Zhejiang University

Silergy Semiconductor Technology (hangzhou) Ltd

Yu Qi

Gang Pan

Yueming Wang

Wei Chen

Chen Zhao

Jianmin Zhang

Gao Fan

Tao Fang

Kedi Xu

Xinyun Zhu

Hang Yu

Junming Zhu

Xiaomeng Yang

A lumped power supply circuit for converting an AC signal into a DC signal, the lumped power supply circuit including: a cascaded H-bridge circuit having N H-bridge sub-circuits connected in series between two input terminals of the AC signal, and being configured to convert the AC signal into N first voltage signals, where N is a positive integer greater than or equal to 2; a high-frequency filtering module configured to filter the N first voltage signals, and to generate N second voltage signals; a DC conversion module to receive the N second voltage signals, and to convert the N second voltage signals into at least one third voltage signal; and a lumped power buffer module having an output terminal coupled to a load, and being configured to receive the at least one third voltage signal, and to filter out part of power frequency fluctuations in the third voltage signal.

Lumped power supply circuit for converting an AC signal into a DC signal

Disclosed in the present invention is a method for decoding Chinese character writing for an invasive brain-computer interface. In a practical application, a corresponding motor neural signal is divided into two states of a writing stroke and a writing stroke break in view of inconsistency of the writing stroke and the writing stroke break during Chinese character writing, and different filters are trained. A hidden markov model (HMM algorithm) and a Viterbi algorithm are used to judge a task state of the motor neural signal, and the corresponding signal is put into a corresponding decoder. The present invention effectively reduces influence of difference of neural data in different states on the decoder, and improves the performance and robustness of the decoder.

Chinese character writing and decoding method for invasive brain-computer interface

A multilevel self-balance control circuit can include: a voltage divider unit configured to receive and divide an input voltage; a voltage-controlled charge source load coupled to an output terminal of the voltage divider unit, and being configured to adaptively adjust charge amount input to the voltage-controlled charge source load based on an output voltage of the voltage divider unit, such that a total amount of charges flowing through the voltage-controlled charge source load during a period of each working state of the voltage divider unit is positively correlated with the output voltage of the voltage divider unit, thereby forming a negative feedback loop to achieve voltage balancing of the voltage divider unit; and a control unit configured to generate control signals for the voltage divider unit and the voltage-controlled charge source load, thereby coordinately controlling the voltage divider unit and the voltage-controlled charge source load.

Multilevel self-balance control circuit, DC/DC conversion system and AC/DC conversion system

A cascade circuit can include: N power conversion units connected in series between two ports of a power supply, where N is a positive integer greater than or equal to 2; a controller connected to one of the N power conversion units, and being configured to send a signal to be transmitted through the connected power conversion unit; where each of the power conversion units is configured to send the signal to be transmitted to a next-stage power conversion unit when the each of the power conversion unit shares a reference voltage with the adjacent next-stage power conversion unit; and where the signal to be transmitted is controlled to be transmitted from a previous-stage power conversion unit to a next-stage power conversion unit in sequence until the signal to be transmitted is received by all of the N power conversion units.

Cascade circuit, control method and integrated circuit thereof

The present invention discloses an adaptive brain-computer interface decoding method based on multi-model dynamic ensemble, where a traditional state-space model is improved, and a set of measurement functions instead of one fixed measurement function are used to dynamically characterize a relationship between observation variables and state variables; and, by using a pool of linear and nonlinear decoders, and in a decoding process of a brain-computer interface system, decoders are automatically switched according to the data, so as to realize adaptive brain signal decoding. Through the above multi-model ensemble strategy, linear and nonlinear decoder capabilities can be integrated, the accuracy and stability of the brain-computer interface system can be improved, and decoding unstability caused by the non-stationary neural signal of the brain-computer interface system can be solved to a certain extent.

Adaptive brain-computer interface decoding method based on multi-model dynamic integration

The present invention discloses an online neuron spike sorting method based on neuromorphic computing, which converts neuron spike signals collected from the cerebral cortex into spike signals through field coding, classifies different waveforms and corresponding time stamps by means of spiking neural networks, and realizes online neuron spike sorting; at the same time, the online update method of spiking neural network is used to adapt to the online changes of neuronal spike waveform and improve the accuracy of long-term online neuronal spike sorting. This method has fast computational speed, which can improve the speed of spike sorting process, maintain high consistency in classification on different datasets, and facilitate the deployment of implanted chips.

Online neuron classification method based on neuromorphic computing

The present invention discloses a brain machine interface decoding method based on spiking neural network, comprising: (1) constructing a liquid state machine model based on a spiking neural network, the liquid state machine model consists of an input layer, an middle layer and an output layer, wherein, a connection weight from the input layer to the middle layer is W, a loop connection weight inside the middle layer is W, a readout weight from the middle layer to the output layer is W; (2) Inputting a neuron spike train signal, and training each weight with the following strategy: (2-1) Using STDP without supervision to train the connection weight Wfrom the input layer to the middle layer; (2-2) Setting the loop connection weight Winside the middle layer by means of distance model and random connection, and obtaining a middle layer liquid information R(t); (2-3) Using ridge regression with supervision to train the readout weight Wfrom the middle layer to the output layer, and establishing a mapping between the middle layer liquid information R(t) and the output motion information, and finally outputting a predicted motion trajectory. The present invention can quickly train a model in a relatively short time, predict the arm motion trajectory in real time, and achieve an improvement in efficiency and accuracy.

Growing community of inventors

Hangzhou, China

Yu Qi