Correctly uses the oms band and spacing for computing the nb of channel
and total power for design per band.
In order to keep the SI values as reference, introduce a new parameter
in SI to indicate wether to use this feature or not.
If "use_si_channel_count_for_design": true, then the f_min, f_max and spacing
from SI are used for all OMSes
else, the f_min, f_max, spacing defined per OMS (design_bands) is used.
This impacts tests where the artificial C-band boudaries were hardcoded, and
it also has an impact on performances when SI's defined nb of channels is larger
than the one defined per OMS. In this case the design was considering a larger
total power than the one finally propagated which resulted in reduced performance.
This feature now corrects this case (if "use_si_channel_count_for_design": false
which is the default setting). Overall autodesign are thus improved.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I471a2c45200894ca354c90b46b662f42414b48ad
tous les test marche et les jeu de tests aussi.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: If25b47aa10f97301fde7f17daa2a9478aed46db2
This will be used to compute the max total power for design per OMS.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I392f06c792af9f32d4a14324c683bd3fae655de8
Introduce a new multi-band element that contains a list of Edfa element:
- reads multiple amps out of the element config.
- deduces frequency band from the amp in the list.
no autodesign yet: multi-band amps must have type_variety.
- checks that type variety of individual EDFAs is consistent with multiband
type variety
- demux and mux spectrum when propagate in multiband
- don't add a preamp or booster if a multiband amp is already defined.
The print of channel number is removed from equipment, since the channel number
may now depend on the path's amplifiers. This changes invocation results layout.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I44e77ff82e622cdee4021a7984d660317cb90cf9
The commit introduces mux/demux functions in amps and ensures that the
propagation is only done on carriers that are in the Amp bandwitdh, ie
with all their spectrum including slot width is in bandwidth.
For consistency, default amp f_min is changed:
Objective is to use amplifiers' band to bound the possible frequencies
to be propagated. Since the current default f_min of Amp in json_io.py is
higher than the SI one, this would result in a different nb of channels
than currently used in tests, and a change in all tests. In order to
avoid this, I preferred to change this value and have consistency
between SI f_min and Amp f_min.
The commits adds a set of functions to make amps band the useable
spectrum on each OMS. Thee OMS generation is changed to use the amp band.
The commit adds filtering functions (demux and mux) to filter out spectrum
which is not in the amplifier band.
Spectrum assignment is also corrected to correctly match the amp bandwidth
constraint with guardband: center frequency index must be within the
usable part of the amp band. This changes a bit the notion of freq_index
and guardband in the functions, but this is transparent to user:
f_min, f_max represent the amp band, while self.freq_index_min/max
represent the center frequency boundary for a reference 50GHz channel.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I225b2b2dc0e1f1992c0460f6e08fa9c9bc641edf
redesign True means that network is redesigned using each request
as reference channel. When False it means that the design is made
once and successive propagation use the settings computed with this
design.
Default propogation is without redesign, so that path-request-script
must use the ----redesign-per-request option to behave as before this
commit.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I0084185106d14659a846136620cd17791d551a7d
gnpy currently uses the same parameter for tx output power and span
input power: this prevents from modelling low tx power effect.
This patch introduces a new tx-cannel-power and uses it to
propagate in ROADM.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: Id3ac75e2cb617b513bdb38b51a52e05d15af46f5
Finally, ref_carrier is not meant to change after design since
it is the carrier used for design. So let's move its definition
to networks function. Only ROADM need the ref_carrier baud rate
so let's define a dedicated variable in ROADM to hold it.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: Ida7e42dd534a04c8df8792b44980f3fd2165ecb6
Remove the visualisation of the effective_pch in amp because actual
and target are the relevant ones. effective_pch was artificially
related to a mix of reference channel and noisy channel (mixed between
on the fly redesign but using actual ROADM equalisation which includes noise
in its actual loss).
the change does no more rely on the target power (which is rounded)
but on the designed gain, which is not rounded.
Propagations are slightly changed for openroadm simulations because of that.
(I verified)
The gain of amp was estimated on the fly with p_spni also in case of
RamanFiber preceding elements. removing p_spani requies that an estimation
of Raman gain be done during design.
This commit also adds this estimation.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I960b85e99f85a7d168ac5349e325c4928fa5673b
input power is computed at design time: so let's record it and
use it instead of p_span_i for ROADM reference channel loss computation.
Note that this loss parameter is only used for visualisation purpose.
No impact on propagation.
Since this loss is computed for the reference channel used for
design, we need to record input power based on input degrees,
and indicate this information within the call function.
Note that this will be also usefull later on to implement ROADM
parameters
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I64d510fc20df72f07158f400964d592d76dc0ce4
Prepare for the next step, to be able to handle lists of candidate assignment
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I2bd78606ce4502f68efb60f85892df5f76d52bb5
Offset power is used for equalization purpose to correct for the
generic equalization target set in ROADM for this particular transceiver.
This is usefull to handle exception to the general equalization rule.
For example in the case of constant power equalization, the user might
want to apply particular power offsets unrelated to slot width or baudrate.
or in constant PSW, the user might want to have a given mode equalized for
a different value than the one computed based on the request bandwidth.
For example consider that a transceiver mode is meant to be equalized with
75 GHz whatever the spacing specified in request. then the user may specify
2 flavours depending on used spacing:
service 1 : mode 3, spacing 75GHz
service 2 : mode 4, spacing 87.5Ghz
avec
{
"format": "mode 3",
"baud_rate": 64e9,
"OSNR": 18,
"bit_rate": 200e9,
"roll_off": 0.15,
"tx_osnr": 40,
"min_spacing": 75e9,
"cost": 1
}
{
"format": "mode 4",
"baud_rate": 64e9,
"OSNR": 18,
"bit_rate": 200e9,
"roll_off": 0.15,
"tx_osnr": 40,
"min_spacing": 87.5e9,
"equalization_offset_db": -0.67,
"cost": 1
}
then the same target power would be considered for mode3 and mode4
despite using a different slot width
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I437f75c42f257b88b24207260ef9ec9b1ab7066e
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
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
A new function is added to build spectrum information based on
the actual mixture of channels to be simulated (baud rate, slot width,
power per frequency).
Propagation function is changed so that, if the user defines a
specific distribution, then it uses it, else it uses as before,
all identical channels based on the initial request. In this case,
as before this change, we assume full load, with same channel for
the spectral info and not the resulting mixt of channels after
routing.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: Icf56396837b77009e98accd27fcebd2dded6d112
When mode is not given and propagation is performed bidir, penalties
corresponding to automatically selected mode are not correctly
initialized in request, and gnpy-path-request fails.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: If045624ff5dad7f0dfdec93eaa05bb5eae86e643
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
The compare_reqs function checks if N and M values of the requests are
None but these attributes may not exist e.g. if a service file does not
define an "effective-freq-slot" for a request.
Signed-off-by: Jonas Mårtensson <jonas.martensson@ri.se>
Change-Id: I786aad97ed658cd703694f164a87525d77b51fe1
Change tests based on M==0 value for response creation and use
instead the blocking_reason attribute existence
result element should have non null M value if request is not blocked.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I67e4222cf9d014201e91d3aefd3624b001264e03
In case user defines trx_mode, it is possible to check consistency of
nb of required slots and the total requested path_bandwidth and raise
a service error, before staring any propagation computation.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I543cab581280faef5d6072eb172da136f2542492
For this commit only the first element from the {N, M} list is read
and assigned.
This is better than not reading this value at all.
the commit also updates test_files and test data files with correct
values for the effective_freq_slot attribute
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I1e60fe833ca1092b40de27c8cbfb13083810414e
if the user has specified a nb of slot and has not specified a mode
it may happen that the nb of slot is finally not large enough to support
the requested traffic: then blocking reason is 'NOT_ENOUGH_RESERVED_SPECTRUM'
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I8d4c4df5fa97e37aefac8d9ee0d93c901505fa55
When a path is blocked for 'NO_FEASIBLE_MODE' reason, and bidir is true,
the request attributes are filled with the last explored mode values
(baudrate notably), and the reversed path is propagated with this last
explored mode specs. if this reversed path is also not feasible the blocking
reason was overwritten with a 'MODE_NOT_FESIBLE' reasonn, because
baudrate is filled in the request attribute.
This change ensure that the blocking reason (if it exists) is not overwritten.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: If80a37d77e2b967a327562c733a44e7f78f1c544
without this change if a request was blocked on main path due to 'NO_FEASIBLE_BAUDRATE_WITH_SPACING'
and if the request was bidir, there was a propagation tentative on the reversed path
despite the fact that no baudrate was selected, which ended up with a program crash.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: Ie4e578aab944e534d8b2d99fb02c4e28a242e717
Previous implementation selected last candidate in case both routing
and disjunction constraints could not be applied. The new implementation
elaborates an alternate list where each feasible paths satisfyng
disjunction constraint but not route constraint is recorded. The algorithm
then preferably selects a feasible path that satisfies all constraints and
if none is found and route constraint is LOOSE, the first set of paths
that satisfy disjunction is selected (instead of the last one).
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: Iba44397d105006a98539494d821cc83dc3e3bbd9
- add the per degree info using the EXACT next node uid as identifier
of the degree when a node is a roadm
- add the degree identifier on the propagate and on the call functions
- use the per degree target_pch_out_db defined in json entry for the
target power in network build
- verifies existence of the per degree power target in order to support
partial per degree target power definition
- correct test data files for expected auto design results that now
should include the per degree information, even if it is the same
for all degree.
- in order to enable per degree power definition on direction where
booster is not defined, enable the declaration of edfas in json without
specifying type variety
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: I5004cbb250ca5fd6b6498ac9d4b9c4f28a65efee
In the previous implementation, when a constraint was added on top of
the disjunction, there could be an infinite loop because the test was
not correctly performed on the request_id but on a request class.
Moreover, there the loop was not needed since the first feasible
candidate is selected if it exist.
This patch removes the loop and adds some comments to explain the code.
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: Icdbee9032f28214b3e03cb62d55ea353477d94bf
Since 30234f913c, the code just added the
"system margins" to each transponder's minimal required OSNR. That's
simple and straightforward, but I think that mutating "equipment
library" in response to a "global simulation parameter" is not really
the right way to go.
Make this explicit; always check the resulting OSNR against the
transponder's minimum plus the margin.
I got into this when checking the fixups that are performed within the
JSON loader. I don't think that the JSON loader is an appropriate place
for this.
Change-Id: Ic8f8385cbbb29dc0d769462f17caad47c2b36940
There are two propagation functions in request.py, propagate and
propagate2. The propagate2 function saves and returns spectral infos
from before and after every element of the path. In most cases only the
final spectral info after propagation is needed (there is currently
nothing in the code using anything more). On the other hand, the
propagate function just returns the path which is unnecessary since that
is modified in place. This patch proposes to instead return the final
spectral info from the propagate function.
Since propagate2 is now redundant, it can be removed.
Change-Id: I3a0f7310a7d5f9c93dc5f257e051c7d45e20c6fe
Signed-off-by: Jonas Mårtensson <jonas.martensson@ri.se>
The parameter add_drop_osnr is specified in the Roadm section of
eqpt_config.json and represents noise added by amplifiers within the
add/drop block of ROADMs. Currently this noise is added to the signal
whether the propagation path includes any ROADMs or not, which does not
make sense to me. This patch proposes to only add the add_drop_osnr when
a path actually includes ROADMs.
See GitHub issue #274.
Change-Id: I58961772c049578eff8879dfb2e53265866d12c4
Signed-off-by: Jonas Mårtensson <jonas.martensson@ri.se>
(Migrated from GitHub PR #311)
Currently the output of transmission_main_example shows the first
transceiver having ONSR = +inf even when a tx_osnr has been specified,
which is confusing.
This patch proposes to update the SNR of the first transceiver with
tx_osnr.
Change-Id: Idab7c92c2f5a12cc92ce5c1c551e5710f30e6a02
Signed-off-by: Jonas Mårtensson <jonas.martensson@ri.se>
This patch proposes a new implementation of the compute_constrained_path function based on the same method as the newly proposed compute_k_constrained_paths function, i.e. using shortest_simple_paths instead of all_simple_paths. This method is more efficient and avoids having to set a cutoff parameter. The new implementation should be identical to the old one from an external perspective, except that it finds a path with include node constraints in more cases.
Change-Id: Ia93b61c0af27076ed5088013bc87787a2920b629
Signed-off-by: Jonas Mårtensson <jonas.martensson@ri.se>
This has been the only place where content from gnyp.tools.service_sheet
was being used outside of tests and examples. It looks like something
which is actually *not* used anywhere (the ResultElement instances are
only ever apended to a list which gets used as-is, so we do not need any
custom comparisons or hashing).
Change-Id: Ib6ddcf55779218d602620e77973d88ad62d0ec7b
