Abstract: A system and method are provided for enhanced multi-way time transfer for time synchronization between at least one slave node and one master node. A slave node sends a first message to the master node to launch a time synchronization between the slave node and the master node. Upon receiving the first message, the master node adds a receiving time on a master clock to the first message to form a second message. The master node sends the second message back to the slave node and the slave node adds a receiving time on the slave clock to the second message to form an updated message. The slave node performs a time adjustment to the slave clock based on the updated message, thereby synchronizing time between the slave node and the master node.

Abstract: A method of generating a deep neural network (DNN) may comprise receiving one or more application-level requirements associated with network communications, translating the one or more application-level requirements into one or more technical constraints, and providing the one or more technical constraints to a control loop that generates a certified DNN architecture based on the technical constraints. The control loop may further comprise a DNN search engine and a hardware synthesis engine. The method may comprise selecting, using the DNN search engine, a candidate DNN architecture based on the technical constraints, and generating, using the hardware synthesis engine, a hardware architecture corresponding to the selected candidate DNN architecture.

Abstract: A hidden chamber detector includes a linear frequency modulated continuous wave (LFMCW) radar, a synthetic aperture radar (SAR) imaging processor, and a time division multiple access (TDMA) multiple input multiple output (MIMO) antenna array, including a plurality of transmitting and receiving (Tx-Rx) antenna pairs. A Tx-Rx antenna pair is selected, in a time division manner, as a Tx antenna and an Rx antenna for the LFMCW radar. The LFMCW radar is configured to transmit an illumination signal, receive an echo signal, convert the echo signal to a baseband signal, collect baseband samples, and send the collected samples to the SAR imaging processor. The SAR imaging processor is configured to receive the collected samples, collect structure/configuration of the antenna array and scanning information, and form an SAR image based on the collected samples, the structure/configuration of the antenna array, and the scanning information.

Abstract: An apparatus includes a camera for capturing an image at a first moment; a range finder for measuring a distance to an object at a center of the image; a rotatable mounting platform, fixedly hosting the camera and the range finder; and a controller. The controller is configured to receive the captured image and the measured distance; determine whether a target of interest (TOI) appears in the image; in response to determining a TOI appearing in the image, determine whether the TOI appears at the center of the image; calculate position parameters of the rotatable mounting platform for centering the TOI in an image to be captured at a second moment, separated from the first moment by a pre-determined time interval; control the rotatable mounting platform to rotate according to the calculated position parameters; and calculate and store the position parameters of the TOI with respect to the apparatus.

Abstract: A system includes at least one slave node and one master node that are for a method of time synchronization between the at least one slave node and the master node. The method includes: sending, by one slave node, a first message to the master node to launch a time synchronization between the slave node and the master node; upon receiving the first message, adding, by the master node, a receiving time on a master clock to the first message to form a second message; sending, by the master node, the second message back to the slave node; adding, by the slave node, a receiving time on the slave clock to the second message to form an updated message; and performing, by the slave node, a time adjustment to the slave clock based on the updated message, thereby synchronizing time between the slave node and the master node.

Abstract: A computing system includes: a memory, containing instructions for a method for anomaly and pattern detection of unstructured big data via semantic analysis and dynamic knowledge graph construction; a processor, coupled with the memory and, when the instructions being executed, configured to: receive unstructured big data associated with social network interactions, events, or activities; parse and structure the unstructured big data to generate structured big data; form a dynamic knowledge base based on the structured big data; and perform sematic reasoning on the dynamic knowledge base to discover patterns and anomalies among the social network interactions, events, or activities; and a display, comprising an interactive graphical user interface (GUI), configured to receive the anomalies and patterns to display real-time actionable alerts, provide recommendations, and support decisions.

Abstract: A hidden chamber detector includes a linear frequency modulated continuous wave (LFMCW) radar, a synthetic aperture radar (SAR) imaging processor, and a time division multiple access (TDMA) multiple input multiple output (MIMO) antenna array, including a plurality of transmitting and receiving (Tx-Rx) antenna pairs. A Tx-Rx antenna pair is selected, in a time division manner, as a Tx antenna and an Rx antenna for the LFMCW radar. The LFMCW radar is configured to transmit an illumination signal, receive an echo signal, convert the echo signal to a baseband signal, collect baseband samples, and send the collected samples to the SAR imaging processor. The SAR imaging processor is configured to receive the collected samples, collect structure/configuration of the antenna array and scanning information, and form an SAR image based on the collected samples, the structure/configuration of the antenna array, and the scanning information.

Abstract: A method for detecting fake images includes: obtaining an image for authentication, and hand-crafting a multi-attribute classifier to determine whether the image is authentic. Hand-crafting the multi-attribute classifier includes fusing at least an image classifier, an image spectrum classifier, a co-occurrence matrix classifier, and a one-dimensional (1D) power spectrum density (PSD) classifier. The multi-attribute classifier is trained by pre-processing training images to generate an attribute-specific training dataset to train each of the image classifier, the image spectrum classifier, the co-occurrence matrix classifier, and the 1D PSD classifier.

Abstract: Various embodiments of the present disclosure provide a method, a device, and a storage medium for performance prediction of a communication waveform in a communication system. The method includes measuring, by a receiver, an actual SNR distribution of a communication link between a transmitter and the receiver; further includes evaluating, by a waveform performance prediction device, a normalized minimum SNR shift required for the communication waveform to operate, where the normalized minimum SNR shift is obtained based on a normalized SNR distribution using a neural network (NN), the normalized SNR distribution corresponding to the actual SNR distribution; and further includes, according to the normalized minimum SNR shift, obtaining, by a waveform performance prediction device, an actual minimum SNR shift for the actual SNR distribution, where according to the actual minimum SNR shift, the communication system is adjusted for operation.

Abstract: A method for rapid discovery of satellite behavior, applied to a pursuit-evasion system including at least one satellite and a plurality of space sensing assets. The method includes performing transfer learning and zero-shot learning to obtain a semantic layer using space data information. The space data information includes simulated space data based on a physical model. The method further includes obtaining measured space-activity data of the satellite from the space sensing assets; performing manifold learning on the measured space-activity data to obtain measured state-related parameters of the satellite; modeling the state uncertainty and the uncertainty propagation of the satellite based on the measured state-related parameters; and performing game reasoning based on a Markov game model to predict satellite behavior and management of the plurality of space sensing assets according to the semantic layer and the modeled state uncertainty and uncertainty propagation.

Abstract: A method for point cloud compression of an intelligent cooperative perception (iCOOPER) for autonomous air vehicles (AAVs) includes: receiving a sequence of consecutive point clouds; identifying a key point cloud (K-frame) from the sequence of consecutive point clouds; transforming each of the other consecutive point clouds (P-frames) to have the same coordinate system as the K-frame; converting each of the K-frame and P-frames into a corresponding range image; spatially encoding the range image of the K-frame by fitting planes; and temporally encoding each of the range images of the P-frames using the fitting planes.

Abstract: A method for emulating a communication system with fast changing link condition, applied to a testbed system including an interference emulator, a transmitter, and a receiver, includes: by the transmitter, generating code-word symbols, in an initial symbol sequence, according to information bits; evaluating an interference condition and a time duration for each code-word symbol; reordering the code-word symbols to a reordered symbol sequence based on interference conditions; evaluating a total time duration of code-word symbols under each interference condition; and transmitting the code-word symbols in the reordered symbol sequence. The method also includes: by the interference emulator, obtaining the total time duration of code-word symbols under each interference condition from the transmitter; and providing an emulated interference environment by ordering the time durations of different interference conditions.

Abstract: The present disclosure provide a system, a method, and a storage medium for distributed joint manifold learning (DJML) based heterogeneous sensor data fusion. The system includes a plurality of nodes; and each node includes at least one camera; one or more sensors; at least one memory configured to store program instructions; and at least one processor, when executing the program instructions, configured to obtain heterogeneous sensor data from the one or more sensors to form a joint manifold; determine one or more optimum manifold learning algorithms by evaluating a plurality of manifold learning algorithms based on the joint manifold; compute a contribution of the node based on the one or more optimum manifold learning algorithms; update a contribution table based on the contribution of the node and contributions received from one or more neighboring nodes; and broadcast the updated contribution table to the one or more neighboring nodes.

Abstract: The present disclosure provides a method for wave propagation prediction based on a 3D ray tracing engine and machine-learning based dominant ray selection. The method includes receiving, integrating, and processing input data. Integrating and processing the input data includes dividing a cone of the original millimeter wave (mmWave) into a plurality of sub cones; determining a contribution weight of rays coming from each sub cone to the received signal strength (RSS) at a receiving end of interest; and determining rays coming from one or more sub cones that have a total contribution weight to the RSS larger than a preset threshold value as dominant rays using a neural network obtained through a machine learning approach. The method further includes performing ray tracing based on the input data and the dominant rays to predict wave propagation.

Abstract: Embodiments of the present disclosure provide a method, a device, and a storage medium for domain adaptation for efficient learning fusion (DAELF). The method includes acquiring data from a plurality of data sources of a plurality of sensors; for each of the plurality of sensors, training an auxiliary classifier generative adversarial network (AC-GAN) by a hardware processor with data from each data source of the plurality of data sources, thereby obtaining a trained feature extraction network and a trained label prediction network for each data source; forming a decision-level fusion network or a feature-level fusion network; and training the decision-level fusion network or the feature-level fusion network with a source-only mode or a generate to adapt (GTA) mode; and applying the trained decision-level fusion network or the trained feature-level fusion network to detect a target of interest.

Abstract: A method for emulating a communication system with fast changing link condition, applied to a testbed system including an interference emulator, a transmitter, and a receiver, includes: by the transmitter, generating code-word symbols, in an initial symbol sequence, according to information bits; evaluating an interference condition and a time duration for each code-word symbol; reordering the code-word symbols to a reordered symbol sequence based on interference conditions; evaluating a total time duration of code-word symbols under each interference condition; and transmitting the code-word symbols in the reordered symbol sequence. The method also includes: by the interference emulator, obtaining the total time duration of code-word symbols under each interference condition from the transmitter; and providing an emulated interference environment by ordering the time durations of different interference conditions.

Abstract: The present disclosure provides a method for joint manifold learning based heterogenous sensor data fusion, comprising: obtaining learning heterogeneous sensor data from a plurality sensors to form a joint manifold, wherein the plurality sensors include different types of sensors that detect different characteristics of targeting objects; performing, using a hardware processor, a plurality of manifold learning algorithms to process the joint manifold to obtain raw manifold learning results, wherein a dimension of the manifold learning results is less than a dimension of the joint manifold; processing the raw manifold learning results to obtain intrinsic parameters of the targeting objects; evaluating the multiple manifold learning algorithms based on the raw manifold learning results and the intrinsic parameters to determine one or more optimum manifold learning algorithms; and applying the one or more optimum manifold learning algorithms to fuse heterogeneous sensor data generated by the plurality sensors.

Abstract: An apparatus includes a camera for capturing an image at a first moment; a range finder for measuring a distance to an object at a center of the image; a rotatable mounting platform, fixedly hosting the camera and the range finder; and a controller. The controller is configured to receive the captured image and the measured distance; determine whether a target of interest (TOI) appears in the image; in response to determining a TOI appearing in the image, determine whether the TOI appears at the center of the image; calculate position parameters of the rotatable mounting platform for centering the TOI in an image to be captured at a second moment, separated from the first moment by a pre-determined time interval; control the rotatable mounting platform to rotate according to the calculated position parameters; and calculate and store the position parameters of the TOI with respect to the apparatus.

Abstract: Various embodiments of the present disclosure provide a method, a device, and a storage medium for performance prediction of a communication waveform in a communication system. The method includes measuring, by a receiver, an actual SNR distribution of a communication link between a transmitter and the receiver; further includes evaluating, by a waveform performance prediction device, a normalized minimum SNR shift required for the communication waveform to operate, where the normalized minimum SNR shift is obtained based on a normalized SNR distribution using a neural network (NN), the normalized SNR distribution corresponding to the actual SNR distribution; and further includes, according to the normalized minimum SNR shift, obtaining, by a waveform performance prediction device, an actual minimum SNR shift for the actual SNR distribution, where according to the actual minimum SNR shift, the communication system is adjusted for operation.

Abstract: The present disclosure provides a method for wave propagation prediction based on a 3D ray tracing engine and machine-learning based dominant ray selection. The method includes receiving, integrating, and processing input data. Integrating and processing the input data includes dividing a cone of the original millimeter wave (mmWave) into a plurality of sub cones; determining a contribution weight of rays coming from each sub cone to the received signal strength (RSS) at a receiving end of interest; and determining rays coming from one or more sub cones that have a total contribution weight to the RSS larger than a preset threshold value as dominant rays using a neural network obtained through a machine learning approach. The method further includes performing ray tracing based on the input data and the dominant rays to predict wave propagation.