A Framework for defining Optical Parameters to be used in WSON networks through GMPLS
draft-martinelli-ccamp-opt-imp-fwk-00

Status of this Memo

By submitting this Internet-Draft, each author represents that any applicable patent or other IPR claims of which he or she is aware have been or will be disclosed, and any of which he or she becomes aware will be disclosed, in accordance with Section 6 of BCP 79.

Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet-Drafts.

Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as “work in progress.”

The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/1id-abstracts.txt.

The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html.

This Internet-Draft will expire on April 29, 2009.

Abstract

In the context of Wavelength Switched Optical Networks (WSON) the problem of selecting a lightpath might be constrained by an evaluation of the optical impairments associated to a wavelength. This is a critical step in a transparent dense wavelength division multiplexing (DWDM) optical islands where a lightpath feasibility has to be assessed between two regenerations nodes.

This memo provides a framework in which optical parameters can be considered a control plane. The document relies on information already present in ITU documents and summarize in term of lightpath constraints.

Table of Contents

1.  Introduction
2.  Motivation for an Impairment-Aware Control Plane
3.  Overview of Optical Impairment Evaluation
    3.1.  Impairments Evaluation Flow
    3.2.  Parameter Classification According to its Information Complexity
    3.3.  Additional Parameter Sharing Properties
4.  Control Plane Considerations
5.  Optical Interface Characteristics
6.  Optical Path Characteristics
    6.1.  Linear Impairments
        6.1.1.  Fiber Losses
        6.1.2.  Insertion Losses (Optical components looses)
        6.1.3.  Amplifier Spontaneous Emission (ASE)
        6.1.4.  Crosstalk
        6.1.5.  Fiber Chromatic Dispersion
        6.1.6.  Polarization Mode Dispersion (PMD)
        6.1.7.  Polarization Dependent Loss (PDL)
    6.2.  Fiber Optical Non-Linearities
7.  Optical Channel Estimation
    7.1.  Optical Power
    7.2.  Optical Signal to Noise Ratio
    7.3.  Residual Chromatic Dispersion
    7.4.  Residual Differtial Group Delay (DGD)
    7.5.  Q-Factor
8.  Acknowledgements
9.  Contributing Authors
10.  IANA Considerations
11.  Security Considerations
12.  References
    12.1.  Normative References
    12.2.  Informative References
Appendix A.  ITU Parameters Missing Information
§  Authors' Addresses
§  Intellectual Property and Copyright Statements

1.  Introduction

Generalized Multi-Protocol Label Switching (GMPLS), [RFC3945] (Mannie, E., “Generalized Multi-Protocol Label Switching (GMPLS) Architecture,” October 2004.), applied to wavelength switched optical networks (WSON) needs to address specific issues related to optical technologies. The framework document [I-D.draft-ietf-ccamp-wavelength-switched-framework] address specific issues related to transparent optical networks and to the routing and wavelength assignment (RWA) problem. Optical impairments, however, are out of the scope of that document.

One of the key aspects while dealing with transparent DWDM optical networks are the physical impairments incurred by non-ideal optical transmission media, and how they accumulate along an optical path. Because of these impairments, even if there is physical connectivity (fibers, wavelengths, and nodes) between the ingress and egress nodes, there is no guarantee that the optical signal (light) reaches the Egress node with acceptable signal quality in terms of Bit Error Rate (BER) or other quality measures (e.g. Q-factor).

Scope of this framework document is to provide an overview of optical impairments and parameters that have to be considered to assess the feasibility of an optical path (lightpath) and provide any useful information for an impairment-aware protocol implementation. For this purpose a classification and properties of optical impairment are provided along with some considerations related to control plane.

The detailed definitions of the physical effects with related mathematical models are, in general, already defined within ITU-T and this documents will refer to ITU-T documentation whenever is possible. This document will only report additional information to determine how this information must be integrated into the control plane.

The document [RFC4054] (Strand, J. and A. Chiu, “Impairments and Other Constraints on Optical Layer Routing,” May 2005.) is another reference targeting optical network routing along with impairments and constrains. The goal there was to provide a survey of all optical constrains along with possible approaches. In this contribution the scope is much more limited, as we only list optical impairments and their salient properties with respect to control plane implementation.

2.  Motivation for an Impairment-Aware Control Plane

GMPLS today is not aware of optical impairments in the DWDM network. This implies that it cannot determine whether an end to end optical path is feasible or not. While there has been much work on adding wavelength constraints and other simple parameters to the control plane ([I-D.draft-ietf-ccamp-rwa-info] and [I‑D.bernstein‑ccamp‑wson‑signaling] (Bernstein, G., “Signaling Extensions for Wavelength Switched Optical Networks,” July 2008.)), this information alone does not help determine the feasibility of the path. In fact, the more impairments are taken into account, the more aggressive the network can be in terms to reach, and the lower the cost of the solution, since less regenerators are needed. Conversely, if a limited number of optical impairments is taken into account, the network must allow for larger margins to account for uncertainty in the missing parameters, and will have to regenerate more frequently.

One question that arises is: why add such information into the control plane and not deal with it through planning tools that today are in charge of optical feasibility determination? The reasons for this are multi-fold:

1. Issues with offline planning tools.
   Planning tools are typically off line tools and do not have accurate information as to the real impairments in the network. This adds uncertainty and requires higher margins, resulting in more regeneration and a higher cost solution. This problem is exacerbated with higher bit rates (40G+) when more accurate feasibility analysis may be needed due the higher transmission challenges. Moreover, if no feasible optical path exists between the endpoints, planning tools cannot identify the currently available regenerators if they do not have real time data from the network, and therefore require multiple tools (e.g., also an EMS) to be reconciled manually to determine the final path.
2. Issues with online planning tools.
   To address the above issues, the planning tool must be constantly online, however this represents another point of failure in the network and requires extra overhead, e.g., for backing it up. This is not desirable for many operators. So can the planning tool be integrated into the EMS, to reduce the overhead for a standalone tool? In certain cases it can, but many operators may prefer using a higher level generic management tool, which is not provided by the DWDM layer vendor.
3. Issues with pre-determining all feasible paths a-priori.
   While the above approaches assume on-demand computation of a feasible path, a different way to tackle the problem is to pre-define a small number of feasible paths between all required end-points a-priori. In this case, all the control plane needs to do is select one of the feasible paths and set up a connection. This approach will work in small networks, but the number of feasible paths that need to be maintained in a large network may be prohibitive as it requires O(N^2) paths. Another issue with this approach is that it reduces the flexibility of choosing a path as not all feasible paths can be defined a-priori (otherwise the number of paths becomes exponential). This will create artificial blocking, when an alternate non-blocking path may exist. An even worse problem is that when the optical layer evolves (e.g., a fiber added), all paths have to be re-computed which makes this set of paths operationally very hard to maintain.

In summary, while there are ways to provide a control plane for an optical network that is not aware of impairments, such a solution has various limitations that imply either high capital cost or high operational cost neither of which are acceptable in many SP networks that are under pressure to optimize both CAPEX and OPEX.

3.  Overview of Optical Impairment Evaluation

3.1.  Impairments Evaluation Flow

The aim of this section is to provide an overview of a typical decision flow for the evaluation of the feasibility of an optical path. The feasibility is evaluated given the transmitter and receiver characteristics, the characteristics of intermediate nodes, and the optical impairments along the path from the lightpath source to its destination.

 +--------+           +--------+           +--------+
 |        |           |        |           |        |
 |  Node  #-----------|  Node  |-----------#  Node  |
 |        |    Link   |        |   Link    |        |
 +--------+           +--------+           +--------+

         TX                               RX
      Interface                       Interface

Figure 1 represents DWDM transparent network can be represented by nodes, links and interfaces.

• Node. It's an optical network element providing wavelength switching and multiplexing functionality. The WSON framework document [I-D.draft-ietf-ccamp-wavelength-switched-framework] completely describes different node types with their peculiarities. An optical node can perturb a lightpath quality by adding impairments due to its internal components. A node may also change the value of some parameters like the optical power.
• Link. The fiber is the physical medium of a lightpath and it adds impairments to the optical signals. In general it causes a degradation in the signal quality.
• Interface. Optical Interfaces contains both a transmitter (TX-Interface) and a receiver (RX-Interface). Depending on its function different parameters must be considered.

The measurement of a BER or a Q-Factor completely describes the quality of an optical signal. However in transparent optical networks there is no direct measurement of such information. Very often the only way is to provide measurement of other parameter's (i.e. impairments) and provide an estimation of the signal quality.

Following Figure 1 the signal get generated at the TX-Interface with certain characteristics. Let's consider, for purposes of illustration, only a couple of simple parameters: the power of the signal and the signal to noise ratio (OSNR). Along the lightpath, the signal goes though fiber which reduces its power and introduces impairments that can be viewed as additional noise added by the fiber characteristics. Traversing an optical node the signal might be amplified so it recover in term of power but this might cause additional noise to the signal. Optical components (e.g. switches) within the node are themselves source of additional noise for the signal. When the signal reaches the RX-Interface at the destination it must have sufficient power and sufficient OSNR to be used by the interface. The acceptable level of the OSNR at the destination as well as the minimum acceptable power might depend on the characteristics of the receiver interface.

The next Figure 2 provides the functional overview of such evaluation process.

+-----------------+        +-----------------+        +------------+
|                 |        |                 |        |            |
| Optical         |        | Optical         |        | Optical    |
| Interface       |------->| Path            |------->| Channel    |
| Characteristics |        | Characteristics |        | Estimation |
|                 |        |                 |        |            |
+-----------------+        +-----------------+        +------------+
                                                          ||
							  ||
                                                       Estimation
						          ||
							  \/
                                                     +------------+
						     |  BER /     |
						     |  Q Factor  |
						     +------------+

Starting from the left the Optical Interface Characteristics represents where the optical signal is transmitted or received and represent the properties at the end points of a lightpath. In principle a path computation with no impairments (RWA only) might use only a minimum set of these parameters to assess the interface compatibility. If impairments are considered additional parameters become interesting. Section 5 (Optical Interface Characteristics) details the parameters related to the interfaces.

Within the block Optical Path Characteristics we represents all kind of impairments affecting a lightpath while it traverse the networks through links and nodes. Section 6 (Optical Path Characteristics) reports a list of of such effects.

When reaching the destination node, an impairment-aware constrained path computation must take a decision if the lightpath is feasible or not. We call this operation Optical Channel Estimation because the real signal quality must be estimated given the impairments along the path. Section 7 (Optical Channel Estimation) reports a set of parameters that can be used to asses such signal quality.

3.2.  Parameter Classification According to its Information Complexity

As discussed in the previous section, the process of determining the feasibility of an optical path includes constraints, impairments and other data that relates to the transmitter (such as modulation format, optical power, supported wavelengths etc.), data that relates to the optical nodes along the path (loss, gain etc.), information that relates to the fiber links between the nodes (fiber loss, dispersion etc), information that relates to other paths that interact with the new path, and finally the capabilities of the receiver (power sensitivity, error correction etc.).

However, not all the data must be shared over the control plane as is. We distinguish between the following cases:

• Some attributes can be kept local to the node. For example, receiver attributes may be kept at the end node, if this node will make the decision on path feasibility. We call these attributes LOCAL in this document.
• Other attributes can be shared in a summarized form since the impairment can be accumulated hop by hop. For example optical power can be added or reduced starting from the power at the transmitter as the signaling message traverses various gain or loss elements along the path. These attributes are termed SCALAR herein
• Some attributes need to be calculated based on all the values of the specific attribute for all the hops along the path. Such values cannot be summarized and need to accumulate in a vector reflecting all the hops along the path. These attributes are termed MULTI-HOP herein (these attributes are called non-linear in the optical jargon)
• Some values are impacted not just by the new path under consideration but also by other channels that interact with it along its route. The most obvious example for such an attribute is wavelength: the available wavelength is a function of other channels at nodes and links along the path. These attributes are termed MULTI-CHANNEL herein.
• Finally, some attributes can be computed accurately only if values are known for every channel and every hop along the path. These attributes are termed MULTI-HOP-CHANNEL herein.

The main contribution of this document is the classification of the various attributes along these cases (LOCAL, SCALAR, MULTI-HOP, MULTI-CHANNEL and MULTI-HOP-CHANNEL). We believe that this information is crucial in order to determine the control plane mechanism needed to carry each attribute. To illustrate this let's assume a particular attribute must be carried in the path signaling message. The attribute will accumulate as follows for the different attribute classes.

• LOCAL attributes will not be carried in the message
• SCALAR attributes will be carried via a single field, with agreed upon rules on how to compute the next value based on the value from the incoming message and the value for the next node and/or link
• MULTI-HOP attributes must be carried in a vector that fills up in serial fashion from hop to hop, in which the value for the n-th hop is carried in v[n]
• MULTI-CHANNEL attributes must be carried in a vector that contains one value per channel and in which all values are updated in parallel at every hop (a good example for this is the wavelength bitmap defined in [ref TBD], in which one bit represents the summarized information on the availability of a specific wavelength along the path), and
• MULTI-HOP-CHANNEL attributes requires a matrix m[h,w] with h rows (one per hop) and w columns (one per wavelength), in which row m[n,*] represents the values for all channels for the n-th hops along the path

3.3.  Additional Parameter Sharing Properties

For the usage in a control plane, other properties for each parameter should also be considered. The association between each parameter and a set of its inherent properties is defined within ITU documents. For a control plane usage however the following value of each property should help in deciding the most appropriate protocol to convey the parameter within he control plane.

• Time Dependency.
  If a parameter is static then we assume it does not change during the life of the network. In case there is a time dependency a minimum refresh rate shall be provided. This may help determine, for example whether the parameter should be considered once during connection setup, or whether the properties should be refreshed (e.g. via an IGP or signaling).
• Wavelength Dependency.
  A parameter can be wavelength dependent if, for each wavelenght it might have different value. On the contrary a parameter might have a single value for all the wavelengths.
• Value type (real, discrete, etc.).
  This kind of information should help in encoding the information within the protocol. Because some of the parameters represent physical quantities discrete values might be used within certain approximations, in other cases real (e.g. IEEE 754 standard) might be required.
• Does the variance of the parameter need to be specified with it (so an error may be computed along with it)?

4.  Control Plane Considerations

[Editorial Note: This section has to be filled up. Current text only as initial placeholder].

The use of optical impairments as path constrains would imply the control plane (GMPLS) to be aware of some additional information coming from the optical layer. The control plane shall be able to convey the proper information for an to allow the optical path feasibility calculation.

• Routing.
  Routing extensions to support GMPLS are defined in [RFC4202] (Kompella, K. and Y. Rekhter, “Routing Extensions in Support of Generalized Multi-Protocol Label Switching (GMPLS),” October 2005.). Needs additional information to defines routing metrics with a certain level of optical-impairment awareness.
• Signaling.
  Optical constrains can be verified along the signaling phase. Different options might be foreseen:
  • Full optical impairment verification. If the routing phase has no awareness of any optical impairments, the signaling phase can verify all of them.
  • Partial optical impairment verification. It might be possible that a certain set of optical impairment are verified during the routing phase while the remaining set can be performed during the signaling phase.
• PCE Architecture [RFC4655] (Farrel, A., Vasseur, J., and J. Ash, “A Path Computation Element (PCE)-Based Architecture,” August 2006.).
  In case a PCE is used to support path computation, an additional solution might foresee a PCE capable of supporting an impairment-aware path computation. Need to evaluate the information to convey to the PCE for such computation.

5.  Optical Interface Characteristics

Placeholder for details optical parameters that specifically refer to the physical interface. Their knowledge is necessary to evaluate the path feasibility in term of optical impairments.

They can be classified for what related to transmitter (e.g. bit rate, modulation format, FEC etc.) and receiver (e.g. stability, CD robustness) interface. The document [ITU.G698.2] (International Telecommunications Union, “Amplified multichannel DWDM applications with single channel optical interfaces,” July 2007.) provide a detailed list of parameteres with their values.

6.  Optical Path Characteristics

This section describes the optical impairments associated to the path optical elements defined in Section 3 (Overview of Optical Impairment Evaluation). For each impairment, the following information is provided:

• A minimum impairment description with a reference to the related ITU-T document where available.
• the impairment classification and properties as defined within Section 3.2 (Parameter Classification According to its Information Complexity) and Section 3.3 ( Additional Parameter Sharing Properties).
• the impairment evaluation/handling techniques.
• the impairment impact on the channel quality estimation as defined by parameters in Section 3.3.

6.1.  Linear Impairments

6.1.1.  Fiber Losses

It's the optical power loss caused by a fiber span [ITU-T ref ?]. It's measured as the ratio (dB) between the input and the output optical power.

The loss introduced by each fiber span depends on:

• fiber attenuation coefficient (dB/Km)
• fiber length (Km). Note that this value is one of the link characteristics defined in [RFC4209] (Fredette, A. and J. Lang, “Link Management Protocol (LMP) for Dense Wavelength Division Multiplexing (DWDM) Optical Line Systems,” October 2005.) (section 2.3.5).

The fiber losses directly affect the signal optical power at the receiver interface: losses are accumulated along the path and compensated by Optical Amplifiers.

The fiber loss can be computed from the fiber nominal values (length and attenuation coefficient), measured by the node (using a probe signal) or provisioned as a link parameter.

The impairment is SCALAR, wavelength independent, time independed (apart from fiber aging but only on long period of times).

6.1.2.  Insertion Losses (Optical components looses)

It's the optical power loss caused by the optical elements crossed by the channel in a node. It's measured as the ratio (dB) between the optical power at the input and output port (see [ITU.G671] (International Telecommunications Union, “Transmission characteristics of optical components and subsystems,” Jannuary 2005.) Section 3.2.9). The possible channel paths in a node are (see [ITU.G680] (International Telecommunications Union, “Physical transfer functions of optical network elements,” July 2007.) Section 3.2.1):

• from the input port to the output port for through channels
• from the input port to the drop port for dropped channels
• form the add port to the output port for added channels

This classification applies to any type of node: PXC, ROADM, etc. The loss value for each path is dependent on the internal node architecture and vendor device characteristics and could be different for each different port pair.

The insertion losses directly affect the signal optical power at the receiver interface: losses are accumulated along the path and compensated by Optical Amplifiers.

The impairment is SCALAR, wavelength independent and time independent.

6.1.3.  Amplifier Spontaneous Emission (ASE)

It's the noise introduced by the node optical amplifiers spontaneous emission that propagates with the amplified signal and is further amplified along the path (see [ITU.G663] (International Telecommunications Union, “Application related aspects of optical amplifier devices and sub-systems,” April 2000.) Section II.6.1). The impairment is considered an OSNR contribution (dB) and directly affects the signal SNR at the receiver interface (see [ITU.G663] (International Telecommunications Union, “Application related aspects of optical amplifier devices and sub-systems,” April 2000.) Section II.6.2).

The ASE noise contribution can be evaluated from the amplifier parameters and the input signal characteristics (signal optical power).

The impairment is SCALAR and wavelength independent. Regarding time-dependency ASE depends on amplifier gain and this may change depending on network stability.

6.1.4.  Crosstalk

It's the effect of signal power leakage from other channels inside the node optical elements (multiplexer, de-multiplexer, optical switches, etc.) and is measured as the ratio of the disturbing power and the signal power (dB) (see [ITU.G692] (International Telecommunications Union, “Optical interfaces for multichannel systems with optical amplifiers,” October 1998.) Section 6.7.1).

The crosstalk value is dependent on the device characteristics and the ratio of the optical power of involved channels. The device characteristics can be considered constant in time, while the channels configuration depends on the network status.

The crosstalk impairment affects the signal SNR and optical power at the receiver interface: the accumulated crosstalk can be converted to an SNR and power penalty.

The impairment is MULTI-CHANNEL (depends on existings active channels) and wavelength independent.

6.1.5.  Fiber Chromatic Dispersion

It's the degradation of the optical signal due to the different propagation delay of the various spectral components causing the broadening of the pulse and is defined by the slope of the delay with respect to the wavelength (ps/nm) (see [ITU.G650] (International Telecommunications Union, “Definition and test methods for the relevant parameters of single-mode fibres,” March 1993.) Section 1.5). The effect can be compensated by means of fiber spans with an inverse dispersion (DCF - Dispersion Compensation Fiber) usually deployed in modules with pre-configured characteristics (DCU - Dispersion Compensation Unit).

The chromatic dispersion introduced by each fiber span is dependent on:

• fiber chromatic dispersion coefficient (ps/(nm * Km)) This coefficient depends on the channel wavelength (it's usually defined as the value for a reference wavelength and a slope coefficient).
• fiber length (Km).

The chromatic dispersion of each fiber span can be evaluated from the fiber characteristics and the channel wavelength.

The chromatic dispersion accumulates along the path compensated by the DCU modules obtaining a residual chromatic dispersion at the receiver interface the affects the signal SNR.

The impairment is SCALAR, Wavelength-Dependent and time-independent (apart from ageing effects over long period of times).

6.1.6.  Polarization Mode Dispersion (PMD)

It's the degradation of the optical signal due to the different propagation delay of the two principal states of polarization (DGD - Differential Group Delay) causing the pulse distortion in shape and width (see [ITU.G663] (International Telecommunications Union, “Application related aspects of optical amplifier devices and sub-systems,” April 2000.) Section II.4.1 and [ITU.G661] (International Telecommunications Union, “Definitions and test methods for the relevant generic parameters of optical amplifier devices and subsystems,” July 2007.) Section 5.1.36) and is measured in ps.

The PMD introduced by a fiber span depends on:

• the square root of the fiber length (Km)
• the fiber PMD coefficient (ps/sqrt(Km))

The PMD introduced by other components (e.g. amplifiers, DCU modules, etc.) is provided as a device parameter.

The PMD coefficient may depend on temperature and operating conditions and can be very variable.

The PMD can be evaluated from the fiber or device parameters; due to the high variability, an upper bound value is usually considered.

The PMD accumulates along the path and affects the signal SNR at the receiver interface.

The impairment is SCALAR.

6.1.7.  Polarization Dependent Loss (PDL)

It's the difference in the signal power among different polarization states caused by the medium irregularities (see [ITU.G663] (International Telecommunications Union, “Application related aspects of optical amplifier devices and sub-systems,” April 2000.) Section II.4.1 and [ITU.G671] (International Telecommunications Union, “Transmission characteristics of optical components and subsystems,” Jannuary 2005.) Section 3.2.23) and is defined as the ratio (dB) of the maximum and minimum peak transmission power with respect to all polarization states. On amplifiers the same effect is called Polarization Dependent Gain (PDG) (see [ITU.G661] (International Telecommunications Union, “Definitions and test methods for the relevant generic parameters of optical amplifier devices and subsystems,” July 2007.) Section 5.1.11).

The PDL introduced by a device appears as a random variation of the signal power and can be managed as a statistical value (function of the number of optical elements traversed) (see [ITU.G680] (International Telecommunications Union, “Physical transfer functions of optical network elements,” July 2007.) Section 9.3.2).

The PDL accumulates along the path and affects the signal SNR and power at the receiver interface.

The impairment is and SCALAR wavelength independent.

6.2.  Fiber Optical Non-Linearities

In this section provide information about optical impairments called Non-Linear. In general they will need a control plane able to exploit multi-channel and multi-hop attibutes. Due to this complexity leave this placeholder for further updates.

• Self Phase Modulation (SPM) (see [ITU.G663] (International Telecommunications Union, “Application related aspects of optical amplifier devices and sub-systems,” April 2000.) Section II.3.1).
  It's the degradation of the optical signal due to the time varying fiber refraction index (the higher intensity portion of the pulse is subject to a higher refractive index than the lower intensity portion) causing the pulse broadening in the frequency domain.
• Cross Phase Modulation (XPM) (see [ITU.G663] (International Telecommunications Union, “Application related aspects of optical amplifier devices and sub-systems,” April 2000.) Section II.3.3).
  It's the degradation of the optical signal due to the time varying fiber refraction index induced by the adjacent channels intensity fluctuations.
• Four Wave Mixing (FWM) (see [ITU.G663] (International Telecommunications Union, “Application related aspects of optical amplifier devices and sub-systems,” April 2000.) Section II.3.5).
  It's the degradation of the optical signal due to the interaction with the spurious optical wavelengths generated by three optical channels that co-propagate inside a fiber.
• Stimulated Brillouin Scattering (SBS) (see [ITU.G663] (International Telecommunications Union, “Application related aspects of optical amplifier devices and sub-systems,” April 2000.) Section II.3.6).
  It's the degradation of the optical signal due to the signal scattering caused by the fiber medium when the input power is above the SBS threshold.
• Stimulated Raman Scattering (SRS) (see [ITU.G663] (International Telecommunications Union, “Application related aspects of optical amplifier devices and sub-systems,” April 2000.) Section II.3.6).
  It's the degradation of the optical signal due to the signal scattering caused by the fiber medium when the input power is above the SBS threshold.

7.  Optical Channel Estimation

The lightpath quality defines the ability of the receiver to correctly decode the signal within a defined error rate (BER). The BER depends on the signal encoding (e.g. FEC, modulation format, etc.), the signal optical characteristics (optical power, OSNR, etc.) and the receiver characteristics.

The goal of this section is to define the set of parameters that need to be evaluated in order to get a direct estimation of the lightpath quality at the receiver node; all the above described impairments can be considered as an effect on these parameters.

These optical parameters shall be considered together with their statistical values: average and variance. This information helps in understanding the error accumulated along with parameter evaluation.

7.1.  Optical Power

The signal optical power must be within the dynamic range of the receiver and above the receiver minimum power (receiver sensitivity). The receiver minimum power is a receiver characteristic and depends also on the signal characteristics as the used forward error correction (FEC), [others ? mod-format ?]

[Q3] (GM: Need to add a formal definition with the ITU reference)

7.2.  Optical Signal to Noise Ratio

The optical signal-to-noise ratio (OSNR) is the ratio of the signal power in the wanted channel to the highest noise power density (referred to 0.1 nm) within the channel frequency range (see [ITU.G661] (International Telecommunications Union, “Definitions and test methods for the relevant generic parameters of optical amplifier devices and subsystems,” July 2007.) Section 5.1.19). The signal OSNR must be above a receiver minimum threshold (receiver sensitivity); the threshold is a receiver characteristic and depends also on the signal characteristics ([which ones ?]).

7.3.  Residual Chromatic Dispersion

The residual chromatic dispersion is the signal chromatic dispersion at the receiver interface (the accumulated and compensated dispersion along). The signal dispersion value must be within a receiver threshold for the signal to be correctly decoded.

The chromatic dispersion effects on the signal quality can be managed as an OSNR or/and an optical power penalty.

7.4.  Residual Differtial Group Delay (DGD)

[Editoral Note: TO BE FILLED]

7.5.  Q-Factor

The Q-factor is a synthetic measure of the signal quality defined as a function of the mean values of the '0' and '1' signal levels and their standard deviation (see [ITU.G976] (International Telecommunications Union, “Test methods applicable to optical fibre submarine cable systems,” July 2007.) Section A.1).

The Q-factor is dependent on the signal OSNR and is directly related to the BER.

The evaluation of the Q-factory requires the estimation of the signal waveform at the receiver interface.

8.  Acknowledgements

Authors would like to thanks Adrian Farrel for all suggestions and reviews of this work.

9.  Contributing Authors

This document was the collective work of several authors. The text and content of this document was contributed by the editors and the co-authors listed below (the contact information for the editors appears in appropriate section and is not repeated below):

   Gabriele Maria Galimberti
   Cisco Systems
   via Philips 12
   Monza  20052
   Italy

   Email: ggalimbe@cisco.com

   Domenico La Fauci                      Maurizio Gazzola
   Cisco Systems                          Cisco Systems
   via Philips 12                         via Philips 12
   Monza  20052                           Monza  20052
   Italy                                  Italy

   Email: dlafauci@cisco.com              Email: mgazzola@cisco.com

   Roberto Cassata                        Zafar Ali
   Cisco Systems                          Cisco Systems
   via Philips 12                         3000 Innovation Drive
   Monza  20052                           Kanata , ONTARIO K2K 3E8
   Italy                                  Canada

   Email: rcassata@cisco.com              Email: zali@cisco.com


   Elio Salvadori                         Yabin  Ye
   CREATE-NET                             CREATE-NET
   via alla Cascata 56 C, Povo            via alla Cascata 56 C, Povo
   Trento  38100                          Trento  38100
   Italy                                  Italy

   Email: elio.salvadori@create-net.org   Email: yabin.ye@create-net.org


   Chava Vijaya Saradhi
   CREATE-NET
   via alla Cascata 56 C, Povo
   Trento  38100
   Italy

   Email: saradhi.chava@create-net.org

   Ernesto Damiani
   University of Milan, Department of Information Technology
   Via Bramante 65, 26013 Crema (CR)
   Italy

   Email: damiani@dti.unimi.it

   Valerio Bellandi
   University of Milan, Department of Information Technology
   Via Bramante 65, 26013 Crema (CR)
   Italy

   Email: bellandi@dti.unimi.it

   Marco Anisetti
   University of Milan, Department of Information Technology
   Via Bramante 65, 26013 Crema (CR)
   Italy

   Email: anisetti@dti.unimi.it

10.  IANA Considerations

This memo includes no request to IANA.

11.  Security Considerations

All drafts are required to have a security considerations section. See RFC 3552 (Rescorla, E. and B. Korver, “Guidelines for Writing RFC Text on Security Considerations,” July 2003.) [RFC3552] for a guide.

12.  References

12.1. Normative References

12.2. Informative References

Appendix A.  ITU Parameters Missing Information

In this appendix we need to collects all information about optical parameters that need to be verified with / requested from ITU.

Authors' Addresses

Full Copyright Statement

Copyright © The IETF Trust (2008).

This document is subject to the rights, licenses and restrictions contained in BCP 78, and except as set forth therein, the authors retain all their rights.

This document and the information contained herein are provided on an “AS IS” basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.

Intellectual Property

The IETF takes no position regarding the validity or scope of any Intellectual Property Rights or other rights that might be claimed to pertain to the implementation or use of the technology described in this document or the extent to which any license under such rights might or might not be available; nor does it represent that it has made any independent effort to identify any such rights. Information on the procedures with respect to rights in RFC documents can be found in BCP 78 and BCP 79.

Copies of IPR disclosures made to the IETF Secretariat and any assurances of licenses to be made available, or the result of an attempt made to obtain a general license or permission for the use of such proprietary rights by implementers or users of this specification can be obtained from the IETF on-line IPR repository at http://www.ietf.org/ipr.

The IETF invites any interested party to bring to its attention any copyrights, patents or patent applications, or other proprietary rights that may cover technology that may be required to implement this standard. Please address the information to the IETF at ietf-ipr@ietf.org.