Abstract: The invention provides a low-power wide area network integrated sensing and communication method based on channel sensing and reinforcement learning, and the method comprises the steps: receiving an uplink signal transmitted by a terminal node through a gateway receiver, demodulating the uplink signal through employing a soft demodulation algorithm, and obtaining a bit log-likelihood ratio; the deep reinforcement learning model completes channel reconstruction of the corresponding link during communication according to the bit log-likelihood ratio; and calculating the current optimal network configuration according to the reconstructed channel so as to distribute the current optimal network configuration to each terminal node during next downlink transmission.

Abstract: A method and a system for recognizing a two-dimensional (2D) movement track based on a smart watch is provided. The method comprises: acquiring accelerometer signal data and gyroscope signal data of the smart watch; estimating a tilt angle of the smart watch by using the accelerometer signal data and correcting the gyroscope signal data by using the tilt angle; and calculating angle value information of the smart watch by using the corrected gyroscope signal data and estimating a coordinate point. According to the present application, the movement track of the smart watch can be accurately estimated by using the accelerometer and the gyroscope built in the smart watch.

Abstract: The invention provides a low-power wide area network integrated sensing and communication method based on channel sensing and reinforcement learning, and the method comprises the steps: receiving an uplink signal transmitted by a terminal node through a gateway receiver, demodulating the uplink signal through employing a soft demodulation algorithm, and obtaining a bit log-likelihood ratio; the deep reinforcement learning model completes channel reconstruction of the corresponding link during communication according to the bit log-likelihood ratio; and calculating the current optimal network configuration according to the reconstructed channel so as to distribute the current optimal network configuration to each terminal node during next downlink transmission.

Abstract: Disclosed is a few-shot gesture recognition method. The method comprises the following steps: customizing, by a user, gesture categories, and acquiring few samples for each gesture category; inputting the acquired samples into a trained few-shot learning model, extracting a feature vector corresponding to each sample, and synthesizing feature vectors belonging to the same gesture to obtain an average feature vector corresponding to each gesture as a prototype vector; acquiring a corresponding sample for a target gesture implemented by the user, and inputting the sample into the few-shot learning model to obtain a feature vector of the target gesture as a query vector; and calculating similarities between the query vector and prototype vectors of different gestures, and selecting a gesture category corresponding to the prototype vector with the highest similarity as a prediction category of the target gesture.

Abstract: The present application discloses a cloud-network integration (CNI) oriented multi-access edge computing (MEC) architecture. An access network (AN) side of the architecture is provided with a plurality of edge computing nodes, a physical channel of the AN side is split into a plurality of subchannels with each of the subchannels supporting a media access control (MAC) access mode, and a software defined network (SDN) controller is arranged in the architecture and is configured to allocate resources of the subchannels at a physical layer and protocols at an MAC layer and control offloading at the edge computing nodes or a cloud center by a terminal user. The present application can carry out fine-grained control and cooperative management on network resources and computing resources, and achieve more effective computation offloading and service enhancement.

Abstract: A method for user recognition and emotion monitoring based on a smart headset is provided. The smart headset includes an earplug part and a main body, wherein the earplug part is provided with a first microphone and a wearing detection sensor, and a housing of the main body is provided with a signal amplification circuit, a communication module, and a microcontroller. The wearing detection sensor is to detect whether a user wears the smart headset, and the first microphone is to obtain a sound signal in an ear canal. The sound signal is amplified, and then is outputted to the microcontroller. The amplified sound signal is transmitted by the microcontroller to a smart terminal paired with the smart headset to extract a heart sound signal characteristic, and legality of—identity of the user is validated and emotional state of the user is inferred according to the heart sound signal characteristic.

Abstract: The present application discloses a deep learning method and system for spectrum sharing of partially overlapping channels. The method comprises the following steps: in response to a received user transmission request, inputting, by a base station, channel state information CSI of a plurality of historical time slices into a trained channel prediction convolutional neural network model and outputting predicted channel state information CSI of a next time slice; and inputting the channel state information CSI of the next time slice into a reinforcement learning model and obtaining a channel allocation strategy of each user equipment in a collision domain of the base station so as to realize a maximum throughput of simultaneous communication of each user equipment, wherein the reinforcement learning model is obtained by training by taking bandwidth efficiency performance as a reward. The present application is high in universality, bandwidth utilization rate and throughput for communication networks.

Abstract: The present application discloses a cloud-network integration (CNI) oriented multi-access edge computing (MEC) architecture. An access network (AN) side of the architecture is provided with a plurality of edge computing nodes, a physical channel of the AN side is split into a plurality of subchannels with each of the subchannels supporting a media access control (MAC) access mode, and a software defined network (SDN) controller is arranged in the architecture and is configured to allocate resources of the subchannels at a physical layer and protocols at an MAC layer and control offloading at the edge computing nodes or a cloud center by a terminal user. The present application can carry out fine-grained control and cooperative management on network resources and computing resources, and achieve more effective computation offloading and service enhancement.

Abstract: The present application discloses a massive MIMO wireless energy transmission method based on dynamic frame transmission. The method comprises the following steps: controlling, by a base station, each antenna to transmit a pilot signal to a user end in a time-sharing mode by using set time-sharing pilot frames; acquiring, by the user end, downlink channel state information from the antennae of the base station to the user end and feeding the downlink channel state information back to the base station; and calculating, by the base station, a precoding matrix based on the downlink channel state information, mapping data from a user layer to an antenna port by using the newly calculated precoding matrix, and performing beam forming calculation with maximization of an energy signal of the user end as a goal.

Abstract: The present application discloses a human-computer interaction method and an interaction system based on capacitive buttons. The method comprises the following steps: acquiring a capacitive charging signal generated by a user finger-touching a capacitive button; processing the capacitive charging signal to extract Mel-frequency cepstrum coefficient characteristics; and inputting the Mel-frequency cepstrum coefficient characteristics into a trained hidden Markov model, identifying a finger type of the user touching the capacitive button, and further implementing human-computer interaction according to an identification result.

Abstract: A vibration signal-based smartwatch authentication method includes generating incremental vibration signals using a vibration motor in a smartwatch; performing frequency band-based hierarchical endpoint segmentation to obtain vibration signals at a plurality of frequency bands; extracting frequency-domain features for the vibration signals at the plurality of frequency bands; training a dynamic time warping model by taking the vibration signals at the plurality of frequency bands as a training data set, training a nearest neighbor model by taking the extracted frequency-domain features as training data; collecting to-be-authenticated vibration signals which are processed to serve as test data signals; discriminating similarities between the test data signals and corresponding training data signals through the dynamic time warping model, giving a classification result through the nearest neighbor model, performing weighted calculation on a discrimination result of the dynamic time warping model and a discrimin

Abstract: Disclosed is a few-shot gesture recognition method. The method comprises the following steps: customizing, by a user, gesture categories, and acquiring few samples for each gesture category; inputting the acquired samples into a trained few-shot learning model, extracting a feature vector corresponding to each sample, and synthesizing feature vectors belonging to the same gesture to obtain an average feature vector corresponding to each gesture as a prototype vector; acquiring a corresponding sample for a target gesture implemented by the user, and inputting the sample into the few-shot learning model to obtain a feature vector of the target gesture as a query vector; and calculating similarities between the query vector and prototype vectors of different gestures, and selecting a gesture category corresponding to the prototype vector with the highest similarity as a prediction category of the target gesture.

Abstract: A monitoring method and system based on a magnetic field of a loudspeaker are provided. This method includes: collecting a magnetic field signal near the loudspeaker, and converting the magnetic field signal into a magnetic field digital signal; converting the magnetic field digital signal into a speech signal and removing an interference from the speech signal to obtain a filtered speech signal; removing noise in the speech signal, and performing short-time inverse Fourier transform on a speech power spectrum to obtain a denoised speech signal; equalizing response of the denoised speech signal to each frequency to convert the speech signal into an intelligible speech audio signal; and recognizing the speech audio signal and converting the same into a text content.

Abstract: A smart device input method based on facial vibration includes: collecting a facial vibration signal generated when a user performs voice input; extracting a Mel-frequency cepstral coefficient from the facial vibration signal; and taking the Mel-frequency cepstral coefficient as an observation sequence to obtain text input corresponding to the facial vibration signal by using a trained hidden Markov model. The facial vibration signal is collected by a vibration sensor arranged on glasses. The vibration signal is processed by: amplifying the collected facial vibration signal; transmitting the amplified facial vibration signal to the smart device via a wireless module; and intercepting a section from the received facial vibration signal as an effective portion and extracting the Mel-frequency cepstral coefficient from the effective portion by the smart device.

Abstract: An optimization method for UAV-based wireless information and energy transmission includes following steps: S1: reporting, by a wireless device, an energy state of the wireless device to a UAV; S2: detecting, by the UAV, a channel state between the UAV and the wireless device; and S3: selecting, by the UAV, an optimal action based on estimated revenue maximization according to an electric quantity of the UAV, an electric quantity of the wireless device, and the channel state. The use of the wireless device can reduce wiring costs, beautify the space, and ensure a smaller size and a lower power. By applying the UAV to information and energy transmission for the wireless devices, the data transmission rate and the energy conversion efficiency of networks are improved.

Abstract: The present application discloses a massive MIMO wireless energy transmission method based on dynamic frame transmission. The method comprises the following steps: controlling, by a base station, each antenna to transmit a pilot signal to a user end in a time-sharing mode by using set time-sharing pilot frames; acquiring, by the user end, downlink channel state information from the antennae of the base station to the user end and feeding the downlink channel state information back to the base station; and calculating, by the base station, a precoding matrix based on the downlink channel state information, mapping data from a user layer to an antenna port by using the newly calculated precoding matrix, and performing beam forming calculation with maximization of an energy signal of the user end as a goal.

Abstract: A method and a system for recognizing a two-dimensional (2D) movement track based on a smart watch is provided. The method comprises: acquiring accelerometer signal data and gyroscope signal data of the smart watch; estimating a tilt angle of the smart watch by using the accelerometer signal data and correcting the gyroscope signal data by using the tilt angle; and calculating angle value information of the smart watch by using the corrected gyroscope signal data and estimating a coordinate point. According to the present application, the movement track of the smart watch can be accurately estimated by using the accelerometer and the gyroscope built in the smart watch.

Abstract: A vibration-based authentication method for an access control system includes: collecting vibration signals generated by a built-in vibration motor in an authentication device; filtering, denoising, and performing endpoint segmentation on the collected vibration signals, and extracting vibration signals containing effective touch; performing an alignment on the segmented vibration signals; performing a fast Fourier transform on the aligned vibration signals to obtain frequency-domain data, extracting frequency-domain features obtained after alignment and features obtained before alignment to construct a training data set, and storing the training data set in a database of the authentication device; using a new unlock signal generated when a user touches the authentication device as test data, and processing the test data to obtain test data containing effective touch; and matching and classifying the test data containing effective touch with the training data set by using a machine learning classification mod

Abstract: The present application discloses a human-computer interaction method and an interaction system based on capacitive buttons. The method comprises the following steps: acquiring a capacitive charging signal generated by a user finger-touching a capacitive button; processing the capacitive charging signal to extract Mel-frequency cepstrum coefficient characteristics; and inputting the Mel-frequency cepstrum coefficient characteristics into a trained hidden Markov model, identifying a finger type of the user touching the capacitive button, and further implementing human-computer interaction according to an identification result.

Abstract: A monitoring method and system based on a magnetic field of a loudspeaker are provided. This method includes: collecting a magnetic field signal near the loudspeaker, and converting the magnetic field signal into a magnetic field digital signal; converting the magnetic field digital signal into a speech signal and removing an interference from the speech signal to obtain a filtered speech signal; removing noise in the speech signal, and performing short-time inverse Fourier transform on a speech power spectrum to obtain a denoised speech signal; equalizing response of the denoised speech signal to each frequency to convert the speech signal into an intelligible speech audio signal; and recognizing the speech audio signal and converting the same into a text content.