Abstract: This invention proposes a method of crowd abnormal behavior detection from video using artificial intelligence, includes three steps: step 1: Data-preprocessing; step 2: Feature extraction and abnormal prediction using a three-dimensional convolution neural network (3D CNN), step 3: Post-processing and synthesizing information to issue warning.

Abstract: A method for multi-object tracking from video. The method includes the following steps: (1) Capturing frames from the streaming source and preprocess the data; (2) Extract video features with three choices: a 3D-CNN backbone followed by a Transformer Encoder, a Video Transformer Encoder, a 2D-CNN Encoder with a stack of frames as input followed by a Transformer Encoder; (3) Multi-object tracking using a new end-to-end multi-task deep learning model named JDAT (Joint Detection Association Transformer), then post-processing and updating tracking state with Temporal Aggregation Module (TAM). The deep learning models in step 2 and step 3 are trained simultaneously end-to-end with a loss function that is accumulated over multiple timesteps (Collective Average Loss—CAL). Also, the model can be pretrained with weakly labeled image dataset in a self-supervised learning manner first, then finetuned on supervised video datasets with full tracking labels.

Abstract: The present invention discloses a DFT (Design for test) methodology to integrate synchronous double-clock latches to synchronize data transferring between two asynchronous clock domains, where the conventional lockup-latches (asynchronous latches) are automatically inserted to connect/combine the two independent asynchronous DFT chains for a more compacted solution. If the asynchronous latches are replaced by the synchronous double-clock ones, the static timing analysis (STA) will check both ends of the latches, therefore, the full combined chain is safe from the setup-hold issues. This method utilizes the automatic flow of asynchronous latches insertion to connect independent DFT chains. Then these conventional latches are replaced by synchronous ones. Both the synthesis, the physical implementation flows in the DFT design and the automatic-test pattern generation (ATPG) scheme have been modified to match these replacements.

Abstract: Described is a design to synchronize and prevent any clock timing issues associated with two clock domain crossing (CDC) in Design for Testing (DFT) and other CDC applications. In order to avoid any meta-stability issues associated with the sensitive edges of the two clock domains, the synchronization scheme splits the incoming data of the first clock domain into three branches, from which a finite state machine is to choose the correct branch for data transfer to the second clock domain. The selected branch will be the ‘safe’ one that guarantees there will be no timing issue at the second clock domain. This methodology can be applied in IC design that needs to optimize area and performance.

Abstract: The method of connection establishment method in data transmission between micro-services in cloud ensures reducing connection complexity between client instances and server instances, increasing the number of connections which helps increase data transmission performance. Besides, the method helps ensure load balancing when server instances change or in time of failure. The proposed method includes the following steps: step 1: data preparation; Step 2: build connection data table; Step 3: select the appropriate connection when receiving a request to send data from the client; step 4: periodically queries the independent service management, updating the current state of the server instances.

Abstract: The invention relates to an explosive mass drop test device. Specifically, the present invention relates to a method of designing an explosive block drop test device to serve the explosion test process and meet the requirements for ensuring safety during the test. The present invention gives an example of a test device for dropping explosive masses in the range 50-300 kg. In addition, the structure is designed to be simple, easy to integrate and disassemble with the explosive block, the manipulation of dropped objects is simplified. The product of the present invention can be used to test the safety of explosive blocks.

Abstract: The invention offers an ultracapacitor-based power system solution with four main functional blocks which are power conditioning block, monitoring block, charge-discharge block and protection block. The proposed system has the advantage of working well in the environment of vibration, high temperature, has a large capacity to provide a large amount and radiates less heat compared to systems using traditional batteries. In addition, the system has functions to protect and stabilize the output voltage, and the operating parameters of the system is monitored continuously.

Abstract: A method to detect, identify and determine the position of laser points from infrared laser emitters in a firearm training simulation system. The method consists of two steps. This first step combines some common tools such as a camera with a filter switcher and edge and corner detecting algorithms in image processing to provide a fast automatic screen calibrating method. The second step is to detect laser point objects in each frame continuously captured by the camera, and then further process these objects to identify laser shots, filter out noisy objects, and eventually detect simulated shots from different laser emitters and determine the final mapped position of the shot.

Abstract: Method of analysis text message syntactically and by content, which entails: step 1; Split syntaxes (made available to subscribers by the network operator) into tokens to store in a Syntax Trie; step 2. Pre-process an incoming text from a subscriber; step 3. Split the text (pre-processed in Step 2) into tokens; step 4. Look up paths that include the tokens (obtained in Step 3) in the Syntax Trie (initialized in Step 1); step 5: Return the look-up result, which is the path in the Syntax Trie that best reflects the user intent.

Abstract: The factual event detection method for online news consists of five main steps: Step 1: Preprocess raw text to split a document into sentences and words, find part-of-speech tags (POS tags), and digitize data; Step 2: Encode digitized words and their POS tags into vectors; Step 3: Detect triggers and their event types; Step 4: Concatenate the surrounding encoded vectors of each trigger into an aggregate vector; Step 5: Classify the realis status of each event. The method chosen is used in systems that automatically collect, extract, and analyze data in order to monitor and early warn about the activities of targets mentioned in digital newspapers.

Abstract: The present invention provides a method of masked face recognition from images by artificial intelligence technology including four steps: step 1: generating the images of face wearing a mask; step 2: training a deep learning model for face detection while wearing a mask; step 3: training the deep learning model for face feature extraction while wearing a mask; step 4: building a full pipeline of masked face recognition from images using the trained models from step 2, step 3, and some post-processing algorithms. The method aims to improve the accuracy of identity verification in the context of wearing masks has become popular and compulsory in life.

Abstract: The present invention provides a method of finding service endpoints that can be customized. The method includes the following steps: step 1: discover the “services” customized by the CSDE method, called the “services” CSDE; step 2: discover groups of running Pods belonging to the respective CSDE “services”; step 3: register the IP of the Pod to the corresponding CSDE “service” according to the requirement to use a specific network range. The method is deployed to the corresponding CSDE application for installation in a container virtualization environment.

Abstract: The invention proposes False targets caused by reflection position determining system and solution for the SSR which helps determine reflected real target's coordinate, reflector's coordinate and features, coordinate and features of the false target (multipath target) and using these data as the basis to build up the multipath target suppressing system, guaranteeing the detecting ability of the radar. The proposed system contains: Input data generating component; False target position (multipath position) determining component; Reflected signal power comparison calculating component; Displaying and data transferring component; Reflected multipath targets suppressing component.

Abstract: The invention provides the method that allows UE to simultaneously use 4G and 5G technology when handover from a eNodeB station to eNodeB ENDC base station that supports data transfer on 5G technology platform effectively and optimize deployment costs for current carriers. The method can be widely applied on software and hardware available.

Abstract: A system and method for live video streaming over tactical wireless networks based on distortion estimation using multilayer perceptron neural network (MLP). The system and method can adaptively adjust the parameters of transport and application layers according to wireless network conditions, based on video distortion estimation using MLP neural network. The system includes a video motion and detail estimation block, a redundancy and video bitrate optimizer, a MLP neural network-based video distortion estimator, and video transport blocks, a packet error rate (PER) and bandwidth measurement block.

Abstract: A self-balancing Two-Rotation Driving Mechanism (TRDM) uses parallelogram and has low weight, high load, high accuracy, quick response and good adaptation with multi equipment. A longitudinal rotation motor assembly is constrained fully with a base assembly and drives a longitudinal rotation assembly through connecting rod parallelogram. The lateral rotation motor assembly is assembled in the hub of a longitudinal rotation assembly and rotates an equipment clamping assembly in a lateral direction. Generally, the TRDM can drive equipment clamping assembly in two rotations, with longitudinal rotation in the range of ±45° and lateral rotation in range of ±20°. A rotation limitation is reached when limitation shafts touch the end of respective grooves.

Abstract: A method of adjusting and self-stabilizing the receiver side in 5G including: initial synchronization set and the set of constructing parameters, collecting the initial defining phase value of the reference channel and data; Determine and evaluate the signal lag and the first crest string capacity on the PRACH (Physical Random Access Channel) offer the first assessment coefficient; Compare the first assessment coefficient to the adjustment requirement if greater than the adjustment threshold continues to process data and evaluate the detailed adjustment of the detailed level on the PUSCH (Physical Uplink Share Chanel) schedule a processing money, giving the second assessment coefficient. If less than the adjustment requirements perform the next step: from the above evaluation coefficients perform late compensation on the conversion card (transfer card-two-way data transfer department from BBU and RRU) and power on RRU with periodic or instant update time according to the adjustment rule.

Abstract: The present invention provides next-hop monitoring method for border gateway protocol in layer 3 virtual private network models. The invention helps determine the fastest possible next hops of a route. If the next-hop is invalid or there is a change in cost when it comes to the next-hop, an update is needed to ensure convergence time.

Abstract: The invention provides a method for automatically analyzing and filtering redundant alarms in radio transceiver systems consisting of the following steps: step 1: the operator shall define the relationship between the alarms, defining rules to filter redundant alarms at the FRDU block; step 2: the FSU block will detect the alarm and send it to the FAFU block; step 3: FAFU block will receive alarms from FSU, based on the rules defined in FRDU block will analyze and filter out redundant alarms; Step 4: The FSMU block will receive the alarms after being filtered, stored in the database and also send these alarms to the EMS system.

Abstract: A method for uplink closed loop power control in advanced wireless systems which adaptively adjusts target signal-to-interference noise ratio (SINRtarget) in order to achieve the best uplink throughput of data services is disclosed. To derive the desired target SINR, the system collects and evaluates various uplink parameters as inputs: real-time signal-to-interference noise ratio of data physical channel, terminal power headroom, and terminal buffer data status and data service requirements.