The code look as if it was trying to prevent direct instantiation of the
SimParams class. However, instance *creation* in Python is actually
handled via `__new__` which was not overridden. In addition, the
`get()` accessor was invoking `SimParams.__new__()` directly, which
meant that this class was instantiated each time it was needed.
Let's cut the boilerplate by getting rid of the extra step and just use
the regular constructor.
This patch doesn't change anything in actual observable behavior. I
still do not like this implicit singleton design pattern, but nuking
that will have to wait until some other time.
Change-Id: I3ca81bcd0042e91b4f6b7581879922611f18febe
Constant power per slot_width uses the slot width instead of
baud rate compared to PSD.
This is the equalization used in OpenROADM
add tests for constant power per slot width equalization
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: Ie350e4c15cb6b54c15e418556fe33e72486cb134
flake8-html is now compatible with the v5 of this package, so let's use
it. Unfortunately, they killed the `--diff` option in v6, so we cannot
use it right now. I understand the reasoning as well as the fact that
it's easy to be broken, but I don't like broken CI that much.
Change-Id: I70dd686e097f411c39bfc53f83d519540687dd64
...mainly to be in sync with oopt-gnpy-libyang that I've been working on
recently, and to allow us to modernize this infrastructure later on.
Change-Id: Id0ed1d7620762fc204300ebe8a190de8e42ae9df
GitHub CI started failing with the following error:
assert (watt2dbm(si.signal) == target - correction).all()
assert False
+ where False = <built-in method all of numpy.ndarray object at 0x7f01c0ca94d0>()
+ where <built-in method all of numpy.ndarray object at 0x7f01c0ca94d0> = array([-25.5, -24.5, -22.5, -25. , -27.5]) == array([-25.5, -24.5, -22.5, -25. , -27.5])
+array([-25.5, -24.5, -22.5, -25. , -27.5])
-array([-25.5, -24.5, -22.5, -25. , -27.5])
Full diff:
array([-25.5, -24.5, -22.5, -25. , -27.5]).all
This is with code which has passed in the Zuul/Vexxhost CI.
It looks very similar to a regression that hit numpy 1.24.0, but the
GitHub action log shows that this happens with numpy 1.24.1. Weird, and
I'm not getting these differences locally, and also not on an ARM64
cloud VM.
Anyway, comparing floating point numbers for strict equality is futile,
so let's use this opportunity to use a proper check for these.
Change-Id: I05683f3116cad78d067bddde2780fe25b5caf768
On a ROADM, the code would previously set the same per-carrier power
regardless of the channel spectrum width. With this patch, carriers are
equalized either by their:
- absolute power (same as before),
- power spectral density (PSD).
Also, it's possible to apply a per-channel power offset (in dB) which
will be applied to a specified channel on top of the selected
power-level or PSD strategy. The same offset can be also selected
through the `--spectrum` option via the `default_pdb` parameter.
The equalization policy can be set via the ROADM model (in the equipment
config) as well as on a per-instance basis.
The PSD is defined as the absolute power over a spectral bandwidth,
where the spectral bandwidth corresponds to the actual spectrum
occupation (without any applicable guard bands), as approximated by the
symbol rate. PSD is specified in mW/GHz. As an example, for a 32 GBaud
signal at 0.01 mW, the PSD is 0.01/32 = 3.125e-4 mW/GHz.
This has some implications on the power sweep and ROADM behavior. Same
as previously (with absolute power targets), the ROADM design determines
the power set points. Target power is usually the best (highest) power
that can be supported by the ROADMs, especially the Add/Drop and express
stages' losses, with the goal to maximize the power at the booster's
input. As such, the `--power` option (or the power sweep) doesn't
manipulate with ROADM's target output power, but only with the output
power of the amplifiers. With PSD equalization, the `--power` option is
interpreted as the power of the reference channel defined in equipment
config's `SI` container, and its PSD is used for propagation. Power
sweep is interpreted in the same way, e.g.:
"SI":[{
"f_min": 191.3e12,
"baud_rate": 32e9,
"f_max":195.1e12,
"spacing": 50e9,
"power_dbm": 0,
"power_range_db": [-1,1,1],
"roll_off": 0.15,
"tx_osnr": 40,
"sys_margins": 2
}],
...and with the PSD equalization in a ROADM:
{
"uid": "roadm A",
"type": "Roadm",
"params": {
"target_psd_out_mWperGHz": 3.125e-4,
}
},
{
"uid": "edfa in roadm A to toto",
"type": "Edfa",
"type_variety": "standard_medium_gain",
"operational": {
"gain_target": 22,
"delta_p": 2,
"tilt_target": 0.0,
"out_voa": 0
}
},
then we use the power steps of the power_range_db to compute resulting
powers of each carrier out of the booster amp:
power_db = psd2powerdbm(target_psd_out_mWperGHz, baud_rate)
sweep = power_db + delta_power for delta_power in power_range_db
Assuming one 32Gbaud and one 64Gbaud carriers:
32 Gbaud 64 Gbaud
roadmA out power
(sig+ase+nli) -20dBm -17dBm
EDFA out power
range[
-1 1dBm 4dBm
0 2dBm 5dBm
1 3dBm 6dBm
]
Design case:
Design is performed based on the reference channel set defined in SI
in equipment config (independantly of equalization process):
"SI":[{
"f_min": 191.3e12,
"baud_rate": 32e9,
"f_max":195.1e12,
"spacing": 50e9,
"power_dbm": -1,
"power_range_db": [0,0,1],
"roll_off": 0.15,
"tx_osnr": 40,
"sys_margins": 2
}],
`delta_p` values of amps refer to this reference channel, but are applicable
for any baudrate during propagation, e.g.:
{
"uid": "roadm A",
"type": "Roadm",
"params": {
"target_psd_out_mWperGHz": 2.717e-4,
}
},
{
"uid": "edfa in roadm A to toto",
"type": "Edfa",
"type_variety": "standard_medium_gain",
"operational": {
"gain_target": 22,
"delta_p": 2,
"tilt_target": 0.0,
"out_voa": 0
}
},
Then the output power for a 64 Gbaud carrier will be +4 =
= lin2db(db2lin(power_dbm + delta_p)/32e9 * 64e9)
= lin2db(db2lin(power_dbm + delta_p) * 2)
= powerdbm + delta + 3 = 4 dBm
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I28bcfeb72b0e74380b087762bb92ba5d39219eb3
The program currently ignores the explicit `index` and reads the
constraints in the JSON order of the list. However in general, it is not
guaranteed that constraints are listed in order.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: Icefe271f5801cf9f7b43311c6666556564587c65
Signed-off-by: Jan Kundrát <jan.kundrat@telecominfraproject.com>
The option is only set for gnpy-transmission-main.
The spectrum file is a list of spectrum objects, each defining
f_min, f_max and spectrum attributes using the same meaning as SI
in eqpt_config.json for baud_rate, roll_off, tx_osnr. slot_width is
used for the occupation of each carrier around their central frequency,
so slot_width corresponds to spacing of SI.
Unlike SI, the frequencies are defined includint f_min and f_max.
The partitions must be contiguous not overlapping.
Pref.p_span0 object records the req_power, while
ref_carrier records info that will be useful for equalization ie baud_rate.
For now, I have not integrated the possibility to directly use
transceivers type and mode in the list.
User can define sets of contiguous channels and a label to identify
the spectrum bands. If no label are defined, the program justs uses
the index + baud rate of the spectrum bands as label.
Print results per spectrum label
If propagated spectrum has mixed rates, then prints results (GSNR and OSNR)
for each propagated spectrum type according to its label.
Print per label channel power of elements
Per channel power prints were previously only showing the noiseless
reference channel power and only an average power.
With this change, we add a new information on the print:
the average total power (signal + noise + non-linear noise).
If there are several spectrum types propagating, the average per
spectrum is displayed using the label.
For this purpose, label and total power are recorded in each element
upon propagation
Note that the difference between this total power and the existing
channel power represents the added noise for the considered OMS.
Indeed ROADMs equalize per channel total power, so that power displayed
in 'actual pch (dBm)' may contain some noise contribution accumulated
with previous propagation.
Because 'reference pch out (dBm)' is for the noiseless reference,
it is exactly set to the target power and 'actual pch (dBm)' is always
matching 'reference pch out (dBm)' in ROADM prints.
Add examples and tests for -spectrum option
initial_spectrum1.json reproduces exactly the case of SI
initial_spectrum2.json sets half of the spectrum with 50GHz 32Gbauds and
half with 75GHz 64 Gbauds. Power setting is not set for the second half,
So that equalization will depend on ROADM settings.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: Ibc01e59e461e5e933e95d23dacbc5289e275ccf7
Existing tests only cover short distances, and effect on accumulated
noise, especially when crossing ROADMs with equalization, are not well
reported on elements power prints.
With this long path, I can catch more printing inconsistencies.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I2d0e8ccbbd387a2cd6c645c07f4b5f75e4617c30
ref_carrier is added in Pref conveys the reference channel type
information ie the channel that was used for design (would it be
auto-design or for a given design). Other attributes (like
slot_width or roll-off) may be added here for future equalization
types.
Pref object already records the req_power, so let's remove it
from ReferenceCarrier and only use ref_carrier to record info that
will be useful for PSD equalization ie baud_rate.
This reference baud_rate is required to compute reference target power
based on spectral density values during propagation. It is thus required
because of on-the-fly evaluation of loss for p_span_i and for printing
loss and target power of ROADM during propagation.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: Ic7441afa12ca5273ff99dea0268e439276107257
This change enables to use a different tx_osnr per carrier.
If tx_osnr is defined via spectrum then use it to define a tx_osnr per
carrier in si else use the tx_osnr of request to set tx_osnr of si.
Then, the propagate function for requests is changed to update OSNR with
tx_OSNR per carrier defined in si.
TODO: The tx_osnr defined in spectrum is not yet taken into account for
the propagate_and_optimize function, because the loop that optimizes
the choice for the mode only loops on baudrate.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I0fcdf559d4f1f8f0047faa257076084ec7adcc77
The idea behind this change is to reproduce the exact same behaviour as
with the scalar, but accounting for variable levels of powers.
- delete the neq_ch: equivalent channel count in dB because with mixed
rates and power such a value has limited utility
- instead creates a vector that records the 'user defined' distribution
of power.
This vector is used as a reference for channel equalization out of the
ROADM. If target_power_per_channel has some channels power above
input power, then the whole target is reduced.
For example, if user specifies delta_pdb_per_channel:
freq1: 1dB, freq2: 3dB, freq3: -3dB, and target is -20dBm out of the
ROADM, then the target power for each channel uses the specified
delta_pdb_per_channel.
target_power_per_channel[f1, f2, f3] = -19, -17, -23
However if input_signal = -23, -16, -26, then the target can not be
applied, because -23 < -19dBm and -26 < -23dBm, and a reduction must be
applied (ROADM can not amplify).
Then the target is only applied to signals whose power is above the
threshold. others are left unchanged and unequalized.
the new target is [-23, -17, -26]
and the attenuation to apply is [-23, -16, -26] - [-23, -17, -26] = [0, 1, 0]
Important note:
This changes the previous behaviour that equalized all identical channels
based on the one that had the min power !!
TODO: in coming refactor where transmission and design will be properly
separated, the initial behaviour may be set again as a design choice.
This change corresponds to a discussion held during coders call. Please look at this document for
a reference: https://telecominfraproject.atlassian.net/wiki/spaces/OOPT/pages/669679645/PSE+Meeting+Minutes
- in amplifier: the saturation is computed based on this vector
delta_pdb_per_channel, instead of the nb of channels.
The target of the future refactor will be to use the effective
carrier's power. I prefer to have this first step, because this is
how it is implemented today (ie based on the noiseless reference),
and I would like first to add more behaviour tests before doing
this refactor (would it be needed).
- in spectralInfo class, change pref to a Pref object to enable both
p_span0 and p_spani to be conveyed during propagation of
spectral_information in elements. No refactor of them at this point.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I591027cdd08e89098330c7d77d6f50212f4d4724
Flexgrid precision is 6.25GHz so --show-channels should be at least 5 digits
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I7de4254ab18508320133371e0d8cc8b5e08f0d2f
Test runs (`linters-diff-ci`) end up with an error:
Traceback (most recent call last):
File "/home/zuul/src/gerrithub.io/Telecominfraproject/oopt-gnpy/.tox/linters-diff-ci/bin/flake8", line 8, in <module>
sys.exit(main())
File "/home/zuul/src/gerrithub.io/Telecominfraproject/oopt-gnpy/.tox/linters-diff-ci/lib/python3.10/site-packages/flake8/main/cli.py", line 22, in main
app.run(argv)
File "/home/zuul/src/gerrithub.io/Telecominfraproject/oopt-gnpy/.tox/linters-diff-ci/lib/python3.10/site-packages/flake8/main/application.py", line 336, in run
self._run(argv)
File "/home/zuul/src/gerrithub.io/Telecominfraproject/oopt-gnpy/.tox/linters-diff-ci/lib/python3.10/site-packages/flake8/main/application.py", line 326, in _run
self.report()
File "/home/zuul/src/gerrithub.io/Telecominfraproject/oopt-gnpy/.tox/linters-diff-ci/lib/python3.10/site-packages/flake8/main/application.py", line 321, in report
self.formatter.stop()
File "/home/zuul/src/gerrithub.io/Telecominfraproject/oopt-gnpy/.tox/linters-diff-ci/lib/python3.10/site-packages/flake8_html/plugin.py", line 245, in stop
self.write_index()
File "/home/zuul/src/gerrithub.io/Telecominfraproject/oopt-gnpy/.tox/linters-diff-ci/lib/python3.10/site-packages/flake8_html/plugin.py", line 281, in write_index
versions=self.option_manager.generate_versions(),
AttributeError: 'OptionManager' object has no attribute 'generate_versions'
ERROR: InvocationError for command /home/zuul/src/gerrithub.io/Telecominfraproject/oopt-gnpy/.tox/linters-diff-ci/bin/flake8 --format html --htmldir linters --exit-zero (exited with code 1)
Bug: https://github.com/lordmauve/flake8-html/issues/30
Change-Id: I755877341dec2d9cd9bdcdab098e2067f783cc27
This was corrected for example-data but not for tests data
(in commit 3a72ce84d0, related
to issue #390)
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I929beeb034166d30aa994439a1d6a26350f5c3e9
Previous check was made on reference channel computation.
Now we use the actual total input power to compute the actual gain.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I3e0db72fdb030a49e2b06cdcfb442b5e642c1777
Mainly changes self.pch_out_db to self.ref_pch_out_dbm in order
to reflect real unit for the value and to remind that this value
is defined for a reference noiseless channel (whose power is recorded
in p_spani in Pref).
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: If0e008c3efc36ce73c9df01c76cf46985543d9fa
In the previous version, when the values of the counter-propagating Raman pump profiles were flipped, the pumps resulted flipped also in frequency.
Change-Id: I66f7c2aff35c72f5dcb4fb11f7a82fe1df2ee3f2
Co-authored-by: Andrea D'Amico <andrea.damico@polito.it>
Burying these behind the tox.ini works fine on CI, but it means that
someone has to ask the users to run an extra `pip install pytest` for
the test suite. Not nice.
This will need a follow-up commit to adjust the GitHub action to use
this new simplified way. That cannot land via Gerrit due to GitHub's
permission model.
Change-Id: I8d2e610c91da728d72c7d19590b25bbd8713f0de
The lumped losses are used in the computation of the loss/gain profile
through the fiber whether the Raman effect is considered or not. The
computed power profile is used to calculate the related NLI impairment.
Using the 'gn_model_analytic' method, the lumped losses are taken into
account as the contribution of an additional total loss at the end of
the fiber span. In case the 'ggn_spectrally_separated' is selected, the
method uses the computed power profile according to the specified z and
frequency arrays. The lumped losses are so considered within the NLI
power evolution along the fiber.
Change-Id: I73a6baa321aca4d041cafa180f47afed824ce267
Signed-off-by: Jan Kundrát <jan.kundrat@telecominfraproject.com>
Both PMD and PDL is set to 0 by default. Values from the OpenROADM MSA
for ILAs are included in corresponding eqpt files.
Signed-off-by: Jonas Mårtensson <jonas.martensson@ri.se>
Change-Id: I79611db3ae798e9dadc47ee39161dc1e242f2595
This fixes#421
As a first step PDL is specified in the eqpt library for ROADMs only.
In a later step, PDL (as well as PMD) should be specified also for amps
and possibly for fibers. PDL values from the OpenROADM MSA for ROADMs
are included in the corresponding eqpt files.
The acculumation rule for PDL is the same as for PMD as shown in:
"The statistics of polarization-dependent loss in optical communication
systems", A. Mecozzi and M. Shtaif, IEEE Photon. Technol. Lett., vol.
14, pp. 313-315, Mar 2002.
PDL penalty is specified and calculated in the same way as for CD and
PMD, i.e. linear interpolation between impairment_value/penalty_value
pairs. This patch includes penalty specification for OpenROADM trx
modes according to the MSA.
Signed-off-by: Jonas Mårtensson <jonas.martensson@ri.se>
Change-Id: Ib0ab383bcaee7d7523ffc3fa9a949d76c8c86ff7
The penalties are calculated and presented separately from the GSNR.
They are also taken into account when optimizing trx mode and verifying
path feasibility in path_requests_run processing.
Penalties are specified in the eqpt_config file as part of trx modes.
This patch includes specifications for OpenROADM trx modes.
Penalties are defined by a list of
impairment_value/penalty_value pairs, for example:
"penalties": [
{
"chromatic_dispersion": 4e3,
"penalty_value": 0
},
{
"chromatic_dispersion": 18e3,
"penalty_value": 0.5
},
{
"pmd": 10,
"penalty_value": 0
},
{
"pmd": 30,
"penalty_value": 0.5
}
]
- Between given pairs, penalty is linearly interpolated.
- Below min and above max up_to_boundary, transmission is considered
not feasible.
This is in line with how penalties are specified in OpenROADM and
compatible with specifications from most other organizations and
vendors.
The implementation makes it easy to add other penalties (PDL, etc.) in
the future.
The input format is flexible such that it can easily be extended to
accept combined penalty entries (e.g. CD and PMD) in the future.
Signed-off-by: Jonas Mårtensson <jonas.martensson@ri.se>
Change-Id: I3745eba48ca60c0e4c904839a99b59104eae9216
In this change, the RamanSolver is completely restructured in order to obtain a simplified and faster solution of the Raman equation. Additionally, the inter-channel Raman effect can be evaluated also in the standard fiber, when no Raman pumping is present. The same is true for the GGN model.
The Raman pump parameter pumps_loss_coef has been removed as it was not used. The loss coefficient value evaluated at the pump frequency can be included within the fiber loss_coef parameter.
This change induces variations in some expected test results as the Raman profile solution is calculated by a completely distinct algorithm. Nevertheless, these variations are negligible being lower than 0.1dB.
Change-Id: Iaa40fbb23c555571497e1ff3bf19dbcbfcadf96b
Gamma and the raman efficiency are calculated using the effective area if not provided. Both these parameters are managed as optional in json_io.py for backward compatibility.
Change-Id: Id7f1403ae33aeeff7ec464e4c7f9c1dcfa946827
Modification of the Fiber and the NliSolver in order to properly propagate the new definition of the spectral information taking advantage of the numpy array structures.
In the previous version, the propagation of the spectral information was implemented by means of for cycles over each channel, in turn.
In this change the propagation is applied directly on the newly defined spectral information attributes as numpy arrays.
Additional changes:
- Simplification of the FiberParameters and the NliParameters;
- Previous issues regarding the loss_coef definition along the frequency are solved;
- New test in test_science_utils.py verifing that the fiber propagation provides the correct values in case of a few cases of flex grid spectra.
Change-Id: Id71f36effba35fc3ed4bbf2481a3cf6566ccb51c
Squeeze function has been replaced by asarray. Using 'get' function
instead of if condition for the dictionaries. Frequency reference
derived from wavelength reference of 1550 nm.
Change-Id: I815ad8591c9e238f3fc9322ca0946ea469ff448f
This change siplifies the structure of the simulation parameters,
removing the gnpy.science_utils.simulation layer, provides some
documentation of the parameters and define a mock fixture for testing in
safe mode.
Jan: while I'm not thrilled by this concept of hidden global state, we
agreed to let it in as a temporary measure (so as not to hold merging of
Andrea's flexgrid/multirate patches). I've refactored this to a more
pytest-ish way of dealing with fixtures. In the end, it was also
possible to remove the MockSimParams class because it was not adding any
features on top of what SimParams can do already (and to what was
tested).
Change-Id: If5ef341e0585586127d5dae3f39dca2c232236f1
Signed-off-by: Jan Kundrát <jan.kundrat@telecominfraproject.com>
On certain values of loss_coef, the computation loss_coef * 1e-3 * 1e3
results in x.0000000000000000y values:
eg if 0.21 is provided in the topology file, then parameters.FiberParams
changes this to _loss_coef = 0.00021
and elements.Fiber.to_json prints 0.21000000000000002
This is a normal floating point behaviour, but is rather confusing.
In order to remove this unwanted decimal, I propose to round the printings
in to_json
https://docs.python.org/3/tutorial/floatingpoint.html
The change also add this round for gain, because in power mode
the gain is computed based on loss and loss may have this floating value.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: Ib3287a794e7da985eabf0af914c0e1ef4914e857
Oops. We are not using gating, which means that changes are tested
against the "current tip of the branch" and might pass fine there, but
once they are merged, there can well be a conflict between them. This
has just happened.
The EDFA which reported a difference had its VOA set to 0.5. Previously,
this was not taken into assumption.
Fixes: ce51a4d1 Take explicitly set out_voa value into account in power calculation
Fixes: 280443f1 add an invocation test with power saturation
Change-Id: Icebbb16d2ef5886d2c9c04cc9a300a6aa08bf245
* changes:
tests: add OpenROADMv5 example propagation
OpenROADM: mark example config files as v4 explicitly
Add an eqpt config file matching latest OpenROADM MSA version
Add updated openroadm amp specifications to eqpt config
add several power reference setting tests to the existing test
the test only checks the power level out of roadm B.
if previous node is a preamp, power level is the one specified in
target_pch_out_db.
if previous node is a fused , power level at roadm input is below
target_pch_out_db and roadm can not increase this power (no amp).
then expected outpower is in_power, which should be equal to -22 + power_dBm
on this particular node.
nota bene
currently, no minimum losss is coded on roadm, so that the applied loss
is 0 dB, and roadm does not affect power in this case.
This behaviour is not correct and should be changed in the future.
But for now, I am only concentrating on existing behaviour tests.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I34fe2dcf2d355b291c27745ab511d3d77057dd94
The recent commit has added support for OpenROADM v5, the latest
published optical spec sheet. Given that the upstream project has
released v10 YANG files (but still just v3, v4 and v5 XLS sheets with
optical performance numbers), I think it would be rather misleading to
have both versioned and non-versioned config files -- especially when
the unversioned one refers to the oldest release, not the newest one.
Change-Id: I04109341724b51d276660d400c923dc28561aef2
- if specified, they must be used except:
- if N and M are not consistant (eg M smaller than the required
spectrum for the demand)
- if N value and M value lead to occupation outside of the band
- if spectrum is occupied
- if any of them is None, the program uses the first fit strategy for M and
the path_bandwidth value to compute minimum required M
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I9160ffb116dd9d7d53ad80638826b609a1367446
