maximum channel insertion loss deviation maximum variation of the insertion loss absolute value within the passband channel frequency range for a DWDM device or channel wavelength range
Basic term definitions
3.1.1 port optical fibre or optical fibre connector attached to a passive device for the entry and/or exit of the optical power
The optical properties of a fiber optic wavelength-selective branching device can be characterized using an n x n transfer matrix of coefficients In this matrix, n represents the number of ports, and the coefficients indicate the fractional optical power transferred among the specified ports.
The transfer matrix is thoroughly explained in Annex A, where the ports are sequentially numbered to illustrate all ports and their possible combinations It is important to note that the numbering of the ports is arbitrary.
Figure 1 illustrates a six-port WDM device featuring two input ports and four output ports This device can also function with four input ports and two output ports due to its reciprocity characteristics Additionally, users can select various combinations of input and output ports, such as one input port with five output ports or three input ports with three output ports, making it particularly suitable for bi-directional transmission system applications For more details, refer to Annex B.
Figure 1 – Example of a six-port device, with two input and four output ports
When there are four operating wavelengths, the transfer matrix is represented as a 6 × 6 × 4 matrix For instance, the optical attenuation at wavelength λ1 from port 1 to port 6 utilizes a value of 161, while the return loss of port 2 at wavelength λ4 employs a value of 224 Additionally, the optical attenuation from port 5 to port 2 at wavelength λ3 is represented by a value of 523.
3.1.3 transfer matrix coefficient element t ij of the transfer matrix
Note 1 to entry: t ij is the number of more than or equal to zero, and less than or equal to one
Note 2 to entry: A detailed explanation is shown in Annex A
3.1.4 logarithmic transfer matrix transfer matrix whose matrix element a ij is a logarithmic value of transfer matrix element t ij a ij is a number of positive and expressed in dB
Note 1 to entry: A detailed explanation is shown in Annex A
3.1.5 conducting port pair port pair consisting of i and j where t ij is nominally greater than zero (ideally t ij is 1 and a ij is
3.1.6 isolated port pair pair i and j consisting where t ij is nominally zero, and a ij is nominally infinite at a specified wavelength
3.1.7 channel wavelength (frequency) band in which an optical signal is transmitted for a WDM device
Note 1 to entry: WDM devices have two or more channels
3.1.8 channel spacing centre-to-centre differences in frequency or wavelength between adjacent channels in a WDM device
Component definitions
A wavelength-selective branching device is a passive component featuring three or more ports that distributes optical power among its ports in a predetermined manner This device operates without amplification or active modulation, relying solely on the wavelength It facilitates the nominal transfer of at least two different wavelength ranges between distinct pairs of ports.
3.2.2 wavelength division multiplexing device wavelength division multiplexer
WDM device synonym for a wavelength-selective branching device
Note 1 to entry: The term of wavelength-selective device is the contrast with the term of non-wavelength-selective branching device The term of WDM device is frequently used
3.2.3 dense wavelength division multiplexing device
WDM device which is intended to operate for a channel spacing equal or less than 1 000 GHz (approximately 8 nm at 1 550 nm and 5,7 nm at 1 310 nm)
3.2.4 coarse wavelength division multiplexing device
WDM device which is intended to operate for channel spacing less than 50 nm and greater than 1 000 GHz
WDM device which is intended to operate for channel spacing equal to or greater than 50 nm
A MUX WDM (DWDM, CWDM, or WWDM) device features multiple input ports and a single output port, designed to combine various optical signals that are distinguished by their wavelengths from the corresponding input ports into one unified output.
A Wavelength Division Multiplexing (WDM) device, including types such as Dense Wavelength Division Multiplexing (DWDM), Coarse Wavelength Division Multiplexing (CWDM), or Wide Wavelength Division Multiplexing (WWDM), features one input port and multiple output ports Its primary function is to separate multiple optical signals, each distinguished by different wavelengths, from a single input port to their respective output ports.
A DWDM device with three ports is designed to separate multiple optical signals based on their wavelengths from a common port It alternately transmits odd channel signals to one branching port and even channel signals to the other, ensuring efficient signal management and distribution.
Note 1 to entry: An interleaver can operate as a wavelength multiplexer (OMUX) by reversing the demultiplexer.
Performance parameter definitions
3.3.1 operating wavelength nominal wavelength λ h at which a WDM device operates with the specified performance
Note 1 to entry: The term "operating wavelength" includes the wavelength to be nominally transmitting, designated attenuating and isolated
Note 2 to entry: Operating frequency is also used for DWDM devices
3.3.2 operating wavelength range specified range of wavelengths including all operating wavelengths
Note 1 to entry: It includes all passbands and isolation wavelength ranges corresponding to all channels
Note 2 to entry: The term "operating wavelength range" is defined for a WDM device, not for each channel or port
3.3.3 channel wavelength range range within which a CWDM or WWDM device operates with less than or equal to a specified optical attenuation for the conducting port pair
For a specific nominal channel center wavelength, denoted as \$\lambda_{nom}\$, the corresponding wavelength range extends from \$\lambda_{imin} = (\lambda_{nom} - \Delta\lambda_{max})\$ to \$\lambda_{imax} = (\lambda_{nom} + \Delta\lambda_{max})\$, where \$\Delta\lambda_{max}\$ represents the maximum deviation from the nominal wavelength.
Note 2 to entry: For CWDM devices, channel centre wavelengths and maximum channel centre wavelength deviations are defined as nominal central wavelengths and wavelength deviations in ITU-T G 694.2
Note 3 to entry: An illustration of channel wavelength range is shown in Figure 2
O pt ic al at tenu at io n (d B)
Channel wavelength range for channel k λ λ λ k kmin kmax
Figure 2 – Illustration of channel wavelength range
3.3.4 channel frequency range frequency range within which a DWDM device is required to operate with less than or equal to a specified optical attenuation for the conducting port pair
Note 1 to entry: For a particular nominal channel frequency, f nomi , this frequency range is from f imin = (f nomi -
∆ f max ) to fimax = (f nomi + ∆ f max ), where ∆ f max is the maximum channel centre frequency deviation
Note 2 to entry: Nominal channel centre frequency and maximum channel centre frequency deviation are defined in ITU-T G.694.1
3.3.5 passband channel passband synonym for channel wavelength range (channel frequency range)
Note 1 to entry: Passband is frequently used
Note 2 to entry: There are two or more passbands for WDM devices Each passband is defined corresponding to each channel
3.3.6 insertion loss maximum value of a ij (where i ≠ j) within the passband for conducting port pair
Insertion loss refers to the optical attenuation measured from a specific port to another port within a conducting port pair of a Wavelength Division Multiplexing (WDM) device This loss is expressed as a positive value in decibels and is calculated accordingly.
P in is the optical power launched into the port;
P out is the optical power received from the other port of the conducting port pair
Note 2 to entry: An illustration of insertion loss is shown in Figure 3
O pt ic al at tenu at io n (d B) Passband channel h for Passband channel k for λ λ λ k kmin kmax
M ax im um ins er tion l os s for c hannel k
M ax im um ins er tion los s for c hannel h
Figure 3 – Illustration of insertion loss
Note 3 to entry: For a WDM device, the insertion loss shall be specified as a maximum value of the insertion losses of all channels
Channel insertion loss refers to the loss of signal strength in a specific channel of Wavelength Division Multiplexing (WDM) devices While it shares a similar meaning with general insertion loss, channel insertion loss specifically pertains to individual channels, whereas insertion loss is a broader term used in the specifications of both WDM devices and their channels.
3.3.8 passband ripple maximum peak-to-peak variation of the insertion loss (absolute value) over the passband (within a channel frequency or wavelength range) (refer to Figure 4 below)
Channel frequency (or wavelength) range
O pt ic al at tenu at io n (d B) Channel frequency
Figure 4a – Ripple at band edges Figure 4b – Ripple in band
The maximum channel insertion loss deviation refers to the maximum absolute variation of insertion loss within the passband, which is the frequency range for Dense Wavelength Division Multiplexing (DWDM) devices or the wavelength range for Coarse Wavelength Division Multiplexing (CWDM) and Wide Wavelength Division Multiplexing (WWDM) devices.
Note 1 to entry: Channel insertion loss deviation should not to be confused with ripple defined in Figure 5 below
Maximum channel insertion loss deviation
Channel centre frequency (or wavelength)
Channel frequency range for DWDM devices, channel wavelength range for CWDM and WWDM devices)
Insertion loss characteristics for three temperatures
0 Frequency (THz) for DWDM devices, wavelength (nm) for CWDM and WWDM devices
O pt ic al at tenu at io n (d B)
Figure 5 – Illustration of channel insertion loss variation
3.3.10 channel non-uniformity insertion loss channel non-uniformity for a specified set of branching ports the difference between the maximum and the minimum insertion loss at the common port
Note 1 to entry: Channel non-uniformity is defined for a MUX (N x 1 WDM device) and a DEMUX (1 x N WDM device) Channel non-uniformity is a positive value, and expressed in dB
For CWDM and DWDM devices, channel non-uniformity refers to the variation in insertion loss, specifically the difference between the maximum and minimum insertion loss at the nominal wavelengths (frequencies) across all channels.
The centre wavelength deviation refers to the difference between the centre wavelength and the nominal wavelength (or frequency) of a specified channel in DWDM devices The centre wavelength is defined as the midpoint of the wavelength range that is x dB below the minimum optical attenuation for the designated passband (channel).
Note 1 to entry: 0,5, 1 or 3 are generally used for x
3.3.12 crosstalk for WDM devices, the value of the ratio between the optical power of the specified signal and the specified noise
Crosstalk, expressed as a negative value in dB, is defined for each output port of WDM devices, specifically for DEMUX (1 x N WDM devices) It is calculated by subtracting the insertion loss from the conducting port pair (port i to port o) from the isolation of the isolated port pair (port j to port o).
WDM devices is defined for a DEMUX (1 x N WDM device) For an MxN WDM device, crosstalk can be defined to as expanding M of a 1 x N WDM device
Note 2 to entry: For WDM devices with three of more ports, the crosstalk should be specified as the maximum value of the crosstalk for each output port
Note 3 to entry: Care should be taken not to confuse crosstalk and isolation
3.3.13 isolation minimum value of a ij (where i ≠ j) within isolation wavelength range for isolated port pair
Note 1 to entry: Isolation is a positive value expressed in dB
The isolation wavelength for a pair of ports \(i\) and \(j\) (where \(i \neq j\)) is defined at a wavelength \(\lambda_h\) This nominal wavelength \(\lambda_k\) represents an operating wavelength for a different pair of ports, ensuring that ports \(i\) and \(j\) remain isolated from each other.
Note 1 to entry: Isolation frequency is also used for DWDM device
O pt ic al at tenu at io n (d B)
Figure 6 – Illustration of isolation wavelength
The isolation wavelength range for a conducting port pair \(i\) and \(j\) at wavelength \(\lambda_h\) is defined as the spectrum of wavelengths from \(\lambda_{k_{\text{min}}}\) to \(\lambda_{k_{\text{max}}}\), centered around an operating wavelength \(\lambda_k\) This operating wavelength corresponds to a different pair of ports, ensuring that ports \(i\) and \(j\) remain isolated.
Note 1 to entry: Isolation frequency range is also used for DWDM devices ij im a a λ k
O pt ic al at tenu at io n (d B)
Figure 7 – Illustration of isolation wavelength range
3.3.16 wavelength isolation value of a ij (where i ≠ j) in the isolation wavelength range
Note 1 to entry: The wavelength isolation shall be defined as the minimum value of wavelength isolation over the isolation wavelength range
3.3.17 adjacent channel isolation isolation with the restriction that x, the isolation wavelength number, is restricted to the channels immediately adjacent to the (channel) wavelength number associated with port o
Note 1 to entry: Adjacent channel isolation is a positive value expressed in dB
Adjacent channel isolation differs from adjacent channel crosstalk, as illustrated in Figure 8 The figure shows an upward-pointing arrow for positive values and a downward-pointing arrow for negative values Typically, there are two adjacent channel isolations: one for the shorter wavelength (higher frequency) side and another for the longer wavelength (lower frequency) side.
3.3.18 adjacent channel crosstalk crosstalk with the restriction that x, the isolation wavelength number, is restricted to the channels immediately adjacent to the (channel) wavelength number associated with port o
Note 1 to entry: Adjacent channel crosstalk is a negative value expressed in dB
Adjacent channel crosstalk differs from adjacent channel isolation, as illustrated in Figure 8 The figure shows that the upward-pointing arrow represents a positive value, while the downward-pointing arrow indicates a negative value Typically, there are two instances of adjacent channel crosstalk: one for the shorter wavelength (higher frequency) side and another for the longer wavelength (lower frequency) side.
O pt ic al at tenu at io n (d B)
Frequency (THz) for DWDM devices, wavelength (nm) for CWDM and WWDM devices
Non-adjacent channel centre frequency
Adjacent channel centre frequency (wavelength)
A dj ac en t c hannel is ol at ion
Non-adjacent channel centre frequency (wavelength)
Adjacent channel centre frequency (wavelength)
A dj ac en t c hannel c ros st al k
Figure 8 – Illustration of adjacent channel isolation
Non-adjacent channel isolation refers to the restriction of isolation wavelengths (or frequencies) to channels that are not immediately adjacent to the channel associated with port O, as illustrated in Figure 9 below.
Non-adjacent channel isolation differs from non-adjacent channel crosstalk, as illustrated in Figure 9, where the upward-pointing arrow signifies a positive value and the downward-pointing arrow represents a negative value.
Non-adjacent channel crosstalk refers to the interference that occurs when the isolation wavelength (or frequency) is limited to channels that are not directly adjacent to the channel linked with port o This phenomenon is illustrated in Figure 9 below.
Non-adjacent channel crosstalk differs from non-adjacent channel isolation, as illustrated in Figure 9, where the upward-pointing arrow signifies a positive value and the downward-pointing arrow represents a negative value.
O pt ic al at tenu at io n (d B)
No n- adj ac ent c hannel c ros st al k a iox a ioc
No n- adj ac ent c hannel is ol at ion
Non-adjacent channel centre frequency
Non-adjacent channel centre frequency (wavelength)
Adjacent channel centre frequency (wavelength)
Adjacent channel centre frequency (wavelength)
Frequency (THz) for DWDM devices, wavelength (nm) for CWDM and WWDM devices
Figure 9 – Illustration of non-adjacent channel isolation
3.3.21 minimum adjacent channel isolation minimum value of a ij within the adjacent operating wavelength (or frequency) range (adjacent channel passband) The minimum adjacent channel isolation is positive in dB
Note 1 to entry: Refer to Figure 10 below Generally, there are two minimum adjacent channel isolations For a channel, the minimum value of two minimum adjacent channel isolations is selected
Classification
Fibre optic WDM devices shall be classified as follows:
WDM devices can be categorized into types:
A 1 x N or N x 1 Wavelength Division Multiplexing (WDM) device functions as a wavelength multiplexer, demultiplexer, or both, while an M x N WDM device serves as a wavelength router or a channel add/drop device The applications of these WDM devices are detailed through transfer matrices, as outlined in Annex C.
• By channel wavelength range or operating wavelength range;
Fibre optic WDM devices can be categorized by various styles, including the type of fibre, connector, cable, housing shape, and configuration The branching device ports are classified based on their specific configurations.
A device containing integral fibre optic pigtails, without connectors (see Figure 16)
Figure 16 – Wavelength-selective branching device
A device containing integral fibre optic pigtails, with a connector on each pigtail (see Figure 17)
Figure 17 – Wavelength-selective branching device
A device containing fibre optic connectors as an integral part of the device housing (see Figure 18)
Figure 18 – Wavelength-selective branching device
A device containing some combination of the interfacing features of the preceding configurations (see Figure 19)
Figure 19 – Wavelength-selective branching device Variant
The wavelength-selective branching device variant identifies the common features which encompass structurally similar components
Examples of features which define a variant include, but are not limited to the following:
Relevant specifications shall specify one or more assessment levels, each of which shall be designated by a capital letter The assessment level defines the relationship between groups
A and B inspection levels and groups C and D inspection periods
The following are the preferred levels
• group A inspection: inspection level II, AQL = 4 %;
• group B inspection: inspection level II, AQL = 4 %;
• group A inspection: inspection level II, AQL = 1 %;
• group B inspection: inspection level II, AQL = 1 %;
• group A inspection: inspection level II, AQL = 0,4 %;
• group B inspection: inspection level II, AQL = 0,4 %;
• group D inspection: 24 month periods where AQL is the acceptable quality level
One additional assessment level (other than those specified above) can be given in the relevant specification When this is done, the capital letter X shall be used
NOTE Groups A and B are subject to lot-by-lot inspection Groups C and D are subject to periodic inspection
Other documents may be referenced.
Documentation
Graphical and letter symbols shall, whenever possible, be taken from the IEC 60027 series, IEC 60050-731 and IEC 61931
This specification is a component of the IEC specification system, which includes subsidiary specifications made up of relevant specifications As illustrated in Table 1, there are currently no sectional specifications available for WDM devices.
Table 1 – Three-level IEC specification structure
Specification level Examples of information to be included Applicable to
Inspection rules Optical measuring methods Environmental test methods Sampling plans
Two or more component families or subfamilies
Specification level Examples of information to be included Applicable to
Identification rule Marking standards Dimensional standards Terminology standards Symbol standards Preferred number series
Specific symbols Specific units Preferred values Marking
Quality assessment procedures Selection of tests
Qualification approval and/or Capability approval procedures
Specific information Completed quality conformance test schedules
A wavelength-selective branching device is defined by a specific specification, which can be developed by any national committee of the IEC This process establishes a particular design for the device as an IEC standard, adhering to the constraints of the generic specification.
Relevant specifications shall specify the following as applicable:
The drawings and dimensions given in relevant specifications shall not restrict details of construction, nor shall they be used as manufacturing drawings
This specification mandates the use of either first angle or third angle projection for all drawings in the documents Consistency is key, as all drawings within a document must utilize the same projection system, and it is essential to clearly indicate which system is being employed.
All dimensions shall be given in accordance with ISO 129-1, ISO 286-1 and ISO 1101
The metric system shall be used in all specifications
Dimensions shall not contain more than five significant digits
When units are converted, a note shall be added in each relevant specification and the conversion between systems of units shall use a factor of 25,4 mm to 1 inch
The measurement method to be used shall be specified in the relevant specification for any dimensions which are specified within a total tolerance zone of 0,01 mm or less
Reference components for measurement purposes, if required, shall be specified in the relevant specification
Gauges, if required, shall be specified in the relevant specification
Test data sheets must be created for every test performed in accordance with the relevant specifications These data sheets are essential components of both the qualification report and the periodic inspection report.
Data sheets shall contain the following information as a minimum:
• title of test and date;
• specimen description including the type of fibre and the variant identification number;
• test equipment used and date of latest calibration;
• all measurement values and observations;
• sufficiently detailed documentation to provide traceable information for failure analysis
Instructions for use, when required, shall be given by the manufacturer.
Standardization system
Performance standards consist of a series of tests and measurements, which may be organized into a specific schedule based on the standard's requirements These tests are designed to be conducted once to demonstrate a product's ability to meet the performance standards Each standard features a unique set of tests and severities that reflect the needs of a particular market sector, user group, or system location.
A product that meets all performance standard requirements can be declared compliant; however, it must be monitored through a quality assurance and quality conformance program.
The performance standards emphasize the importance of selecting appropriate tests and severity levels from measurement standards to ensure compatibility between products, particularly concerning attenuation and return loss Compliance of each product with these standards will be rigorously verified.
Reliability standards are intended to ensure that a component can meet performance specifications under stated conditions for a stated time period
For each type of component, the following need to be identified (and appear in the reliability standard):
• failure modes (observable general mechanical or optical effects of failure);
• failure mechanisms (general causes of failure, common to several components), and failure effects (detailed causes of failure, specific to component)
These are all related to environmental and material aspects
After manufacturing, components undergo an "infant mortality phase" where many may fail if deployed immediately To prevent early failures in the field, manufacturers implement a screening process that subjects components to environmental stresses, including mechanical, thermal, and humidity factors This controlled testing accelerates known failure mechanisms, allowing them to manifest sooner than they would in unscreened components As a result, components that pass this screening process exhibit a significantly reduced failure rate, ensuring greater reliability for consumers.
Screening is an optional aspect of manufacturing, distinct from a testing method, and does not influence the "useful life" of a component, which is the duration it operates within specifications Over time, various failure mechanisms emerge, leading to an increased failure rate that surpasses a certain threshold This marks the end of the useful life and the onset of the "wear-out region," indicating that the component requires replacement.
At the start of a component's useful life, performance testing is conducted by the supplier, manufacturer, or a third party to verify that it meets performance specifications across various intended environments In contrast, reliability testing ensures that the component adheres to performance standards for a defined minimum useful lifetime or a maximum failure rate This testing typically builds upon performance testing by extending the duration and severity to expedite the failure mechanisms.
Reliability theory connects the testing of component reliability to their parameters and the associated lifetime or failure rates It extrapolates these findings to predict performance under less demanding service conditions The reliability specifications outline the necessary component parameters to guarantee a minimum lifetime or a maximum failure rate during operation.
To ensure effective product standardization, it is essential to establish interface, performance, and reliability standards Table 2 presents various options that can be utilized once these three standards are implemented.
Interface standard Performance standard Reliability standard
Product A is fully IEC standardised having a standard interface and meeting defined performance standards and reliability standards
Product B is a product with a proprietary interface but which meets a defined IEC performance standard and reliability standard
Product C is a product which complies with an IEC standard interface but does not meet the requirements of either an IEC performance standard or reliability standard
Product D is a product which complies with both an IEC standard interface and performance standard but does not meet any reliability requirements.
Design and construction
Devices must be constructed from materials that comply with the relevant specifications If non-flammable materials are necessary, this requirement should be clearly stated in the specifications, and the IEC 60695-11-5 test should be referenced.
All components and related hardware must be produced to consistent quality standards, ensuring they are free from sharp edges, burrs, or any defects that could impact their durability, functionality, or aesthetic appeal Special emphasis should be placed on the precision and quality of marking, plating, soldering, and bonding processes.
Performance requirements
Fibre optic WDM devices shall meet the performance requirements specified in appropriate IEC performance standard.
Identification and marking
Components, associated hardware, and packages shall be permanently and legibly identified and marked when this is required by the relevant specification
Each variant in a relevant specification shall be assigned a unique identification number This number shall be set out as follows:
Component marking must be detailed in the applicable specification, following this preferred order: a) port identification, b) manufacturer's part number (including serial number if necessary), c) manufacturer's identification mark or logo, d) manufacturing date, e) variant identification number, and f) any additional markings specified.
In cases where space limitations prevent the necessary markings on a component, each unit must be individually packaged with a data sheet that includes all required information that cannot be marked directly.
Several fibre optic WDM devices may be packed together for shipment
Package marking requirements should be detailed in the relevant specification, with the preferred order of marking including: a) manufacturer's identification mark or logo, b) manufacturer's part number, c) manufacturing date code (year/week as per ISO 8601), d) variant identification number(s), e) type designation, f) assessment level, and g) any additional markings mandated by the specification.
Individual unit packages must be labeled with the reference number of the certified record of released lots, the manufacturer's factory identity code, and the component identification when applicable.
Safety
Fibre optic WDM devices can emit potentially hazardous radiation from an uncapped or unterminated output port or fibre end when utilized in an optical fibre transmission system or equipment.
Manufacturers of fibre optic WDM devices must provide adequate information to inform system designers and users about potential hazards, as well as outline necessary precautions and best working practices.
In addition, each relevant specification shall include the following:
When handling small diameter fibers, it is crucial to avoid skin punctures, particularly around the eyes It is not advisable to directly view the end of an optical fiber or connector while it is transmitting energy unless safety assurances regarding the output energy level have been confirmed.
Reference shall be made to IEC 60825-1, the relevant standard on safety
General
The optical characteristics of a fiber optic wavelength-selective branching device can be represented by an n x n matrix of coefficients, where n indicates the number of ports and the coefficients denote the fractional optical power transferred among specific ports For instance, a six-port device with two input ports and four output ports, as illustrated in Figure A.1, allows for a total of 36 possible combinations of port connections, which are organized in a matrix format.
Figure A.1 – Example of a six-port device, with two input and four output ports
Transfer matrix
In general, the transfer matrix T is:
The ratio of optical power transferred from output port \( j \) to input port \( i \) is defined as \( t_{ij} = \frac{P_{ij}}{P_i} \), where \( t_{ij} \) ranges from greater than zero to one (0 ≤ \( t_{ij} \) ≤ 1) In a wavelength-selective branching device, the coefficient \( t_{ij} \) depends on the wavelength and may also vary with the input polarization or modal power distribution.
Single-mode fibre optic WDM devices can function coherently with multiple inputs, meaning that the transfer coefficients are influenced by the relative phase and intensity of coherent optical power inputs at two or more ports.
The transfer matrix coefficient is influenced by wavelength, denoted as \( t_{ijk} \), where \( k \) represents the wavelength number, \( \lambda_k \) A more general expression for the transfer matrix is provided below.
Transfer matrix coefficient
An element t ij of the transfer matrix (refer to Figure A.2 below)
Figure A.2 – Illustration of transfer matrix coefficient
Logarithmic transfer matrix
In general, the logarithmic transfer matrix is:
The optical power reduction in decibels from port \( j \) with unit power into port \( i \) is represented by the equation \( a_{ij} = -10 \log t_{ij} \), where \( t_{ij} \) is the transfer matrix coefficient and \( a_{ij} \) is a non-negative value This relationship highlights the connection between the transfer matrix coefficient and the logarithmic transfer matrix, providing a more comprehensive understanding of optical power dynamics.
Specific performances of WDM devices for bidirectional transmission system (example)
Generic
The standard configuration for Wavelength Division Multiplexing (WDM) devices is 1 x N, which can function as both multiplexers (MUXs) and demultiplexers (DEMUXs) Additionally, 1 x N WDM devices are utilized in bidirectional transmission systems For instance, Figure B.1 illustrates both unidirectional and bidirectional transmission applications of a 1 x 2 WDM device, with Figure B.1a depicting a unidirectional transmission system (DEMUX application) and Figure B.1b showcasing a bidirectional transmission system application.
Figure B.1a – Unidirectional transmission system application of a 1x2 WDM device (DEMUX) λ λ 1 λ λ 2
Figure B.1b – Bidirectional transmission system application of a 1x2 WDM device
Figure B.1 – Uni-directional and bi-directional transmission system application of a 1 x 2 DM device
In the unidirectional transmission system illustrated in Figure B.1a, port 1 serves as the input port, while ports 2 and 3 function as output ports, with wavelength λ 1 directed to port 2 and wavelength λ 2 to port 3 In this context, far-end crosstalk (XT FE) for port 2 (λ 1) is defined using a formula based on the logarithmic transfer matrix coefficient, highlighting the significance of "far-end" as referring to the opposite side in a WDM device.
For port 3 and λ 2 , the far-end crosstalk is a 132 – a 131 Far-end crosstalk is negative value expressed in dB
Far-end isolation can also be defined Far-end isolation is a 122 for port 2, a 131 for port 3 Far- end isolation has the same meaning as “commonly-used“ isolation
In the bidirectional transmission system illustrated in Figure B.1b, port 1 serves as the input for wavelength λ 1 and the output for wavelength λ 2 This setup allows for the definition of near-end isolation and near-end crosstalk, where "near-end" refers to the same side of the system Specifically, the near-end isolation for port 2 (λ 1) is measured at 322 Additionally, near-end crosstalk (XT NE) for port 2 (λ 1) is expressed using a logarithmic transfer matrix coefficient formula.
Definition of near-end isolation and near-end crosstalk
More generic definitions of near-end isolation and near-end crosstalk are explained as shown below
B.2.1 bidirectional (near-end) isolation for a bidirectional WDM multiplexer (MUX)/demultiplexer (DMUX) device, bidirectional (near- end) isolation is defined as
BCA represents a mox, which is an element of the logarithmic transfer matrix In this context, \( m \) denotes the MUX input port number, \( o \) signifies the DMUX output port number, and \( x \) indicates the wavelength number linked to port \( m \).
Bidirectional near-end crosstalk in a WDM-MUX/DEMUX device is determined by subtracting the logarithmic transfer matrix coefficient of the conducting port pair with the active channel from the bidirectional near-end isolation.
Bidirectional WDM-MUX/DMUX devices feature input and output channels on the same side, allowing input light from one direction to potentially appear on the output port of the opposite direction This phenomenon is referred to as bidirectional (near-end) crosstalk.
In the context of the logarithmic transfer matrix, the equation \$I_B = a_{mox} - a_{doc}\$ defines the relationship between various elements Here, \$a_{mox}\$ and \$a_{doc}\$ represent specific elements of the matrix, while \$d\$ denotes the DMUX input port number and \$o\$ signifies the DMUX output port number Additionally, \$c\$ corresponds to the channel wavelength number linked to port \$o\$, and \$m\$ indicates the MUX input port number, with \$x\$ representing the wavelength number associated with port \$m\$.
Note 2 to entry: In the example given below of a four-wavelength bidirectional system, wavelengths 1 and 2 travel from left to right and wavelengths 3 and 4 from right to left (see Figure B.2)
Figure B.2 – Illustration of a four-wavelength bidirectional system
For the example given above, the bidirectional isolation of port 2 to wavelength 3 is a 423 – a 121
Transfer matrix as applications of WDM devices (example)
Generic
Wavelength Division Multiplexing (WDM) devices play a crucial role in fiber optic communication systems, including applications such as wavelength multiplexers, demultiplexers, multiplexer/demultiplexers, routers, and channel add/drops These WDM functions are essential for efficient optical transmission and can be represented using transfer matrices.
The schematic diagrams which follow do not necessarily correspond to the physical layout of the branching device and its ports
The diagrams illustrate the direction of optical power travel, indicated by arrows on the ports Ports without arrows are typically isolated from the corresponding launched port.
The following devices include only those which are in common use within industry at present They do not include every possible form of transfer matrix
For the definition of the transfer matrix refer to 3.1.2
The transfer coefficients are nominally equal to zero or greater than zero The nominal values of the transfer coefficients are indicated.
Wavelength multiplexer
A wavelength-selective branching device is designed to merge N distinct optical signals, each identified by its wavelength, from N input ports into a single output port, referred to as Port 0.
Figure C.1 – Example of a wavelength multiplexer
The wavelength dependent transfer matrix is:
For non-zero indices \(i\), each coefficient \(t_{i0}\) should ideally equal 1 at wavelength \(i\) and 0 at all other wavelengths The coefficients \(t_{ij}\) (where \(i, j \neq 0\) and \(i \neq j\)) are associated with directivity, while the coefficients \(t_{ii}\) pertain to return loss.
Wavelength demultiplexer
A wavelength-selective branching device is designed to separate N distinct optical signals based on their wavelengths, directing them from a single input port to N corresponding output ports, with Port 0 serving as the input.
Figure C.2 – Example of a wavelength demultiplexer
The wavelength dependent transfer matrix is:
For i ≠ 0 each coefficient t 0i is ideally 1 at wavelength i and 0 at all other operating wavelengths t ii is related to the return loss.
Wavelength multiplexer/demultiplexer
A wavelength-selective branching device which performs functions both of a wavelength multiplexer and demultiplexer Port 0 is the output for the multiplexer and input for the demultiplexer (see Figure C.3) λ 1 λ 2
Figure C.3 – Example of a wavelength multiplexer/demultiplexer
The wavelength dependent transfer matrix is:
For non-zero indices \(i\), the ideal coefficients \(t_{0i}\) and \(t_{i0}\) should equal 1 at wavelength \(i\) and 0 at all other wavelengths The coefficients \(t_{ij}\) (where \(i, j \neq 0\) and \(i \neq j\)) are associated with directivity, while the coefficients \(t_{ij}\) are linked to return loss.
Wavelength router
A wavelength-selective branching device routes N operating wavelengths to designated output ports based on selected input ports, enabling efficient transmission through the device.
Figure C.4 – Example of a wavelength router
The wavelength dependent transfer matrix is:
In zones A and B of the matrix, the coefficients \( t_{ii} \) are associated with return loss, while the coefficients \( t_{ij} \) (where \( i \neq j \)) pertain to directivity Zones C feature nominally symmetric and identical matrices, with \( t_{ii} \) typically equal to 1 at the operating wavelength.
2] N+1 (where [M] N defines the function M mod N) and 0 at all other operating wavelengths.
Wavelength channel add/drop
A wavelength-selective branching device is designed to drop (N-1) channels from a set of M operating wavelengths (where N ranges from 2 to M + 1) while simultaneously inserting (N-1) channels at the same nominal operating wavelength as the dropped channels.
Figure C.5 – Example of wavelength channel add/drop
The wavelength dependent transfer matrix is:
In zone A of the matrix, the transfer coefficients are nominally zero, with the coefficients \( t_{ii} \) associated with return loss and \( t_{ij} \) (where \( i \neq j \)) linked to near-end crosstalk Conversely, in zone B, the coefficient \( t_{1(N+1)} \) is consistently nominally 1 across all measurements.
The operating wavelengths are defined such that M – N + 1 wavelengths, denoted as λ_i, are distinct from λ_j and λ_k, while being nominally zero at all other wavelengths The coefficients t_j (N+1) are set to 1 at the operating wavelength λ_j and 0 at all other wavelengths, mirroring the coefficients t_1j for j ≠ (N+1) All other coefficients, which pertain to directivity, are nominally zero.
Example of technology of thin film filter WDM devices
General
A Wavelength Division Multiplexing (WDM) device utilizing thin film filter (TTF) technology features a thin film filter applied to a substrate, typically a glass plate It includes optical fibers for input and output ports, along with lenses that transform divergent light into collimated light.
Figure D.1 – Schematic configuration of a thin film filter WDM device
Thin film filter technology
A thin-film filter operates on the principles of the Fabry-Perot etalon, functioning as a bandpass filter It allows signals at the passband wavelength to transmit while reflecting other wavelengths with high efficiency The cavity length of the filter is crucial, as it determines the center wavelength of the passband.
Multilayer thin-film filters, also referred to as wavelength selective optical filters, consist of alternating layers of optical coatings applied to a glass substrate By adjusting the thickness and quantity of these layers, the passband wavelength of the filter can be precisely tuned, allowing for customization of its width.
Key d k thickness; n k refractive index; θ k incident angle
Figure D.2 – Structure of multilayer thin film
Typical characteristics of thin film filter
Figure D.3 shows the typical characteristics of a 1 510 nm and C-band WDM device that uses thin film filter technology This device has three ports
Figure D.3 – Typical characteristics of 1 510 nm and C-band WDM device using thin film filter technology d 1 d 2 d 3 d k - 1 n 1 n 2 n 3 n k - 1 n k n 0 θ 1 θ 0 θ 2 θ 3 θ k θ θ θ θ θ
In se rt io n L os s ( dB )
Common port to Pass port Common port to Reflect port
Common port to Pass port Common port to Reflect port
Example of technology of fibre fused WDM devices
General
A fused coupler is a crucial passive component in optical telecommunication systems, serving essential functions such as light branching and splitting in optical fiber circuits, multiplexing (MUX) and demultiplexing (DEMUX), filtering, wavelength-independent splitting, and polarization-selective splitting.
A basic 2 x 2 bidirectional fusion coupler is created by fusing two independent single mode fibers, utilizing the fundamental principle of coupling between parallel optical waveguides This process involves fusing the claddings of each fiber over a small area, necessitating that the fibers are positioned closely together.
The core principle in waveguide theory is the transfer of power between two waveguides, which can be partial or complete, due to energy transfer This optical power exchange is facilitated by the optical coupling between the evanescent wave of the guided mode in one core and the natural mode in the second core.
The uniformly spaced parallel interaction region is crucial for the coupling process, featuring a longitudinally invariant structure The optical coupling occurring in this waist region can be effectively analyzed through coupling mode analysis.
Figure E.1 – Structure of a fused bi-conical tapered 2x2 coupler
One of the various packaging schemes of the fused couplers is shown in Figure E.2 The package generally involves a double-layered structure designed to protect the fused bi-conical region
The performance of the coupler is significantly affected by the material properties of the primary packaging substrate due to varying environmental and thermal conditions Synthetic quartz (SQ) is the optimal choice for the primary package, as it behaves similarly to the fiber The substrate is designed in a semi-cylindrical shape with a rectangular groove, allowing for easy positioning and secure fixation using a positioning stage and adhesive at the ends of a parallel region After the initial packaging, the bare fused coupler requires further encapsulation for protection The device is then placed inside a metal tube, sealed at both ends with sealants to ensure an airtight environment A metal alloy, which has a coefficient of thermal expansion comparable to that of SQ, is used for the main body.
Figure E.2 – Typical scheme for a fused coupler
Typical characteristics of fibre fused WDM devices
Figure E.3 shows typical wavelength dependent characteristics of the transmittance for bar ports and cross ports
Figure E.3 – Typical characteristics of a fibre fused WDM device
Example of arrayed waveguide grating (AWGs) technology
General
An arrayed waveguide grating is an optical dispersive element that utilizes planar lightwave circuit technology This device integrates two slab waveguides along with input and output waveguides on a single chip Functioning similarly to a spectrometer, the integrated chip serves as a multi/demultiplexer in dense wavelength division multiplexing (DWDM) transmission systems.
An Arrayed Waveguide Grating (AWG) utilizes a configuration where incoming light diffracts in a slab waveguide and enters an array of channel waveguides of varying lengths, creating a wavelength-dependent phase shift This light then converges in another slab waveguide, similar to a concave mirror, with the focusing position determined by the input light's wavelength Consequently, the wavelength multiplexed input light is effectively demultiplexed to designated output ports AWG chips are often constructed from silica glass on a silicon substrate, which minimizes propagation loss and allows for efficient coupling with single-mode optical fibers.
Figure F.1 – Basic configuration of AWG
Typical characteristics of AWG
The optical attenuation spectrum of an AWG wavelength multi/demultiplexer, designed for 100 GHz-spacing in 40-channel DWDM systems, is illustrated in Figure F.2 Each spectral curve exhibits a Gaussian profile around its peak transmission wavelength.
A flat non-Gaussian spectrum can be realized by installing a parabolic input waveguide aperture or a Mach-Zehnder interferometer in front of the input side slab waveguide
Figure F.2 – Example of AWG characteristics
Example of FBG filter technology
General
A fibre Bragg grating (FBG) selectively reflects specific light wavelengths while allowing others to pass through When paired with an optical circulator, it effectively captures these reflected wavelengths, as illustrated in Figure G.1.
Figure G.1 – Usage of fibre Bragg grating filter
A Fiber Bragg Grating (FBG) features a periodic variation in the refractive index of the fiber core, creating a wavelength-specific mirror This unique property allows FBGs to function effectively as optical filters or wavelength-specific reflectors.
Figure G.2 – Function and mechanism of fibre Bragg grating
The core concept of a Fiber Bragg Grating (FBG) is based on Bragg reflection, where the refractive index shows a periodic variation along a specific length The wavelength that is reflected, known as the Bragg wavelength (\( \lambda_B \)), is determined by the relationship involving the grating period (\( \Lambda \)).
B =2n λ where n is the average refractive index of the grating and Λ is the period of the refractive index variation
The bandwidth (Δλ), is given by
∆ n where δn 0 is the variation in the refractive index, and η is the power fraction in the core
The peak reflection (P B (λ B )) is approximately given by
Typical characteristics of FBG filter
In se rt io n L os s ( dB )
R ef le ct io n ( dB )
O pt ic al at tenuat ion (dB )
Figure G.3 – Example of FBG filter characteristics
ITU-T Recommendation G.694.1, Spectral grids for WDM applications: DWDM frequency grid
ITU-T Recommendation G.694.2, Spectral grids for WDM applications: CWDM wavelength grid