithm to automatically negotiate the band and for any cross charging;

- it may rely on a temporal synchronisation of several DSLAM apparatuses (physical unbundling) or, if a single DSLAM apparatus is used (frequency unbundling), on the synchronisation of the streams of operators managing their own bands.

With reference to Fig. 4, the operation of a DSLAM apparatus comprising the aforesaid device 1 is as follows.

In downstream transmission DS, the multiple signal received by two or more operators on a single optical fibre with wavelength multiplexing (or, likewise, on the same number of fibres as the number of operators) is first converted into an electrical signal and then demultiplexed, with appropriate demultiplexers DEMUX, to be inserted into two or more separate vectoring groups (the same number as the number of operators involved).

The FEXT processor (DS Vectoring - FEXT precoder), with appropriate symbol coders (Symbol Cod.), pre-compensates and then routes the DMT streams in parallel to the DSL carriers (for example, with band of 4 kHz), selected based on the command from the control device 1 , the input of which receives the control signals from the operators, locally or from a remote location.

In the case of static allocation of the carriers, this operation is carried out during installation and is destined to remain until the configuration of the operators connected to the cabinet or to the Distribution point changes. In the case of dynamic allocation, a continuous negotiation is performed at slot level (in one or more sub-bands, such as 5 MHz, or similar), which can be periodically exchanged among the users.

In upstream US transmission, the streams coming from the modems in the customer's premises, on separate carriers (4 kHz), are demodulated individually, with appropriate demodulators DEM DMT, and sent to the vectoring processor (US Vectoring - FEXT canceller) that operates as FEXT canceller, under the static or dynamic control of the device 1. Downstream they follow the symbol decoders (Symbol Dec.) and the multiplexer MUX, before conversion of the electrical signal into optical signal and wavelength multiplexing or, similarly, transmission on the same number of fibres as the number of operators.

Further system architectures can involve the use of different DSLAM apparatuses for each operator, if necessary also in different cabinets or one at the CAB and the other at the DP, but interconnected, each controlled by its own device 1 , capable of coordinating with the others through a specific communication protocol.

In a possible simplified system architecture, without the device 1 and without altering the configurations of existing CABs, several operators switch to the use of apparatuses equipped with the functions described above without suffering limitations in the choice of vendor (which may be different) and without the need for coordination (except for a prior division of the band according to an agreement between the parties or a rule established by the national regulatory authorities (NRA)), autonomously managing the data streams of its customers, without specific operating limitations in movement of the metallic pairs downstream of the apparatus, and in any case achieving considerable benefits in the performances to offer its customers.

With reference to Fig. 5, this architecture can be used both with current DSLAM VDSL2 in the band 0 - 17.66 MHz and with future extended band DSLAM DSL, operating above 17.66 MHz, and by virtue of removal of FEXT, it allows several operators to increase the data-rate offered to customers without acting, for example, on the existing multi-CAB configurations.

This variant has the advantage of an operating flexibility that enables operators to evolve their access networks without coordinating future system choices with one another. For example, in compliance with the rules established by the national regulatory authority (NRA), an operator connected to a cabinet (CAB) of reference can extend, separately from the other operators, the band used without affecting, for technical reasons, their choices. Moreover, each operator can extend its fibre optic in the vicinity of the buildings (with DSL, G.Fast or other technologies) maintaining the availability of the sub-bands allocated to them, again without technically affecting the other operators.

Finally, this variant enables the implementation of: - operating scenarios that are not backward compatible (nbc);

- operating scenarios that are backward compatible (be).

In the case of operating scenarios that are not backward compatible, the operators will decide to coordinate with one another in time, to reallocate, by mutual consent, the spectrum on which each of them, for the portion available to them, will implement vectoring, and the NT (network Termination) home router of all customers will be replaced, passing for example from the standard according to the profile 17a to the new system operating at frequencies higher than 17.66 MHz.

In the case of operating scenarios that are backward compatible, each operator will implement vectoring in its own sub-bands separately from the others, without the need to synchronise or coordinate with the others and will only replace the home routers of its own customers.

The method and the system of the invention enable vectoring to be used where, for example, the national regulatory authority (NRA) imposes the obligation of Sub Loop Unbundling (SLU) to promote competition. In this case, according to a predetermined criterion, the national regulatory authority (NRA) allocates some sub-bands to the operators of electronic communications that request them, which are free to operate within the spectrum allocated to them.

The method and the system of the invention are applicable to any DSL standard, including the G.Fast standard, which in the future may suffer, just as VDSL2, from limitations of performances if adopted in the presence of several operators while the SLU regulatory obligation is in force.

The method and the system of the invention allow the implementation of system solutions suitable to achieve the objectives established in Europe and in other countries for ultra-broadband infrastructures, thus expanding the technological alternatives available to operators.

The method and the system of the invention ensure true advantages in the presence of multi-operator scenarios with extended band solutions as FEXT depends greatly on the transmission frequency used and alien-FEXT tends to limit the benefit of spectral expansion, in the presence of separate vectoring groups.

To appreciate the severity of the penalisation caused by alien-FEXT the results of the calculations indicated graphically in Figs. 6-7 can be observed. These calculations take as reference the spectral plan of the VDSL2 (profile 17a), assuming that the band is extended above the limit of 7.66 MHz and that, for simplicity, the whole of the additional spectrum is allocated to the DS direction of the connection. Therefore, they refer to the use of a standard AWG 24 cable, which represents the most frequent choice in the Italian access network. These calculations also assume that two operators {m = 2) access the cable and that the same number of customers (n = 4, 10) are part of each of the two separate vectoring groups.

For two distance values from the roadside cabinet (CAB) to the home modem-router (NT - Network Termination) at the customer's home, d(m) = 100, 200, Fig. 6 compares the data-rate values in the case of SBV (Sub Band Vectoring) division of the transmission band and FBV (Full Band Vectoring) sharing of the whole band by the two vectoring groups. Fig. 6 confirms that expansion of the band is in fact inefficient in the presence of vectoring, unless alien-FEXT is eliminated and, at the same time, that the number of interferents, J, has little influence, as even only a few interfering pairs are sufficient to penalise the whole vectoring group.

In the same architectural conditions, Fig. 7 considers four band values β(ΜΗζ) = 35.32; 52.8; 70.4; 105.6 and introduces the gain, G, in terms of data-rate obtainable in DS direction as a function of the CAB-NT (Cabinet - Network Termination) distance, c/(m), both with SBV division of the transmission band and without it, with shared use of the FBV spectrum. The gain shown in the ordinate is defined as follows: G = BR [B(MHz)] / BR [17.66 MHz], where BR expresses the data-rate value.

Fig.7 shows how, even with only two operators, the gain G can be modest (0.3÷0.5) even for high bands of the DSLAM, while its value becomes very large (6÷8) with the use of transmission band division, made possible by the method and by the system of the invention.

In conclusion, while alien-FEXT disturbance is so high that it cancels the benefit of the increased band, in the case of band division each channel becomes AWGN (Additive White Gaussian Noise) type, as alien- FEXT is cancelled completely. In this last case, among the other benefits, the powerful channel codes, well known in the literature, also prove to be fully effective and can produce high coding gain values. Therefore, the combination of the use of channel coding and the division of the transmission band, made possible by the method and by the system of the invention, restores the optimal nature of AWGN channel to each metallic pair of the cable and can thus be the solution of the ultra-broadband coverage of a country at modest infrastructuring costs for several operators equipped with their own infrastructures.