Abstract: A brain-inspired cognitive learning method can obtain good learning results in various environments and tasks by selecting the most suitable algorithm models and parameters based on the environments and tasks, and can correct wrong behavior. The framework includes four main modules: a cognitive feature extraction module, a cognitive control module, a learning network module, and a memory module. The memory module includes a data base, a cognitive case base, and an algorithm and hyper-parameter base, which store data of dynamic environments and tasks, cognitive cases, and concrete algorithms and hyper-parameter values, respectively. For dynamic environments and tasks, the most suitable algorithm model and hyper-parameter combination can be flexibly selected. In addition, with “good money drives out bad”, mislabeled data is corrected using correctly labeled data, to achieve robustness of training data.

Abstract: An elastic cooperative inference architecture and method for UAV cluster can dynamically update the scheduling policy according to the status of each node, and can deal with the failure of some nodes. In addition, the elastic cooperative inference architecture and method can process larger scale complex models on the embedded devices with limited performance carried by UAVs by means of cooperative inference, so as to improve the accuracy of intelligent applications. At the same time, the elastic cooperative inference architecture and method can adaptively update the allocation strategy when some nodes are unavailable or recovered, and improve the survivability of UAV cluster through elastic coordination.

Abstract: A three-dimensional (3D) spectrum situation completion method and device based on a generative adversarial network includes performing graying and coloring preprocessing based on incomplete 3D spectrum situations from historical or empirical spectrum data obtained by a UAV through sampling a target region, obtaining three-channel incomplete 3D spectrum situation maps displayed in colors, forming a training set based on the incomplete 3D spectrum situation maps; training the generative adversarial network based on the training set and obtaining a trained generator network in the generative adversarial network, performing graying and coloring preprocessing based on a measured incomplete 3D spectrum situation obtained by the UAV through sampling a specified measurement region, obtaining a three-channel measured incomplete 3D spectrum situation map displayed in colors, and using the measured incomplete 3D spectrum situation map as input data to the generator network to obtain a three-channel measured complete 3D

Abstract: A method and a device for measuring a four-dimensional (4D) radiation pattern of an outdoor antenna based on an unmanned aerial vehicle (UAV) are provided. The device includes a measurement path planning unit, a UAV platform unit, a radiation signal acquisition unit, a data command processing unit, and a ground data processing unit. The measurement path planning unit, the radiation signal acquisition unit, and the data command processing unit each are suspended from the UAV platform unit by using a pod. The present disclosure applies to the radiation pattern measurement of an outdoor antenna.

Abstract: A joint link-level and network-level intelligent system and method for dynamic spectrum anti-jamming are provided. The system includes a link-level anti-jamming subsystem and a network-level anti-jamming subsystem. The link-level anti-jamming subsystem sets a reward value as system transmission throughput in a single decision cycle, and a user makes an intelligent frequency usage decision based on the obtained reward value to skip a frequency band in which jamming exists. The network-level anti-jamming subsystem performs reasonable frequency band allocation and management for lower-level sub-users when link-level anti-jamming fails to further enhance a frequency domain anti-jamming capability of the entire system. The users make intelligent frequency usage decisions through a dynamic spectrum anti-jamming algorithm based on reinforcement learning to effectively avoid external malicious jamming, realize dynamic spectrum access, and enhance a frequency domain anti-jamming capability of the system.

Abstract: A method and a device for measuring a four-dimensional (4D) radiation pattern of an outdoor antenna based on an unmanned aerial vehicle (UAV) are provided. The device includes a measurement path planning unit, a UAV platform unit, a radiation signal acquisition unit, a data command processing unit, and a ground data processing unit. The measurement path planning unit, the radiation signal acquisition unit, and the data command processing unit each are suspended from the UAV platform unit by using a pod. The present disclosure applies to the radiation pattern measurement of an outdoor antenna.

Abstract: An intelligent data and knowledge-driven method for modulation recognition includes the following steps: collecting spectrum data; constructing corresponding attribute vector labels for different modulation schemes; constructing and pre-training an attribute learning model based on the attribute vector labels for different modulation schemes; constructing and pre-training a visual model for modulation recognition; constructing a feature space transformation model, and constructing an intelligent data and knowledge-driven model for modulation recognition based on the attribute learning model and the visual model; transferring parameters of the pre-trained visual model and the pre-trained attribute learning model and retraining the transformation model; and determining whether training on a network is completed and outputting a classification result.

Abstract: A method for jointly estimating gain-phase error and direction of arrival (DOA) based on an unmanned aerial vehicle (UAV) array includes: equipping each UAV with an antenna, and forming a receive array through a swarm of multiple UAVs to receive source signals; when an observation baseline of the swarm remains unchanged, changing array manifold through movement of the UAVs, and re-sensing the source signals; for each sensed source signals, calculating a covariance matrix, and obtaining a corresponding noise subspace through eigenvalue decomposition; and constructing a quadratic optimization problem based on the noise subspace and array steering vector, constructing a cost function, and implementing joint estimation of the gain-phase error and the DOA through spectrum peak search. The method can jointly estimate the DOA and gain-phase error and calibrate the gain-phase error, thereby improving accuracy of passive positioning.

Abstract: A method for jointly estimating gain-phase error and direction of arrival (DOA) based on an unmanned aerial vehicle (UAV) array includes: equipping each UAV with an antenna, and forming a receive array through a swarm of multiple UAVs to receive source signals; when an observation baseline of the swarm remains unchanged, changing array manifold through movement of the UAVs, and re-sensing the source signals; for each sensed source signals, calculating a covariance matrix, and obtaining a corresponding noise subspace through eigenvalue decomposition; and constructing a quadratic optimization problem based on the noise subspace and array steering vector, constructing a cost function, and implementing joint estimation of the gain-phase error and the DOA through spectrum peak search. The method can jointly estimate the DOA and gain-phase error and calibrate the gain-phase error, thereby improving accuracy of passive positioning.

Abstract: A brain-inspired cognitive learning method can obtain good learning results in various environments and tasks by selecting the most suitable algorithm models and parameters based on the environments and tasks, and can correct wrong behavior. The framework includes four main modules: a cognitive feature extraction module, a cognitive control module, a learning network module, and a memory module. The memory module includes a data base, a cognitive case base, and an algorithm and hyper-parameter base, which store data of dynamic environments and tasks, cognitive cases, and concrete algorithms and hyper-parameter values, respectively. For dynamic environments and tasks, the most suitable algorithm model and hyper-parameter combination can be flexibly selected. In addition, with “good money drives out bad”, mislabeled data is corrected using correctly labeled data, to achieve robustness of training data.

Abstract: The present invention provides a radar target detection method based on estimation before detection (EBD), which comprises: obtaining pre-detect targets (PDTs) based on conventional pulse-Doppler processing and pre-detection; estimating ranges and speeds of PDTs, i.e., performing parameter EBD; establishing a dimension-reduction observation model of a received signal based on PDTs and parameter thereof; reconstructing a target vector in the dimension-reduction observation model based on a sparse recovery algorithm; and designing a generalized likelihood ratio detector based on the reconstruction result for target detection. The method of the present invention can significantly reduce the radar signal processing loss, and the target detector used in the method has the constant false alarm rate (CFAR) property, so that the weak target detection performance can be greatly improved.

Abstract: A joint link-level and network-level intelligent system and method for dynamic spectrum anti-jamming are provided. The system includes a link-level anti-jamming subsystem and a network-level anti-jamming subsystem. The link-level anti-jamming subsystem sets a reward value as system transmission throughput in a single decision cycle, and a user makes an intelligent frequency usage decision based on the obtained reward value to skip a frequency band in which jamming exists. The network-level anti-jamming subsystem performs reasonable frequency band allocation and management for lower-level sub-users when link-level anti-jamming fails to further enhance a frequency domain anti-jamming capability of the entire system. The users make intelligent frequency usage decisions through a dynamic spectrum anti-jamming algorithm based on reinforcement learning to effectively avoid external malicious jamming, realize dynamic spectrum access, and enhance a frequency domain anti-jamming capability of the system.

Abstract: A protein antigen combination for detecting Alzheimer's disease autoantibodies in human serum sample and an application thereof are disclosed. The protein antigen combination includes at least two protein fragments selected from the following proteins: MAPT, ADARB1, P21, DNAJC8, RAGE, ASXL1, and JMJD2D. Based on the protein antigen composition, a kit for diagnosis, especially for early diagnosis of Alzheimer's disease or its related risks, can be prepared. By using the protein antigen composition provided by the invention, early Alzheimer's disease can be quickly and accurately diagnosed, which has important practical significance.

Abstract: Described are methods of saccharifying starch-containing materials using a glucoamylase, the methods of producing fermentation products and the fermentation products produced by the method thereof.

Abstract: The invention relates to a cognitive SIMO network access method based on cooperative relay, wherein a cognitive base station collects channel responses in a network and judges whether the cognitive users can cooperate with a primary user to achieve the target transmission rate required by the primary user or not, if so, a cognitive SIMO network is accessible to a frequency band licensed to the primary user; otherwise, the cognitive SIMO network is non-accessible. The access method can enable a plurality of the cognitive users and the primary user to simultaneously use the same licensed spectrum in the same geographical position, on the premise of ensuring the target transmission rate of the primary user, and further improve the utilization efficiency of the spectrum as far as possible.

Abstract: A resource distribution method for throughput maximization in a cooperative cognitive SIMO network. The primary user sends data after receiving a cooperation confirmation message, and the cognitive network keeps silent, receives the information data of the primary user and simultaneously decodes the data. The cognitive users which successfully decode the data send the data of themselves to the cognitive base station and forward the data of the primary user; the cognitive users which cannot successfully decode the data only send the data of themselves and do not forward the data of the primary user. The cognitive base station eliminates the interference of the data of primary user and performs beamforming for the signals. According to the combined adjustment the maximum throughput performance can be realized in the cognitive network.

Abstract: The invention relates to a cognitive SIMO network access method based on cooperative relay, wherein a cognitive base station collects channel responses in a network and judges whether the cognitive users can cooperate with a primary user to achieve the target transmission rate required by the primary user or not, if so, a cognitive SIMO network is accessible to a frequency band licensed to the primary user; otherwise, the cognitive SIMO network is non-accessible. The access method can enable a plurality of the cognitive users and the primary user to simultaneously use the same licensed spectrum in the same geographical position, on the premise of ensuring the target transmission rate of the primary user, and further improve the utilization efficiency of the spectrum as far as possible.

Abstract: The invention relates to a resource distribution method for throughput maximization in a cooperative cognitive SIMO network. The primary user sends data after receiving a cooperation confirmation message, and the cognitive network keeps silence, receives the information data of the primary user and simultaneously decodes the data. The cognitive users which successfully decode the data utilize part of their transmission power to send the data of themselves to the cognitive base station, the remaining transmission power is used for helping in forwarding the data of the primary user; the cognitive users which cannot successfully decode the data only send the data of themselves and do not forward the data of the primary user. The cognitive base station eliminates the interference of the data of primary user in the received mixed signals and performs beamforming on the signals after eliminating the interference.