scottk/oopt-gnpy
1
0
Fork 0
mirror of https://github.com/Telecominfraproject/oopt-gnpy.git synced 2025-10-29 09:12:37 +00:00
Code Issues Packages Projects Releases 10 Wiki Activity
Files
master
oopt-gnpy/docs/amplifier_models_description.rst
EstherLerouzic d5491c9ace fix documentation: harmonize titles 
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I827b4dcd1418017d925b63e50f95514dc1a0eed8
2025-03-21 10:00:25 +01:00

319 lines
11 KiB
ReStructuredText
Raw Permalink Blame History

.. _amp_models:
**********************************
Amplifier models and Configuration
**********************************
1. Equipment configuration description
======================================
Equipment description defines equipment types and parameters.
It takes place in the equipment library such as **eqpt_config.json** file defined in example-data folder.
By default **gnpy-transmission-example** uses **eqpt_config.json** file and that
can be changed with **-e** or **--equipment** command line parameter.
2. Amplifier parameters and subtypes
====================================
Several amplifiers can be used by GNpy, so they are defined as an array of equipment parameters in **eqpt_config.json** file.
- *"type_variety"*:
Each amplifier is identified by its unique *"type_variety"*, which is used in the topology files input to reference a specific amplifier. It is a user free defined id.
For each amplifier *type_variety*, specific parameters are describing its attributes and performance:
- *"type_def"*:
Sets the amplifier model that the simulation will use to calculate the ase noise contribution. 5 models are defined with reserved words:
- *"advanced_model"*
- *"variable_gain"*
- *"fixed_gain"*
- *"dual_stage"*
- *"multi_band"*
- *"openroadm"*
*see next section for a full description of these models*
- *"default_config_from_json"*:
Use a custom per frequency dynamic gain tilt, gain and noise ripple arrays defined in the file specified with
this option, instead of the default values from GNPy.
- *"advanced_config_from_json"*:
**This parameter is only applicable to the _"advanced_model"_ model**
json file name describing:
- nf_fit_coeff
- f_min/max
- gain_ripple
- nf_ripple
- dgt
*see next section for a full description*
- *"gain_flatmax"*:
amplifier maximum gain in dB before its extended gain range: flat or nominal tilt output.
If gain > gain_flatmax, the amplifier will tilt, based on its dgt function
If gain > gain_flatmax + target_extended_gain, the amplifier output power is reduced to not exceed the extended gain range.
- *"gain_min"*:
amplifier minimum gain in dB.
If gain < gain_min, the amplifier input is automatically padded, which results in
NF += gain_min - gain
- *"p_max"*:
amplifier max output power, full load
Total signal output power will not be allowed beyond this value
- *"nf_min/max"*:
**These parameters are only applicable to the _"variable_gain"_ model**
min & max NF values in dB
NF_min is the amplifier NF @ gain_max
NF_max is the amplifier NF @ gain_min
- *"nf_coef"*:
**This parameter is only applicable to the *"openroadm"* model**
[a, b, c, d] 3rd order polynomial coefficients list to define the incremental OSNR vs Pin
Incremental OSNR is the amplifier OSNR contribution
Pin is the amplifier channel input power defined in a 50GHz bandwidth
Incremental OSNR = a*Pin³ + b*Pin² + c*Pin + d
- *"preamp_variety"*:
**This parameter is only applicable to the _"dual_stage"_ model**
1st stage type_variety
- *"booster_variety"*:
**This parameter is only applicable to the *"dual_stage"* model**
2nd stage type_variety
- *"out_voa_auto"*: true/false
**power_mode only**
**This parameter is only applicable to the *"advanced_model"* and *"variable_gain"* models**
If "out_voa_auto": true, auto_design will chose the output_VOA value that maximizes the amplifier gain within its power capability and therefore minimizes its NF.
- *"allowed_for_design"*: true/false
**auto_design only**
Tells auto_design if this amplifier can be picked for the design (deactivates unwanted amplifiers)
It does not prevent the use of an amplifier if it is placed in the topology input.
.. code-block:: json
{"Edfa": [{
"type_variety": "std_medium_gain",
"type_def": "variable_gain",
"gain_flatmax": 26,
"gain_min": 15,
"p_max": 23,
"nf_min": 6,
"nf_max": 10,
"out_voa_auto": false,
"allowed_for_design": true
},
{
"type_variety": "std_low_gain",
"type_def": "variable_gain",
"gain_flatmax": 16,
"gain_min": 8,
"p_max": 23,
"nf_min": 6.5,
"nf_max": 11,
"out_voa_auto": false,
"allowed_for_design": true
}
]}
3. Amplifier models
===================
In an opensource and multi-vendor environnement, it is needed to support different use cases and context. Therefore several models are supported for amplifiers.
5 types of EDFA definition are possible and referenced by the *"type_def"* parameter with the following reserved words:
- *"advanced_model"*
This model is refered as a whitebox model because of the detailed level of knowledge that is required. The amplifier NF model and ripple definition are described by a json file referenced with *"advanced_config_from_json"*: json filename. This json file contains:
- nf_fit_coeff: [a,b,c,d]
3rd order polynomial NF = f(-dg) coeficients list
dg = gain - gain_max
- f_min/max: amplifier frequency range in Hz
- gain_ripple : [...]
amplifier gain ripple excursion comb list in dB across the frequency range.
- nf_ripple : [...]
amplifier nf ripple excursion comb list in dB across the frequency range.
- dgt : [...]
amplifier dynamic gain tilt comb list across the frequency range.
*See next section for the generation of this json file*
.. code-block:: json-object
"Edfa":[{
"type_variety": "high_detail_model_example",
"type_def": "advanced_model",
"gain_flatmax": 25,
"gain_min": 15,
"p_max": 21,
"advanced_config_from_json": "std_medium_gain_advanced_config.json",
"out_voa_auto": false,
"allowed_for_design": false
}
]
- *"variable_gain"*
This model is refered as an operator model because a lower level of knowledge is required. A full polynomial description of the NF cross the gain range is not required. Instead, NF_min and NF_max values are required and used by the code to model a dual stage amplifier with an internal mid stage VOA. NF_min and NF_max values are typically available from equipment suppliers data-sheet.
There is a default configuration to enforce 0 tilt and ripple values because GNPy core algorithm is a multi-carrier propagation.
- gain_ripple =[0,...,0]
- nf_ripple = [0,...,0]
- dgt = [...] generic dgt comb
.. code-block:: json-object
"Edfa":[{
"type_variety": "std_medium_gain",
"type_def": "variable_gain",
"gain_flatmax": 26,
"gain_min": 15,
"p_max": 23,
"nf_min": 6,
"nf_max": 10,
"out_voa_auto": false,
"allowed_for_design": true
}
]
- *"fixed_gain"*
This model is also an operator model with a single NF value that emulates basic single coil amplifiers without internal VOA.
if gain_min < gain < gain_max, NF == nf0
if gain < gain_min, the amplifier input is automatically padded, which results in
NF += gain_min - gain
.. code-block:: json-object
"Edfa":[{
"type_variety": "std_fixed_gain",
"type_def": "fixed_gain",
"gain_flatmax": 21,
"gain_min": 20,
"p_max": 21,
"nf0": 5.5,
"allowed_for_design": false
}
]
- *"openroadm"*
This model is a black box model replicating OpenRoadm MSA spec for ILA.
.. code-block:: json-object
"Edfa":[{
"type_variety": "openroadm_ila_low_noise",
"type_def": "openroadm",
"gain_flatmax": 27,
"gain_min": 12,
"p_max": 22,
"nf_coef": [-8.104e-4,-6.221e-2,-5.889e-1,37.62],
"allowed_for_design": false
}
]
- *"dual_stage"*
This model allows the cascade (pre-defined combination) of any 2 amplifiers already described in the eqpt_config.json library.
- preamp_variety defines the 1st stge type variety
- booster variety defines the 2nd stage type variety
Both preamp and booster variety must exist in the eqpt libray
The resulting NF is the sum of the 2 amplifiers
The preamp is operated to its maximum gain
- gain_min indicates to auto_design when this dual_stage should be used
But unlike other models the 1st stage input will not be padded: it is always operated to its maximum gain and min NF. Therefore if gain adaptation and padding is needed it will be performed by the 2nd stage.
.. code-block:: json
{
"type_variety": "medium+low_gain",
"type_def": "dual_stage",
"gain_min": 25,
"preamp_variety": "std_medium_gain",
"booster_variety": "std_low_gain",
"allowed_for_design": true
}
- *"multiband"*
This model enables the definition of multiband amplifiers that consist of multiple single-band
amplifier elements, with each amplifier responsible for amplifying a different portion of the spectrum.
The types of single-band amplifiers that can be included in these multiband amplifiers are specified,
allowing for multiple options to be available for the same spectrum band (for instance, providing
several permitted type varieties for both the C-band and the L-band). The actual element utilizing the
type_variety must implement only one option for each band.
4. advanced_config_from_json
============================
The build_oa_json.py library in ``gnpy/example-data/edfa_model/`` can be used to build the json file required for the amplifier advanced_model type_def:
Update an existing json file with all the 96ch txt files for a given amplifier type
amplifier type 'OA_type1' is hard coded but can be modified and other types added
returns an updated amplifier json file: output_json_file_name = 'edfa_config.json'
amplifier file names
Convert a set of amplifier files + input json definiton file into a valid edfa_json_file:
nf_fit_coeff: NF 3rd order polynomial coefficients txt file
nf = f(dg) with dg = gain_operational - gain_max
nf_ripple: NF ripple excursion txt file
gain_ripple: gain ripple txt file
dgt: dynamic gain txt file
input json file in argument (defult = 'OA.json')
the json input file should have the following fields:
.. code-block:: json
{
"nf_fit_coeff": "nf_filename.txt",
"nf_ripple": "nf_ripple_filename.txt",
"gain_ripple": "DFG_filename.txt",
"dgt": "DGT_filename.txt"
}
Reference in New Issue View Git Blame Copy Permalink