METHOD AND ARRANGEMENT FOR SIGNAL TRANSMISSION AND COMPENSATION OF BACK REFLECTIONS IN OPTICAL ACCES PON SYSTEMS

Abstract

In a network terminal (ONU) of an optical network near end crosstalk (NEC) is compensated by a digital generated cancellation signal. To establish a connection with another terminal (OLT) signals avoiding NEC are transmitted and the compensation is performed while the power of the transmitted signal is increased in steps.

Description

FIELD OF THE INVENTION

The invention relates to a method and arrangement for signal transmission and compensating back reflections in an optical communication system. In particular, the invention relates to back reflections in NGOA (Next Generation Optical Access) and PON (Passive Optical Networks).

BACKGROUND OF THE INVENTION

The context of NGOA (Next Generation Optical Access) and PON (Passive Optical Networks) is known fin the art. The ONUs are located in the subscribers' homes, while the OLT is located in the central office.

FIG. 1 shows an ONU with the connection to the PON at the left. The numbers are taken from the use case, where attenuation in the PON is at maximum and the receive level at the ONU is just −45 dBm. The Tx signal has to have maximum strength in this case to overcome the PON attenuation in upstream direction. Therefore this is the worst case for near-end crosstalk: The Rx signal is weak and the Tx signal is strong.

In FIG. 1 the near-end crosstalk is assumed to be −20 dB. Such a back reflection in a connector has no significant influence to the transmission of the signals; especially the attenuation due to a bad connector can be neglected. Please note, that other systems like GPON are not vulnerable to near-end crosstalk and connector reflections. If, for example, existing GPON PONS are migrated to NGOA the operator may experience problems, because some of the new NGOA ONUs may not work after migration.

Reflections at connectors are specified by the suppliers of the optical connectors. Nevertheless these specifications are only valid, if the connectors are cleaned before they are plugged. In the field the operator cannot guarantee, that these cleaning procedures are done perfectly. This problem is increased in fiber-to-the-home applications, where the end user does the last connection manually.

This may have the following consequences:

• • The ONU may not be able to establish a bi-directional connection to the OLT due to near-end crosstalk.
  • The ONU may not be able to detect the source of the problem.
  • The ONU may not be able to inform the OLT about the problem; the operator has no hint 42 where the problem is located in the field.
  • The tolerance to near-end crosstalk of an NGOA ONU may be too low for real-world applications.

In general there are three principles to avoid near-end crosstalk (NEC):

• • Wide spectral distance between Rx and Tx spectrum. For example Rx and Tx can be in different optical bands. This allows filtering in the optical domain to suppress NEC. GPON is an example.
  • Two fibers. Rx and Tx happens at different fibers. Long-haul systems are an example.
  • Ping-Pong: Rx and Tx happens at different time intervals.

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 schematic representation of an ONU connected to a PON.

FIG. 2 is a schematic representation of the near-end crosstalk spectrum.

FIG. 3 is a schematic representation of near-end crosstalk spectrum according to an embodiment of the invention.

FIG. 4 is a schematic representation of the cancelation path according to an embodiment of the invention.

FIG. 5 is a schematic representation of the cancelation path including the Rx filter according to an embodiment of the invention.

FIG. 6 is a schematic representation of the cancelation path according to an embodiment of the invention.

DESCRIPTION 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 particular applications, it should be understood that the invention is not limited thereto. Those having ordinary skill in the art and access to the teachings provided herein will recognize additional modifications, applications, and embodiments within the scope thereof and additional fields in which the present invention would be of significant utility.

To establish a connection the ONU scans the optical spectrum until it finds a downstream signal. During this scan the Tx power of the ONU is shut off and near-end crosstalk is not a problem. After the ONU has found a downstream signal and has locked to it the next step is to establish the upstream connection.

The ONU starts with a very small Tx power and increases the Tx power step by step until the OLT signals that the Tx power is ok. This Tx power control requires that the ONU shuts off its signal as soon as it loses the frame lock to the OLT downstream signal. Such a loss of frame lock may have many reasons and with the existing protocol there is no way for the ONU to decide whether the root cause is a back reflection or not.

The ONU switches off the Tx power immediately after losing downstream frame lock. It tries to re-establish frame lock. If possible, it put its local oscillator into a hold-over mode, where it is no longer under control of the downstream signal. Experience from current demonstrator measurements showed, that the ONU can receive the downstream signal in the hold-over mode for several seconds. The ONU uses this hold-over time to switch on and off its Tx power several times. The ONU correlates the Tx power with the frame lock respective the frame loss of the downstream signal. If there is a strong correlation, the ONU can decide, that the root cause of the problem is a back reflection from the PON.

The ONU may use a Tx on-off pattern, which is unique per ONU. E.g. it can be a bit sequence derived from its serial ID or from a hash of the serial ID. It can be a random sequence also. This allows the ONU to differentiate between its own near-end crosstalk and the near-end crosstalk from another ONU.

The ONU may use the Tx off phases of the Tx on-off pattern to recover the local oscillator lock. This allows Tx on-off patterns which are longer than several seconds. In principle the Tx on-off patterns can be indefinitely long.

The Tx on-off pattern described above can be used to increase the Tx power in the on-pulses of the pattern stepwise, until the OLT signals that it can receive an upstream signal. After that it is possible for the ONU to transmit data upstream to the OLT for several seconds. The ONU is not able to receive the downstream signal during this time due to the near-end crosstalk, but it is sufficient to signal the ONU-ID and the problem description to the OLT.

The following data is of interest for an operator:

• • The ONU-ID: This allows the operator to determine the subscriber identity and therefore the location of the ONU in trouble (street, building and apartment). This information is necessary to send service persons directly to the fault location and to reduce OPEX (operational expenditure).
  • The receive power of the downstream signal and the Tx power, at which the ONU loses the downstream signal. This allows estimating the connector quality.

In this mode a non-duplex signal transmission is possible. Such a “Ping-Pong” protocol has a reduced data rate of course and the latency is increased significantly. Nevertheless it is possible to offer a fall-back data transmission to the subscriber until the fault has been removed by the operator. This may reduce OPEX, because the operator has much more time to solve the problem.

The spectrum of the Rx signal and the Tx signal are shifted by an intermediate frequency in the order of 1 GHz. As long as the spectral width of the Tx signal is smaller than this intermediate frequency the near-end crosstalk NEC can be filtered at the ONU receiver.

This situation is shown in FIG. 2. The near end crosstalk can be reduced significantly by suppressing frequencies outside the Rx spectrum. For the ONU transmitter it is difficult to achieve a Tx spectrum, which is as narrow as shown in the figure above. The shape of the Tx spectrum depends on the Tx modulator used. A phase modulator based upon a SOA shows a much broader Tx spectrum.

The resulting crosstalk spectrum is shown in FIG. 3. Since parts of the Tx spectrum are folded into the Rx spectrum filtering cannot handle this situation.

One embodiment of the invention includes a novel cancelation path to the Tx/Rx signal processing as shown in FIG. 4 depicting e.g. an ONU. A data signal generator 1 is connected via a digital-analog-converter 2 with an optical transmitter 3 emitting a modulated optical signal TX_opt. A downstream optical signal RX_opt is received from the OLT by an optical receiver 4 and demodulated into an electrical signal, which is fed via a analog-digital-converter 5 and via a subtractor to a digital receiver 6 retrieving the data. Near end crosstalk is coupled from the output of the optical transmitter 3 to the optical receiver 4.

The new cancelation path 7-9 includes a modulation simulator 7, a phase shifter 8, and a attenuator 9 connected in series. To compensate the NEC signal the modulation simulator 7 has to generate a compensation signal C_dig with the spectrum of the NEC signal. A good approximation is the emitted optical signal but transformed to the “electrical” domain (frequency band) respectively digital domain. In this embodiment the “electrical” signals are processed in digital form. The phase and attenuation of the compensation signal C_dig is adjusted by the phase shifter 8 and attenuator 9 to the same values as the actual near-end crosstalk path has. The phase shifter can be realized with a digital all-pass filter and the attenuator is basically a multiplication. The achieved cancellation signal C_electrical is subtracted from the received signal (TX_electrical) in the electrical or digital domain which comprises data and NEC.

The phase and amplitude of the cancelation path is adjusted by the following scheme:

1. Tx power of the ONU is reduced to a value slightly below the critical Tx power where frame 120 loss of the downstream signal happens.

2. The bit error rate (BER) of the downstream signal is measured.

3. A minimum controller adjusts phase and amplitude of the cancelation path to the minimum of BER.

4. After BER is minimized, the Tx power is increased stepwise; for every step the BER is minimized by adjusting the cancelation path. This procedure guarantees that BER measurement is possible always, which is a pre-requisite to use a minimum controller.

5. In addition, the spectrum of the generated compensation signal may be optimized.

The scheme in the figure above does not take analogue filters into account which are in the Rx path between the Rx photo diode and the ADC. Such filtering can improve crosstalk tolerance (see prior art above).

FIG. 5 shows, how the cancelation path can be enhanced to handle an Rx filter. A new component is the Rx filter 11 in front of the ADC 5. The invention compensates for this new component by adding a digital compensation filter 12 with the same characteristics as the cancelation path.

The following calculation shows the mathematical transformation the Tx signal is undertaken when it is reflected in the PON. The same mathematical transformations have to be done in the cancelation path to cancel out the near end crosstalk.

Modulator:

TX_optic(t)=sin (2·πƒ_optic·t−TX_electrical(t))

Crosstalk path (Delay Δt and attenuation a):

NEC_optic(t)=α·sin (2·π·ƒ_optic(t−Δt)−TX_electrical(t−Δt))

Homodyne receiver:

${\mathrm{NEC}}_{\mathrm{electrical}}\left(t\right)={\mathrm{NEC}}_{\mathrm{optic}}·\sin \left(2·\pi ·f_{\mathrm{optic}}·t\right)$ ${\mathrm{NEC}}_{\mathrm{electrical}}\left(t\right)=\frac{a}{2}·\cos \left(2·\pi ·f_{\mathrm{optic}}·t-2·\pi ·f_{\mathrm{optic}}\left(t-\Delta t\right)+{\mathrm{TX}}_{\mathrm{electrical}}\left(t-\Delta t\right)\right)-\frac{a}{2}·\cos \left(2·\pi ·f_{\mathrm{optic}}·t+2·\pi ·f_{\mathrm{optic}}\left(t-\Delta t\right)-{\mathrm{TX}}_{\mathrm{electrical}}\left(t-\Delta t\right)\right)=\frac{a}{2}·\cos \left(2·\pi ·f_{\mathrm{optic}}·\Delta t+{\mathrm{TX}}_{\mathrm{electrical}}\left(t-\Delta t\right)\right)-\frac{a}{2}·\cos \left(2·\pi ·2·f_{\mathrm{optic}}·t-2·\pi ·f_{\mathrm{optic}}\Delta t-{\mathrm{TX}}_{\mathrm{electrical}}\left(t-\Delta t\right)\right)$

By filtering the optical frequencies due to the low-pass characteristic of the photo diode the remaining crosstalk signal is:

${\mathrm{NEC}}_{\mathrm{electrical}}\left(t\right)=\frac{a}{2}·\cos \left(2·\pi ·f_{\mathrm{optic}}·\Delta t+{\mathrm{TX}}_{\mathrm{electrical}}\left(t-\Delta t\right)\right)$

The cancelation path must produce the same mathematical transformation, based upon three parameters, which need to be controlled by the minimum controller:

• • The optical carrier phase o_je.
  • The baseband delay Δt.
  • The attenuation a.

$C_{\mathrm{electrical}}\left(t\right)=\frac{a}{2}·\cos \left(o_{\mathrm{je}}·\Delta t+{\mathrm{TX}}_{\mathrm{electrical}}\left(t-\Delta t\right)\right)$

The index “electrical” is used for signals in the electrical and in the digital domain. The optical carrier phase o_je has a limited stability due to the limited coherence length of the laser in the ONU. For example today a typical NGOA laser has a coherence length of 400 m. This restricts the maximum length of the fiber between ONU and the location of the reflection to 100 . . . 200 m, depending on the required NEC suppression. This is sufficient to handle a reflection due to the cabling inside a larger multi-dwelling building. If the reflection is further away from the subscriber's home the specification of laser line width and coherence length need to be improved.

The baseband delay can be estimated by switching Tx on and off. The crosstalk will increase signal amplitude. Therefore it is possible to determine the correlation between on-off state and measured amplitude for different delays. This correlation can be found best with the maximum possible Tx power. The method to find the maximum possible Tx power in NGOA even if NEC does not allow bidirectional communication is described above.

After estimating the parameter At the other parameters can be optimized with a minimum controller based upon BER measurements. The principle of NEC cancelation has been shown without considering polarization-diversity receivers.

FIG. 6 shows, that two independent cancelation paths 7-9 and 13 can handle polarization-diversity receivers as well. A second receiver path includes a second optical receiver 24, a second analog-digital converter, a second subtractor and a second digital receiver 21 connected in series and shown as functional units.

According to an embodiment of the invention both cancelation paths 7-9 and 13 share the same baseband delay At, but a and o_je are completely independent and are controlled by independent minimum controllers.

The calculation above has been made for a phase modulator. The same principle is possible for IQ-modulators or simple on-off-keying also by adapting the modulator simulation in the cancelation path.

The block diagrams above handle the use case, where a single point of reflection has to be cancelled out. By adding multiple cancelation paths or corresponding calculations it is also possible to handle multiple points of 186 reflections in the PON.

The cancelation paths shown above are implemented completely in the digital signal processing domain (left from the ADCs and DACs in the figures above). This may require the Rx ADC to handle both the receive signal and the NEC in its dynamic range. If the NEC is significantly larger than the signal itself this increases the required number of ENOBs of the ADC. An alternative embodiment includes the computation of the cancelation signal in the digital domain and route it via a dedicated DAC into the analogue domain. Here it can be subtracted from the analogue input signal in front of the ADC (a subtractor is indicated by dashed lines). The resulting input signal requires less ENOBs with this hybrid digital/analogue approach than the pure digital approach.

The present invention is not limited to the details of the above described principles. The scope of the invention is defined by the appended claims and all changes and modifications as fall within the equivalents of the scope of the claims are therefore to be embraced by the invention. Mathematical conversions or equivalent calculations of the signal values based on the inventive method or the use of analogue signals instead of digital values are also incorporated.

ABBREVIATIONS AND REFERENCE SIGNS

NEC near end crosstalk

Tx transmitter

Rx receiver

TX_electrical transmitter signal, electr. or digital domain

RX_electrical receiver signal, electr. or digital domain

TX_opt emitted optical signal

RX_opt received optical signal

1 data signal generator

2 digital-analog-converter

3 optical transmitter

4 optical receiver

5 analog.digital-converter

6 digital receiver

7 modulation simulator

8 phase shifter

9 attenuator

10 subtractor

11 Rx filter

12 compensation filter

13 second cancellation path

20 second subtractor

21 second digital receivers

24 adapted optical receiver

25 second analog-digital-converter

Claims

1. A method for data processing of an optical network unit, comprising:

receiving configuration information at the optical network unit;

adjusting a light signal to a wavelength or wavelength range indicated by the configuration information;

demodulating an incoming optical signal by means of the light signal;

mixing the demodulated incoming optical signal with a signal generated by an oscillator; and

generating a modulated optical upstream signal modulating the light signal by means of a software radio, so that the resulting optical upstream frequency is shifted with respect to the frequency of the local oscillator by a programmable amount.

2. The method according to claim 1, wherein the light signal is provided by a laser.

3. The method according to claim 1, wherein generating the modulated optical upstream signal includes multiplying the light signal by the factor of e−2πiΔ71.

4. The method according to claim 1,

wherein the signal generated by the oscillator is set to a predefined or arbitrary frequency;

wherein the light signal is adjusted to improve reception of the incoming optical signal and to receive the configuration information.

5. The method according to claim 1, wherein bandwidth of the modulated optical upstream signal is lower than bandwidth of the incoming optical signal.

6. The method according to claim 1, wherein the configuration information is provided by an optical line termination.

7. The method according to claim 1, wherein the light signal used for upstream modulation is adjusted to a wavelength or wavelength range indicated by the configuration information by reducing a signal-to-noise ratio.

8. The method according to claim 1, wherein the optical network unit is a subscriber unit of a passive optical network.

9. A device, comprising:

a light source providing a light signal;

a receiver to which the light signal is fed, wherein the receiver provides an electrical output signal;

wherein the light source is adjustable to a wavelength or wavelength range provided by received configuration information received;

an oscillator and a mixer, wherein a signal of the oscillator and the electrical output signal from the receiver are conveyed to the mixer and wherein the signal of the oscillator is tunable to improve a reception of the incoming signal at the device; and

a modulator to which the light signal of the light source is fed to provide a modulated optical upstream signal.

10. The device according to claim 9, wherein said device is an optical network unit.

11. the device according to claim 9, wherein said device includes a Digital Signal Processor.

12. The device according to any of claim 9, wherein the signal of the oscillator is tunable to improve a signal-to-noise ratio of an output signal of the mixer.

13. A communication system comprising at least one device according to claim 9.