Atout PI Laplace

A method of characterizing or controlling a quantum system (1), wherein the dynamics of the quantum system is describable via a stochastic master equation having a plurality of physical parameters, and wherein the quantum system comprises: (i) one or more quantum-state-hosting structure(s) (6); (ii) a detector (8) coupled to at least one of the quantum-state-hosting structure(s) (6) and configured to measure a continuous signal output from the quantum-state-hosting structure(s) (6) to provide a measurement signal from the quantum system (1), wherein the detector (8) is configured to digitize the continuous signal over time bins of duration Δt; and (iii) one or more signal generator(s) (11,13) coupled to the at least one portion and configured to input control signal(s) to physically control the dynamics of the quantum system. The method comprises: (I) physically measuring, with the detector, a continuous signal from the quantum system to provide a plurality of digitized discrete-time binned signals {I1 , ... , Ik} up to time t = kΔt, wherein k is an integer denoting the k-th time bin of the detector; (II) computing a binned state of the quantum system at time t = kΔt, ρk =E ρ k Δ t I 1 , … , I k <img class="EMIRef" id="a3ed2f85-c5f5-464f-b330-0908b35248cb-ia01" /> , wherein the binned state ρk is the average of a plurality of solutions ρkΔt of the stochastic master equation at time t = kΔt each for different possible continuous signals wherein, for each of the plurality of solutions ρkΔt , the value of the possible continuous signal integrated over a given time bin is substantially equal to the discrete-time binned signal for said given time bin; and (III) using the binned state ρk to: assign a value to at least one of the plurality of physical parameters which characterizes the quantum system; and/or assign a value to at least one of the plurality of physical parameters which parametrizes the input control signal(s) to physically control the dynamics of the quantum system; and/or optimize a readout discriminator of a quantum state readout protocol. There is provided a quantum characterisation or control unit, a computer program product, and a computer-readable medium.

A system and method of encoding a binary datastream for wireless transmission combining characteristics of OFDM and FBMC type systems is proposed, based on a precoding schemes adapted to maintaining real orthogonality so that the output of the OTFS processing consists of real-valued symbols. The precoder is intricately linked to the definition of a subsequent 2d linear phase rotation. Two possible precoder architectures are proposed, in which interleaver and conjugate operations process partwise or en bloc the incoming data matrix, before the original matrix is concatenated with the interleaved and conjugated matrix (or its components) to obtain a new matrix for subcarrier allocation and filter bank encoding prior to transmission. Corresponding receiving mechanisms are proposed.

Un procédé de mise en sécurité de N équipements (23) dotés d'un système de stockage électrochimique connecté à un système de conversion électrique bidirectionnel (22) pour la charge/décharge, comprend, suite à une augmentation de température : évaluation d'un niveau de dangerosité par équipement (23) ; calcul de puissances optimales de décharge respectives pour des systèmes de stockage électrochimique de certains au moins des équipements (23), en fonction au moins des niveaux de dangerosité respectifs évalués pour ces équipements ; commandes aux systèmes de conversion électrique bidirectionnels (22), pour décharger les systèmes de stockage électrochimique des équipements en fonction des puissances optimales de décharge calculées.

There is provided a method of performing a data-data controlled-unitary gate between a first data qubit as control in a first basis (|e₀〉, |e₁〉) and a second data qubit as target, and wherein the unitary gate of the controlled-unitary matrix is defined by a unitary matrix U, the method comprising the following operations: (i) providing the first data qubit in the first basis (|e₀〉, |e₁〉), the second data qubit, and an ancilla qubit, wherein each data qubit is connected to the ancilla qubit and wherein the first and second data qubits are not directly connected to each other, wherein the first data, second data, and ancilla qubits are all either (A) physical qubits each hosted in a respective physical mode of one or more physical resonators or (B) logical qubits each hosted by a plurality of physical qubits which are configured to perform an error correction code to encode the respective logical qubits; (ii) configuring the ancilla qubit to be able to perform a controlled-unitary gate with either of the first and second data qubits. The method further comprises the following operations: (iii) performing a first one-qubit state teleportation of the first data qubit to the ancilla qubit from the first basis (|e₀〉, |e₁〉) to a second basis (|g₀〉, |g₁〉) by: (a) preparing a two-qubit state entangling the first data qubit and the ancilla qubit, and (b) subsequently performing a measurement-based uncomputation of the first data qubit; (iv) performing an ancilla-data controlled-unitary gate between the ancilla qubit as control in the second basis (|g₀〉, |g₁〉) and the second data qubit as target; (vi) performing a second one-qubit state teleportation of the ancilla qubit to the first data qubit from the basis (|g₀〉,|g₀〉) to the basis (|e₀〉, |e₁〉) by: (a) preparing a second two-qubit state entangling the ancilla qubit and the first data qubit, and (b) subsequently performing a measurement-based uncomputation of the ancilla qubit. Operation (iv) is performed at any time before operation (vi)(b), operation (vi)(a) is performed after operation (iii)(b), and wherein (I) the unitary gate of the ancilla-data controlled-unitary gate is defined by the unitary matrix U, or (II) the unitary gate of the ancilla-data controlled-unitary gate is defined by another unitary matrix V and the method comprises a sequence of one or more gates that, in combination with the unitary V-gate of the ancilla-data controlled-unitary gate, performs the function of applying the unitary U-gate on the second data qubit. There is also provided a quantum system, a computer program product, and a computer-readable medium.

Method of performing a CNOT gate between a control qubit having a control qubit resonance frequency and a target qubit having a target qubit resonance frequency, in which said control qubit and said target qubit are cat qubits hosted respectively in a control resonator and in a target resonator of a superconducting quantum circuit, and are stabilized therein by means of a command circuit arranged for selectively applying radiation to said superconducting quantum circuit, and in which said control resonator and said target resonator are connected via an ancilla resonator coupled to said command circuit for stabilizing a respective ancilla qubit having an ancilla qubit resonance frequency, and wherein said control qubit and said target qubit are not directly connected. Said method comprises the following operations: a) preparing (400, 700) the ancilla qubit in a state e^iϕ1Z|+〉 where ϕ 1 is a phase comprised in the range [0; π] measured around the Z axis of the ancilla qubit, b) performing a CNOT gate (410, 710) between said control qubit and said ancilla qubit, with the control qubit being the target and the ancilla qubit being the control, c1) performing a S gate (432, 732) on said target qubit, c2) performing a CNOT gate (435, 735) between said target qubit and said ancilla qubit, with the target qubit being the target and the ancilla qubit being the control, c3) performing a conjugate transpose S gate (436, 736) on said target qubit, c4) performing a S gate (438, 738) on said ancilla qubit, c5) performing a CNOT gate (439, 739) between said target qubit and said ancilla qubit, with the target qubit being the target and the ancilla qubit being the control, d) preparing (440, 740) the control qubit in a state |0) or |1〉, e) performing a CNOT gate or a zero-controlled NOT gate (450, 750) between said control qubit and said ancilla qubit, with the control qubit being the target and the ancilla qubit being the control, f) performing an MZ operation (420, 720) on said control qubit, g) deriving a quantum correction to be performed depending on the result of operation f), h) performing a measurement in the basis {e^iϕ1Z|+〉,e^iϕ1Z|-〉} (460, 760) on said ancilla qubit, i) deriving a corrective quantum Z gate to be performed depending on the result of operation h), wherein operation f) is performed after operation b) but before operation d), and wherein operation h) is performed after operations e) and c5). There is also provided a quantum system, a computer program product, and a computer-readable medium.

L'invention a pour objet un Générateur (D1) de bit probabiliste (p-bit) comprenant : - une jonction tunnel magnétique (MTJ) ayant une résistance qui fluctue entre au moins deux états résistifs distincts (P, AP) selon sa magnétisation; - un circuit de polarisation (POL) configuré pour injecter un courant de commande (Ic) à travers la jonction tunnel magnétique variable selon une première tension d'entrée (V<in1>) ; le circuit de polarisation (POL) comprenant : ∘ un transistor de contrôle (T<1>) monté en série avec la jonction tunnel magnétique entre deux nœuds d'alimentation et comprenant une couche isolante (BOX) enterrée dans un substrat semi-conducteur (2) formant une grille arrière (G2,<T1>) ; - des moyens de contrôle (CONT) configurés pour appliquer la première tension d'entrée (V<in1>) sur ladite grille arrière; - un circuit de détection (DET) configuré pour générer un signal de détection (s1) variable selon l'état résistif de la jonction tunnel magnétique.

There is provided a serializer (3) configured to receive an input feature map represented by a Q-dimensional tensor and to convert the received input feature map into a series of 1D vectors, each corresponding to a convolutional patch. The serializer comprises a memory device (300) comprising a plurality of ports and of memory banks adapted to store data words, the ports comprising a least a write port and at least two read ports. The serializer (3) is adapted to perform write operations consisting in writing a number of 1D vectors to a number of banks of said memory device (300) using the writing port, while being adapted for reading in parallel convolutional patches corresponding to stored 1D vectors from banks of the memory device (300) using the at least two reading ports, the extracted convolutional patches being fed to a convolutional filtering device (90) implementing convolutional filters.

There is provided an accelerator (1) implementing a convolutional neural network, CNN, to determine an output feature map, in response to the receipt of an input feature map. The accelerator (1) comprises at least one serializer (3) to serialize the input feature map into a series of 1D serialized vectors, and a plurality of layersets (2). The serializer (3) broadcasts the serialized vectors to at least one of the layersets, the layersets being interconnected in a pipelined fashion, each layerset corresponding to a pipeline stage. Each layerset (2) comprises filter cores (22) to perform the equivalent computation of a CNN layer. The output of a given interconnected layerset is directly transmitted to the next interconnected layerset, without storage of the given layerset output in an external memory, the last interconnected layerset being configured to provide the output feature map. This enables merging CNN layers.

A method of measuring self-Kerr in a physical quantum system configured to host a bosonic code qubit, comprises the following operations:1) providing a physical quantum system configured to host a bosonic code qubit, said physical quantum system comprising (i) a quantum circuit comprising oscillators hosting a buffer or readout mode and a memory mode, and a non-linear element coupling the buffer or readout mode to the memory mode, and (ii) a control circuit configured to apply one or more control signals to the quantum circuit to stabilize the bosonic code qubit in the memory mode and further configured to receive measurement signals from at least the buffer or readout mode,2) define a coherent state (α) for the memory mode, and, for each of a plurality of coherent state amplitudes and each of a plurality of durations (t),a. prepare, using the control circuit, said coherent state (α) in the memory having one of said plurality of coherent state amplitudes,b. wait for a period of time having one of said duration (t),c. apply, using the control circuit, one or more control signals to said quantum circuit such that the quantum circuit is in a resonant regime |i * ω_a - ω_b| = j * |ω_CS| if said one or more control signals comprise an AC component having an angular frequency ω_CS or |i * ω_a - ω_b| = 0 if said one or more control signals comprise only DC components, resulting in an interaction, mediated via said non-linear element, in said physical quantum system which results in a Hamiltonian having a leading term of the general formula H_i,(j,0) ∝ ξ_(j,0)(a^†)ⁱb + h.c. ... , where is the annihilation operator of said memory mode, i is an integer superior or equal to 1, ω_a is the angular frequency of the memory mode whilst the quantum circuit is in the resonant regime, j is an integer superior or equal to 1, b is the annihilation operator of the buffer or readout mode, ω_b is the angular frequency of the buffer or readout mode whilst the quantum circuit is in the resonant regime, and ξ_(j,0) is the strength of said interaction, andd. simultaneously to operation 2)c., using the control circuit, receiving measurement signals comprising at least the mode phase and/or mode amplitude of the buffer or readout mode,3) derive an effective detuning or a decayed mode amplitude for each coherent state amplitude, by fitting of the measurements of operation 2)d. which are associated with said each coherent state amplitude,4) derive at least one of self-Kerr coefficients, stark-shift detuning, and memory dephasing rate from the plurality of effective detunings or decayed mode amplitudes of operation 3) by fitting of a function of the absolute magnitude squares of coherent state amplitude in said memory mode.

A non-linear superconducting quantum circuit (1) including a memory mode (3), a buffer mode (5), a SNAIL (7) coupled to both the memory mode (3) and the buffer mode (5), a SNAIL flux line (9) inductively coupled to the SNAIL (7), and a DC SQUIDs flux line (11) inductively coupled to a tunable inductive element (21). The non-linear superconducting quantum circuit (1) is configured to operate in: • a resonant dissipative mode in which the SNAIL flux line (9) causes the SNAIL (7) to implement three-wave mixing between the memory mode (3) and the buffer mode (5), and the DC SQUIDs flux line (11) varies the buffer resonant frequency to be equal to twice the memory resonant frequency, and • a Kerr Hamiltonian mode in which the SNAIL flux line (9) causes the SNAIL (7) to implement four-wave mixing between the memory mode (3) and the buffer mode (5).