FIELD OF THE INVENTION

The invention refers to an apparatus for preventing signal interference in a passive optical network and to a corre- sponding method for removing signal interference in a passive optical network.

BACKGROUND OF THE INVENTION

Passive Optical Networks (PON) are networks of optical fiber lines in which optical light signals are transmitted and dis ^¬ tributed without interposition of active amplifying compo ^¬ nents .

FIG 1 is a diagrammatic representation of a conventional pas ^¬ sive optical network 1. The passive optical network 1 shown is a so-called metro-access network by which data from a pub ^¬ lic optical fiber network 5 are distributed to a plurality of user units 4 also called ONU (Optical Network Unit) . The op ^¬ tical network comprises a so-called OLT 2 (Optical Line Ter ^¬ minal) as access point, which is the transmission interface between the passive optical network 1 and the overlying pub ^¬ lic optical backbone fiber network 5. Moreover, the passive optical network 1 comprises a splitting unit 3 which connects a central cable 6 coming from the OLT 2 to a plurality of arms 7, each arm 7 ending at an ONU 4.

Conventional PONs are bidirectional networks, in particular they distribute downstream traffic from the optical line ter- minal (OLT) to optical network units (ONUs) in a broadcast manner while the ONUs send upstream data packets.

In fiber-optic communications, wavelength-division multiplex- ing (WDM) is a technology which multiplexes multiple optical carrier signals on a single optical fiber by using different wavelengths of laser light to carry different signals. This allows for a multiplication in capacity, in addition to enabling bidirectional communications over one strand of fiber.

Conventionally passive optical networks use a specific wave ^¬ length for downstream communication (from the OLT to the ONUs) whereas for upstream traffic (from the ONUs to the OLT) different solutions are known in the art. According to one known solution all ONUs transmit over the same wavelength (time division multiplexing technique) , whereas, according to another known solution, wavelength-division multiplexing (WDM) is employed (the so-called WDM-PON approach) . However, in both architectures the same OLT and the same central cable are shared by a plurality of ONUs

Unfortunately, in such passive optical networks, interference in a single ONU can cause a collapse of the entire access network during a data transmission from the ONUs to the OLT whenever a disturbed signal coming from an ONU reaches the central cable and there causes a disturbed overall signal which can no longer be processed correctly.

A particularly critical situation is given when a single ONU starts to transmit permanently or casually light at a wave ^¬ length reserved for the upstream communication of another ONU. Especially for networks using a single wavelength for the shared upstream information flow, this would mean that the communication between ONUs and OLT is significantly af- fected or even impossible.

As a consequence, although the opposite direction (OLT to ONUs) is still working properly, the operation of the PON would be significantly limited or even interrupted. With the increasing number of user units (ONUs) connected to a single OLT this scenario would be particularly inconvenient for the enterprise users or wireless network operators.

These disturbances are usually referred to as attacks, even if these events are not necessarily restricted to intentional activities . When such signal interference occurs in conventional optical networks, the fault must be eliminated by a service techni ^¬ cian who examines and individually disconnects the individual arms at the splitting unit of the optical network until the causal fault has been found. This procedure can be very time- consuming and leads to a long failure time of the optical network .

A large number of methods for removing signal interference in a passive optical network (PON) are known from the prior art. In one known method, for example, the restoration of the path affected by the attack includes the manual substitution of hardware elements, which can be very time-consuming and leads to a long re-activation time. In other known methods the disconnection of the ONU affected by the attack includes the si- multaneous disconnection of an entire group of users, which could be particularly inconvenient for the enterprise users or wireless network operators.

Based upon the above discussions, it is concluded that there is a need in the art, for an improved method and apparatus for removing signal interference in a passive optical network (PON) . The improved method should be capable of removing sig ^¬ nal interference or injection attacks in a passive optical network in an effective and automated way without that the removal of the interference causes the disconnection of an entire group of users. SUMMARY OF THE INVENTION In order to overcome the above-described need in the art, the present invention discloses a method for removing signal in ^¬ terference in a passive optical network, the passive optical network including an optical line terminal, a splitting unit coupled with the optical line terminal, an optical network unit coupled with the splitting unit, and an identification signal uniquely associated with the optical network unit; the method comprising the following steps: sending a first signal, detecting the first signal; comparing the detected first signal with the identification signal decoupling the optical network unit from the splitting unit if the comparing step results in a mismatch.

In a next embodiment the method further comprises the step in which the optical network unit sends the first signal.

In other embodiments of the present invention the method fur ^¬ ther comprises the step according to which the splitting unit detects the first signal, compares the detected first signal with the identification signal and decouples the optical net- work unit from the splitting unit if the comparing step results in a mismatch.

In other embodiments of the present invention the method fur ^¬ ther comprises the step according to which the splitting unit maintains the optical network unit coupled with the splitting unit if the comparing step results in a match.

In a further embodiment the method further comprises the step according to which the splitting unit recouples the optical network unit with the splitting unit if the optical network unit is not coupled with the splitting unit and the comparing step results in a match. In a further embodiment the method further comprises the step according to which switching element maintains the optical network unit decoupled from the splitting unit if the optical network unit is not coupled with the splitting unit and if the comparing step in the first controlling means results in a mismatch

It is also an embodiment, that the splitting unit further comprises a switch unit for coupling and re-coupling the optical network unit with the splitting unit and for decoupling the optical network unit from the splitting unit.

In a further embodiment, the switch unit further comprises a first detecting element for detecting the first signal, a first controlling means for storing the identification signal in a first recording medium and for comparing the detected first signal with the identification signal stored in the first recording medium and a switching element for coupling and re-coupling the optical network unit with the splitting unit and for decoupling the optical network unit from the splitting unit.

In other embodiments of the present invention the method fur ^¬ ther comprises the steps according to which the first detect ^¬ ing element detects the first signal, the first controlling means compares the detected first signal with the identifica ^¬ tion signal stored in the first recording medium, and the switching element decouples the optical network unit from the splitting unit if the comparing step in the first controlling means results in a mismatch. In a further embodiment, the method further comprises the step according to which the switching element maintains the optical network unit coupled with the splitting unit if the comparing step in the first controlling means results in a match .

In a further embodiment the method further comprises the step according to which the switching element recouples the opti- cal network unit with the splitting unit if the optical net ^¬ work unit is not coupled with the splitting unit and if the comparing step in the first controlling means results in a match . It is also an embodiment, that the optical line terminal com ^¬ prises a second detecting element for detecting the first signal, and a second controlling means for storing the identification signal in a second recording medium and for comparing the detected first signal with the identification sig- nal stored in the second recording medium.

It is also an embodiment, that the passive optical network further comprises a supervisory channel linking the optical line terminal with the splitting unit.

In a further embodiment, the method further comprises the following steps: the second detecting element detects the first signal, the second controlling means compares the de ^¬ tected first signal with the identification signal stored in the second recording medium, and the optical line terminal sends through the supervisory channel a request to the split ^¬ ting unit to decouple the optical network unit from the splitting unit if the comparing step in the second control ^¬ ling means results in a mismatch. In a next embodiment, the identification signal comprises a continuous wave optical signal modulated at a low frequency.

In an alternative embodiment of the invention, the continuous wave optical signal modulated at a low frequency (41) is ex ^¬ pressed by the followin :

CW being a constant term representing a continuous wave, Vo being a low frequency tone, t being the time, T being the duration of the identification signal, and recti— \ being a

T T

function being equal to one for -— <t<— and being equal to zero elsewhere. It is also an embodiment, that the identification signal fur ^¬ ther comprises an acknowledgment data packet.

In a further embodiment, the method further comprises the following steps: the optical line terminal stores an updated identification signal in the second recording medium, the optical line terminal sends the updated identification signal to the splitting unit through the supervisory channel, the splitting unit stores the updated identification signal in the first recording medium and the optical line terminal sends the updated identification signal to the optical net ^¬ work unit.

The problem stated above is also solved by a system for re ^¬ moving signal interference in a passive optical network, com- prising: an optical line terminal, a splitting unit coupled with the optical line terminal, an optical network unit cou- pled with the splitting unit, an identification signal uniquely associated with the optical network unit, wherein the optical network unit is configured to send a first signal and the splitting unit is configured to detect the first sig- nal, compare the detected first signal with the identifica ^¬ tion signal and decouple the optical network unit from the splitting unit if the comparing results in a mismatch.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention is explained by way of example in more detail below with the aid of the attached drawings.

FIG 1 is a diagrammatic representation of a conventional pas ^¬ sive optical network;

FIG 2 is a diagrammatic representation of a passive optical network according to one embodiment of the invention;

FIG 3 is a diagrammatic representation of a switch unit ac ^¬ cording to one embodiment of the invention;

FIG 4 is a representation of the normalized Fourier transform of the identification signal versus frequency difference (v - v ₌ ) according to one embodiment of the invention;

FIG 5 is a representation of the required duration T of the identification signal versus frequency according to one em ^¬ bodiment of the invention.

FIG 6 is a diagrammatic representation of a passive optical network according to one embodiment of the invention. DESCRIPTION OF THE INVENTION

As regards the description of FIG 1, reference is made to the background of the invention.

Illustrative embodiments will now be described with reference to the accompanying drawings to disclose the teachings of the present invention. While the present invention is described herein with reference to illustrative embodiments for par ^¬ ticular applications, it should be understood that the inven- tion is not limited thereto. Those having ordinary skill in the art and access to the teachings provided herein will rec ^¬ ognize additional modifications, applications, and embodi ^¬ ments within the scope thereof and additional fields in which the present invention would be of significant utility.

FIG 2 is a diagrammatic representation of a passive optical network 21 according to one embodiment of the invention. The passive optical network 21 shown in FIG 2 is a so-called metro-access network by which data from a public optical fi- ber network 25 are distributed to a plurality of user units 24 also called ONU (Optical Network Unit) . The optical net ^¬ work comprises an OLT 22 (Optical Line Terminal) as access point, which is the transmission interface between the pas ^¬ sive optical network 21 and the overlying public optical fi- ber network 25. Moreover, the passive optical network 21 comprises a splitting unit 23 which connects a central cable 26 coming from the OLT 22 to a plurality of arms 27, each arm 27 ending at an ONU 24. Each ONU 24 may represent a single sub ^¬ scriber, and in FIG 2, as an example, only four ONUs 24 are shown for the purpose of a clearer representation. However, more than 256 ONUs 24 may be connected to a single splitting unit 23. In the embodiment of the invention shown in FIG 2, the splitting unit 23 comprises an optical splitter 29, for connecting the central cable 26 to the plurality of arms 27, and a plurality of switch units 28 for coupling and decoup ^¬ ling the optical network units 24 with the optical splitter 29. In particular, the splitting unit 23 may comprise one switch unit 28 for each of the arm 27.

In the present invention each ONU 24 is uniquely associated with an identification signal which, in one embodiment of the invention may be represented by a continuous wave optical signal modulated at a low frequency (e.g. kilo-Hertz to hun- dreds of Mega-Hertz) , the frequency tone being the unique identifier of the single ONU 24. In other embodiments of the invention the identification signal may be also be repre ^¬ sented by a sequence and, in particular, by a finite se ^¬ quence .

During normal operation an ONU 24 sends a request for commu ^¬ nication to the OLT 22, followed by a first signal which, if the ONU 24 is working properly, should correspond exactly to the identification signal (i.e. the continuous wave optical signal modulated at a low frequency) . The splitting unit 23 detects the first signal sent by the specific ONU 24 and com ^¬ pares this detected first signal with the identification sig ^¬ nal which corresponds to the specific ONU 24. If this compar ^¬ ing step corresponds to a match it means that the ONU 24 is working properly, i.e. no interference or attack occurred, and in this case splitting unit ensures that the connection to the ONU is established. Conversely, if the comparing step corresponds to a mismatch it means that the specific ONU 24 is not working properly, i.e. an interference or an attack occurred, and in this case the splitting unit 23 decouples the specific ONU 24 from the passive optical network 21. Therefore, according to the present embodiment of the inven ^¬ tion, the splitting unit 23 guarantees for the security of the network by checking the identification signal of each specific ONU 24 independently. As a consequence, a possible interference will be efficiently identified and rapidly re ^¬ moved by the splitting unit 23.

FIG 3 is a diagrammatic representation of a switch unit 28 according to one embodiment of the invention. The switch unit 28 shown in FIG 3 is a particular implementation of the switch unit 28 shown in FIG 2 according to one embodiment of the invention and is suitable for coupling and decoupling a specific ONU 24 with the optical splitter 29 through the arm 27. The switch unit 28 shown in FIG 3 includes a first de ^¬ tecting element 34 for detecting the first signal sent by the ONU 24 through the arm 27, a first controlling means 33 for storing the identification signal in a first recording medium 37 and for comparing the detected first signal with the iden- tification signal stored in the first recording medium 37 and a switching element 31 for coupling and decoupling the optical network unit from the optical splitter 29. Moreover, the first detecting element 34 includes a frequency detector 36, a photodiode 35 coupled with the frequency detector 36 through the link 311 and a tap 32 coupled with the photodiode 35 through the link 312 for detecting the first signal in the arm 27. Moreover the first controlling means 33 is coupled with the frequency detector 36 through the control link 38 and with the switching element 31 through the control link 39.

During normal operation the ONU 24 sends a request for commu ^¬ nication, containing a first signal which, if the ONU 24 is working properly, should correspond exactly to the identifi- cation signal (e.g. the continuous wave optical signal modu ^¬ lated at a low frequency) . The first detecting element 34 de ^¬ tects the first signal sent by the specific ONU 24 through the tap 32 and the photodiode 35, while the frequency detec ^¬ tor 36 detects the frequency tone included in the first sig- nal . The first controlling means 33, which may be a switch control unit, compares the detected first signal with the identification signal related to the specific ONU 24 which is stored in the first recording medium 37 which corresponds to the specific ONU 24.

If this comparing step corresponds to a match it means that the ONU 24 is working properly, i.e. no interference or at ^¬ tack occurred, and in this case for the switching element 31 ensures that the connection to the ONU is established. Con- versely, if the comparing step corresponds to a mismatch it means that the specific ONU 24 is not working properly, i.e. an interference or an attack occurred, and in this case the switching element 31 decouples the specific ONU 24 from the passive optical network 21.

As soon as the interference or the attack has been removed, and the specific ONU 24 sends a new request for communica ^¬ tion, containing a first signal, this new request together with the first signal is detected again from the first de- tecting element 34. Consequently the first controlling means 33 compares the detected first signal with the identification signal related to the specific ONU 24 stored in the first re ^¬ cording medium 37, and if the comparing step corresponds to a match the switching element 31 ensures that the connection between the splitting unit 23 and the ONU 24 is established. In particular the splitting unit 23 maintains the optical network unit (24) coupled with the splitting unit if the comparing step results in a match or recouples the optical net ^¬ work unit (24) with the splitting unit 23 if the optical net- work unit (24) is not coupled with the splitting unit (for example because the optical network unit 24 has been previ ^¬ ously decoupled from the splitting unit) and the comparing step results in a match. Therefore, according to the present embodiment of the inven ^¬ tion, the splitting unit 23, and in particular the switch unit 28, guarantees for the security of the network by check ^¬ ing the identification signal of each specific ONU 24 inde- pendently. As a consequence, a possible interference will be efficiently identified and rapidly removed by the switch unit 28.

The embodiment of the invention shown in FIG 3 can be imple- mented in different ways. As shown in FIG 3 the implementa ^¬ tion may require the use of tap couplers, photodiodes and op ^¬ tical switches. In particular the optical switching element 31 may be designed as "Normally Closed" (NC) component wherein the specific arm 27 is connected to the passive opti- cal network if no electrical signal is applied to the switch ^¬ ing element 31. The use of discrete components and micro- optics may represent an option as well as the use of Planar Lightwave Circuits (PLC) which offers several advantages like, for example, reduced size, reduced costs due the capa- bility of integrating several optical components, and low power supply. In the PLC configuration, the switching functionality may be obtained by means of a thermal effect which may require a heather element for the switch unit 28. FIG 4 is a representation of the normalized Fourier transform S ₊ (v) 41 of the identification signal versus frequency dif ^¬ ference (v - v ₌ ) 42 according to one embodiment of the inven ^¬ tion. In the present embodiment, the identification signal may be represented by a continuous wave optical signal modu- lated at a low frequency (e.g. kilo-Hertz to hundreds of Mega-Hertz) , the frequency tone being the unique identifier of the single ONU 24. In particular, in the present embodiment, the continuous wave optical signal modulated at a low frequency 41) may be ex ^¬ pressed by the f llowing formula: ( 1 )

Where CW may represent a continuous wave (e.g. the continu ^¬ ous wave generated by the ONU 24 laser) , Vo being a low fre ^¬ quency tone, t being the time, T being the duration of the identification signal, and rect (i—AI being a function being

T T

equal to one for -— < t <— and being equal to zero elsewhere. In the following, since the CW term is constant, it will be neglected without loss of generality.

The Fourier transform of the continuous wave optical signal modulated at a low frequency is:

S(v) = T ^■ sinc[(v + v ₀ ) · Τ] + T ^■ sinc[(v - v ₀ ) · Τ] , where the sinc{x) function is defined as a ^' /¾c(x)= ^^ .

S(v) is a symmetric function around the zero frequency, so that in the following only its normalized part:

S ₊ (v) =sinc[(v - v ₀ )- T]

will be analyzed.

In FIG 4, the normalized Fourier transform S ₊ (v) 41 of the continuous wave optical signal modulated at a low frequency is shown for different values of the duration T of the iden ^¬ tification signal, in particular for T = 0.0001 sec 45 and for T = 0.002 sec 44. It can be seen that the spectrum 41 becomes narrower when increasing T , making the identification of the frequency easier for the splitting unit.

The zeros of S ₊ (v) are located at the following frequencies: n being an integer number, with n≠0

In order to obtain a quantification of the minimum required duration T of the identification signal, the relative uncertainty on the peak frequency Sv is defined as ratio between the interval between the zero of order n and —n and v ₀ , that can be expressed as

And therefore

1 2 - n

v ₀ δν

i.e. at a given frequency v ₀ the required duration T of the identification signal increases by decreasing δν . Furthermore, the larger the uncertainty δν is, the larger the electrical bandwidth required by the ensemble of the pi ^¬ lot tones will be. Based on the above analysis, the minimum value of the total required electrical bandwidth can be com ^¬ puted as follows:

f In

TotalBandwidth = « · —

Where n represents the number of ONUs 24 of the splitting unit (i.e. the splitting ratio) .

The duration T of the identification signal can be expressed in terms of number ( n _T ) of periods of the pilot tone, leading to :

A v 2 - n

δν = =

v ₀ n _T TotalBandwidth = n

These relationships indicate that the relative accuracy de ^¬ pends only on the number of periods contained in the identi ^¬ fication signal and not on the absolute value of the tone frequency. As a consequence, for a given accuracy, the dura ^¬ tion of the identification signal can be reduced by increas ^¬ ing the frequency of the pilot tone v ₀ at the expense of an increased total electrical bandwidth.

FIG 5 is a representation of the required duration T 51 of the identification signal versus frequency 52 according to one embodiment of the invention and for n = 3 , and Sv = l% 53, Sv = 5% 54, Sv = l0% 55, Sv = 30% 56 as a function of the low frequency tone v ₀ 52. For this evaluation, v ₀ is assumed to be representative of the frequency range of the ensemble of the low frequency signals related to the arms of the split ^¬ ting unit. It is worthwhile to note, that for tone frequen ^¬ cies in the MHz range, relative accuracies below 1% can be achieved by keeping the duration T 51 of the identification signal below 1ms, so that the overall performance of the net ^¬ work is almost unaffected.

FIG 6 is a diagrammatic representation of a passive optical network 61 according to one embodiment of the invention. The passive optical network 61 shown in FIG 6 is a so-called metro-access network by which data from a public optical fi ^¬ ber network are distributed to a plurality of user units 24 also called ONU (Optical Network Unit) . The optical network comprises an OLT 22 (Optical Line Terminal) as access point, which is the transmission interface between the passive opti ^¬ cal network 61 and the overlying public optical backbone fi- ber network. Moreover, the passive optical network 61 comprises a splitting unit 23 which connects a central cable 26 coming from the OLT 22 to a plurality of arms 27, each arm 27 ending at an ONU 24. Each ONU 24 may represent a single sub- scriber, and in FIG 6, as an example, only four ONUs 24 are shown for the purpose of a clearer representation. However, more than 256 ONUs 24 may be connected to a single splitting unit 23. The OLT 22, according to the present embodiment of the invention, comprises a multiplexer/demultiplexer 616 for multiplexing the downstream signal travelling along the optical link 614 to the central cable 26 and for demultiplexing a plurality of upstream signals 615 coming from the same cen ^¬ tral cable 26. The OLT 22, according to the present embodi ^¬ ment of the invention, further comprises a second detecting element 62 which includes a frequency detector 64, a photodi ^¬ ode 65 coupled with the frequency detector 64 through the link 618 and a tap 615 coupled with the photodiode 65 through the link 619 for detecting the first signal in each of the upstream signals 613. The second detecting element 62, and in particular its frequency detector 64, is further coupled through the control link 617 with a second controlling means 69 which is suitable for storing the identification signal in a second recording medium 69 and for comparing the detected first signal with the identification signal stored in the second recording medium 69. The OLT 22, according to the present embodiment of the invention, further comprises an opti ^¬ cal port 68 for the supervisory channel to convey this chan ^¬ nel from the optical line terminal 22 with the splitting unit 23. In particular the supervisory channel is an Unidirec- tional Optical Supervisory Channel (OSC) and transport con ^¬ trol information from the second controlling means 69 to the splitting unit 23 through the multiplexer/demultiplexer 616 and the central cable 26 so that the Unidirectional Optical Supervisory Channel (OSC) is multiplexed by together with the downstream signal (is usually one signal/wavelength) 614 in the central cable 26.

Conventional Coarse Wavelength Division Multiplexing (CWDM) filters and Array Waveguide Grating (AWG) base components can be used to implement the multiplexer/demultiplexer unit 616 and the demultiplexer unit 611.

In the present embodiment of the invention, the splitting unit 23 comprises a demultiplexer 611 for demultiplexing the Unidirectional Optical Supervisory Channel (OSC) 68 and the cable conveying the downstream and upstream signals 612. The splitting unit 23 further comprises an optical splitter 29 for connecting the cable conveying the downstream and up- stream signals 612 to the plurality of arms 27, and a plural ^¬ ity of switch units 28 for coupling and decoupling the optical network units 24 with the optical splitter 29. In particular, the splitting unit 23 may comprise one switch unit 28 for each of the arm 27. The splitting unit 23 further com- prises an optical supervisory channel detection unit 66 for detecting the signals conveyed in the Optical Supervisory Channel, and an Optical Supervisory Channel switch control unit 67 coupled with the Optical Supervisory Channel detec ^¬ tion unit 66 through the link 620 for controlling the cou- pling and de-coupling operation of the switch units 28 and linked to these switch units 28 through the control links 621.

According to the present embodiment of the invention, as shown in FIG 6, the identification signal further comprises an acknowledgment data packet. Therefore, in the present em ^¬ bodiment, the identification signal comprises a continuous wave optical signal modulated at a low frequency represented for example by the formula (1), and an acknowledgment data packet conveying information to be compared with the one stored in the second recording medium 69 (e.g. in a look-up table) . According to the present embodiment of the invention, during normal operation the ONU 24 sends a request for communica ^¬ tion, beginning with the identification signal (e.g. the continuous wave optical signal modulated at a low frequency) , followed by data packets including acknowledgements data packets detected by means of a receiver connected to line 613 but not shown in the diagram. In this way, the information contained in the acknowledgement data packets and indicating which ONU is sending the request is extracted by the OLT 22. In the present embodiment the comparing step includes compar- ing both the frequency tone related to the continuous wave optical signal modulated at a low frequency and the acknowl ^¬ edgment data packet.

As an effect, the present embodiment allows the detection and the removal of interference also in the case in which the first signal includes a matching frequency tone but a wrong acknowledgment data packet. According to the present embodi ^¬ ment of the invention, the OLT 22 still guarantees for the security of the network by checking the identification signal of each specific ONU 24 independently, while the splitting unit 23, and in particular the switch unit 28, merely exe ^¬ cutes the requests to decouple the ONU 24 from the network if an interference occurs . According to another embodiment of the invention, which can be also represented in FIG 6, the identification signal can be changed and updated on a regular basis, in particular the tone frequency of the continuous wave optical signal modu ^¬ lated at a low frequency can be changed and updated regu- larly. According to the present embodiment, the OLT 22 up ^¬ dates the tone frequency related to a particular ONU 24 and stores the new value in the second recording medium 69, at the same time the OLT sends the updated tone frequency re- lated to a particular ONU 24 to the splitting unit (23) through the supervisory channel via the link 68 and the splitting unit (23) stores the updated identification signal in the first recording medium (37), contemporaneously the op ^¬ tical line terminal (22) sends the updated value of the fre- quency of the identification signal to the optical network unit (24) which sets the frequency of the new tone to the up ^¬ dated value.

One of the effects of some embodiments of the invention, is that the number of arms 27 of the splitting unit 23 can be increased beyond the limitation imposed by the number of ports of AWG- Demultiplexers or CWDM-Demultiplexers . Thus, each OLT 22 is allowed to serve a higher number of ONUs 24 reducing in this way the reduced cost per arm 27.

A further effect of some embodiments of the invention, is that no spectrally separated optical control signals are re ^¬ quired to control the connection of each arm 27, so that the number of ports can be increased without the use of demulti- plexer (AWGs or CWDM filters) . As a consequence, neither an array of fixed wavelength laser nor tuneable lasers are nec ^¬ essary in the OLT 22.

A further effect of some embodiments of the invention, is that the OLT does not require an optical amplifier and the related requirement on the maximum allowed distance between the OLT 22 and the splitting unit 23 becomes unnecessary. A further effect of some embodiments of the invention, is the capability to exclude each port from the network separately. In the present invention the number of ONUs 24 served by a single OLT 22 can be increased to 256 or beyond by simultane ^¬ ously guaranteeing the possibility to disconnect each indi ^¬ vidual port.

A further effect of some embodiments of the invention, is that the splitting unit 23 guarantees for the security of the network by checking the identification signal of each arm 27 independently. As a consequence, an attacking port will be identified in a fast way by the splitting unit 23. As a fur ^¬ ther consequence, the attacking port does not need necessar ^¬ ily to be identified by the OLT 22 and a corresponding proce ^¬ dure is intrinsically not necessary.