Octave
An Octave is represented in QuAM through the quam.components.octave.Octave class. Below we describe the three steps needed to configuring an Octave in QuAM:
- Creating the Octave
- Adding frequency converters
- Attaching channels
Creating the Root QuAM Machine
Before we get started, we need a top-level QuAM class that matches our components:
from typing import Dict
from dataclasses import field
from quam.core import QuamRoot, quam_dataclass
from quam.components import Octave, OctaveUpConverter, OctaveDownConverter, Channel
@quam_dataclass
class QuAM(QuamRoot):
octave: Octave = None
channels: Dict[str, Channel] = field(default_factory=dict)
machine = QuAM()
This will be used later to generate our QUA configuration
Creating the Octave
Below we show how an Octave is instantiated using some example arguments:
octave = Octave(name="octave1", ip="127.0.0.1", port=80)
machine.octave = octave
We can next retrieve the Octave config QmOctaveConfig, used to create the QuantumMachinesManager
octave_config = octave.get_octave_config()
# The calibration_db and device_info are automatically configured
qmm = QuantumMachinesManager(host={opx_host}, port={opx_port}, octave=octave_config)
At this point the channel connectivity of the Octave hasn't yet been configured. We can do so by adding frequency converters.
Adding Frequency Converters
A frequency converter is a grouping of the components needed to upconvert or downconvert a signal. These typically consist of a local oscillator, mixer, as well as IF, LO, and RF ports. For the Octave we have two types of frequency converters:
- OctaveUpConverter: Used to upconvert a pair of IF signals to an RF signal
- OctaveDownConverter: Used to downconvert an RF signal to a pair of IF signals
We can add all relevant frequency converters as follows:
octave.initialize_frequency_converters()
octave.print_summary()
octave.print_summary() output
Octave (parent unknown):
name: "octave1"
ip: "127.0.0.1"
port: 80
calibration_db_path: None
RF_outputs: QuamDict
1: OctaveUpConverter
id: 1
channel: None
LO_frequency: None
LO_source: "internal"
gain: 0
output_mode: "always_off"
input_attenuators: "off"
2: OctaveUpConverter
id: 2
channel: None
LO_frequency: None
LO_source: "internal"
gain: 0
output_mode: "always_off"
input_attenuators: "off"
3: OctaveUpConverter
id: 3
channel: None
LO_frequency: None
LO_source: "internal"
gain: 0
output_mode: "always_off"
input_attenuators: "off"
4: OctaveUpConverter
id: 4
channel: None
LO_frequency: None
LO_source: "internal"
gain: 0
output_mode: "always_off"
input_attenuators: "off"
5: OctaveUpConverter
id: 5
channel: None
LO_frequency: None
LO_source: "internal"
gain: 0
output_mode: "always_off"
input_attenuators: "off"
RF_inputs: QuamDict
1: OctaveDownConverter
id: 1
channel: None
LO_frequency: None
LO_source: "internal"
IF_mode_I: "direct"
IF_mode_Q: "direct"
IF_output_I: 1
IF_output_Q: 2
2: OctaveDownConverter
id: 2
channel: None
LO_frequency: None
LO_source: "internal"
IF_mode_I: "direct"
IF_mode_Q: "direct"
IF_output_I: 1
IF_output_Q: 2
loopbacks: QuamList = []
We can see five OctaveUpConverter elements in Octave.RF_outputs, and two OctaveDownConverter elements in Octave.RF_inputs, matching with the number of RF outputs / inputs, respectively.
It is important to specify the LO_frequency of the frequency converters that are used, otherwise they will not add information to the QUA configuration when it is generated.
At this point, our Octave does not yet contain any information on which OPX output / input is connected to each OctaveUpconverter / OctaveDownConverter.
This is done in the third stage
Attaching Channels
Once the frequency converters have been setup, it is time to attach the ones that are in use to corresponding channels in QuAM.
In the example below, we connect an IQChannel to the OctaveUpconverter at octave.RF_outputs[1]
from quam.components import IQChannel, InOutIQChannel
machine.channels["IQ1"] = IQChannel(
opx_output_I=("con1", 1),
opx_output_Q=("con1", 2),
frequency_converter_up=octave.RF_outputs[1].get_reference()
)
octave.RF_outputs[1].channel = machine.channels["IQ1"].get_reference()
octave.RF_outputs[1].LO_frequency = 2e9 # Remember to set the LO frequency
Similarly, we can connect an InOutIQChannel to a combination of an OctaveUpConverter and OctaveDownConverter
machine.channels["IQ2"] = InOutIQChannel(
opx_output_I=("con1", 3),
opx_output_Q=("con1", 4),
opx_input_I=("con1", 1),
opx_input_Q=("con1", 2),
frequency_converter_up=octave.RF_outputs[2].get_reference(),
frequency_converter_down=octave.RF_inputs[1].get_reference()
)
octave.RF_outputs[2].channel = machine.channels["IQ2"].get_reference()
octave.RF_inputs[1].channel = machine.channels["IQ2"].get_reference()
octave.RF_outputs[2].LO_frequency = 2e9
octave.RF_inputs[2].LO_frequency = 2e9
Generating the Config
Once everything is setup, we can generate the QUA configuration
qua_config = machine.generate_config()
qua_config
{
"version": 1,
"controllers": {
"con1": {
"analog_outputs": {
"1": {"offset": 0.0},
"2": {"offset": 0.0},
"3": {"offset": 0.0},
"4": {"offset": 0.0},
},
"digital_outputs": {},
"analog_inputs": {"1": {"offset": 0.0}, "2": {"offset": 0.0}},
}
},
"elements": {
"IQ1": {
"operations": {},
"intermediate_frequency": 0.0,
"RF_inputs": {"port": ["octave1", 1]},
},
"IQ2": {
"operations": {},
"intermediate_frequency": 0.0,
"RF_inputs": {"port": ["octave1", 2]},
"smearing": 0,
"time_of_flight": 24,
"RF_outputs": {"port": ["octave1", 1]},
},
},
"pulses": {
"const_pulse": {
"operation": "control",
"length": 1000,
"waveforms": {"I": "const_wf", "Q": "zero_wf"},
}
},
"waveforms": {
"zero_wf": {"type": "constant", "sample": 0.0},
"const_wf": {"type": "constant", "sample": 0.1},
},
"digital_waveforms": {"ON": {"samples": [[1, 0]]}},
"integration_weights": {},
"mixers": {},
"oscillators": {},
"octaves": {
"octave1": {
"RF_outputs": {
"1": {
"LO_frequency": 2000000000.0,
"LO_source": "internal",
"gain": 0,
"output_mode": "always_off",
"input_attenuators": "off",
"I_connection": ["con1", 1],
"Q_connection": ["con1", 2],
},
"2": {
"LO_frequency": 2000000000.0,
"LO_source": "internal",
"gain": 0,
"output_mode": "always_off",
"input_attenuators": "off",
"I_connection": ["con1", 3],
"Q_connection": ["con1", 4],
},
},
"IF_outputs": {
"IF_out1": {"port": ["con1", 1], "name": "out1"},
"IF_out2": {"port": ["con1", 2], "name": "out2"},
},
"RF_inputs": {
"1": {
"RF_source": "RF_in",
"LO_frequency": 2000000000.0,
"LO_source": "internal",
"IF_mode_I": "direct",
"IF_mode_Q": "direct",
}
},
"loopbacks": [],
}
},
}
Combined Example
from typing import Dict
from dataclasses import field
from quam.core import QuamRoot, quam_dataclass
from quam.components import Octave, OctaveUpConverter, OctaveDownConverter, Channel
@quam_dataclass
class QuAM(QuamRoot):
octave: Octave = None
channels: Dict[str, Channel] = field(default_factory=dict)
machine = QuAM()
octave = Octave(
name="octave1",
ip="127.0.0.1",
port=80,
)
machine.octave = octave
octave.initialize_frequency_converters()
octave.print_summary()
from quam.components import IQChannel, InOutIQChannel
machine.channels["IQ1"] = IQChannel(
opx_output_I=("con1", 1),
opx_output_Q=("con1", 2),
frequency_converter_up=octave.RF_outputs[1].get_reference()
)
octave.RF_outputs[1].channel = machine.channels["IQ1"].get_reference()
octave.RF_outputs[1].LO_frequency = 2e9
machine.channels["IQ2"] = InOutIQChannel(
opx_output_I=("con1", 3),
opx_output_Q=("con1", 4),
opx_input_I=("con1", 1),
opx_input_Q=("con1", 2),
frequency_converter_up=octave.RF_outputs[2].get_reference(),
frequency_converter_down=octave.RF_inputs[1].get_reference()
)
octave.RF_outputs[2].channel = machine.channels["IQ2"].get_reference()
octave.RF_inputs[1].channel = machine.channels["IQ2"].get_reference()
octave.RF_outputs[2].LO_frequency = 2e9
octave.RF_inputs[1].LO_frequency = 2e9
qua_config = machine.generate_config()