Abstract: An apparatus/method for extracting multiple features from time series data collected from a plurality of sensors and for performing transfer learning on them. There is provided an apparatus including: a multi-feature extraction unit for extracting multiple features from a data stream for each sensor inputted from the plurality of sensors; a transfer-learning model generation unit for extracting useful multi-feature information from a learning model which has finished pre-learning, for the multiple features for forwarding the extracted multi-feature information to a multi-feature learning unit to generate a learning model that performs transfer learning on the multiple features; and the multi-feature learning unit for receiving learning variables from the learning model for each of the multiple features and for performing parallel learning for the multiple features, to calculate and output a loss. In addition, there is provided an apparatus for detecting leaks in plant pipelines.

Abstract: Provided are an apparatus and method for detecting an anomaly in a plant pipe using multiple meta-learning. When a multi-sensor data stream about a plant pipe is received, each of a plurality of meta-learning modules for processing different packet section ranges, extracts one or more preset types of features from sensor data of packet section ranges set according to trend from an arbitrary reception time point, generates 2D image features of the features according to multi-sensor-specific times, generates 3D volume features by accumulating the 2D image features in a depth direction according to multiple sensors, and learns the 3D volume features in parallel through multi-sensor-specific learning modules. Results of the learning of the meta-learning modules are aggregated, and it is determined whether there is an anomaly in a plant pipe according to a learning result selected based on an optimal combination of multiple features, multiple sensors, and multiple packet sections.

Abstract: Provided are an apparatus and method for detecting an anomaly in a plant pipe using multiple meta-learning. When a multi-sensor data stream about a plant pipe is received, each of a plurality of meta-learning modules for processing different packet section ranges, extracts one or more preset types of features from sensor data of packet section ranges set according to trend from an arbitrary reception time point, generates 2D image features of the features according to multi-sensor-specific times, generates 3D volume features by accumulating the 2D image features in a depth direction according to multiple sensors, and learns the 3D volume features in parallel through multi-sensor-specific learning modules. Results of the learning of the meta-learning modules are aggregated, and it is determined whether there is an anomaly in a plant pipe according to a learning result selected based on an optimal combination of multiple features, multiple sensors, and multiple packet sections.

Abstract: In a sensor network including a sink node and a plurality of sensor nodes, the sink node receives a routing message including routing information from the sensor nodes, analyzes a structure of the sensor network based on routing information, and calculates a transmission schedule of each sensor node based on the analyzed network structure. Schedule information according to the calculated transmission schedule includes a wakeup time of the sensor node, a time interval that is consumed for packet transmission, and a standby time for preventing a packet collision based on the time interval.

Abstract: Provided is a private electric generator including: a first heat absorbing panel that absorbs heat corresponding to temperature; a second heat absorbing panel that absorbs heat corresponding to ground temperature or water temperature; and a thermoelectric generator that is disposed between the first and second heat absorbing panels and uses a temperature difference in the heat absorbed in the first and second heat absorbing panels to generate power, thereby generating power based on a difference in temperature and ground temperature or water temperature according to a daily temperature range.

Abstract: An apparatus and method are provided, which predict total nitrogen using general water quality data measured in real time. The total nitrogen prediction apparatus may include a regression model selection unit to select a regression model comprising general data of at least one water quality based on a correlation coefficient of the general data of at least one water quality, a quality-of-fit evaluation unit to evaluate quality of fit of the selected regression model, a regression model change unit to determine whether to change the regression model based on the quality of fit and change the regression model according to the determination result, and a total nitrogen prediction unit to predict total nitrogen of a body of water based on the regression model.

Abstract: Provided are a method and a system for measuring total phosphorus that may predict total phosphorus of a river valley using multi-parameter water quality that are measured in real time through a multi-parameter water quality measuring unit and the like, and may increase the accuracy thereof. The total phosphorus measuring method according to the present disclosure includes: computing a correlation between the multi-parameter water quality and the total phosphorus using multi-parameter water quality data and total phosphorus data measured for a predetermined period; selecting upper parameters having a high correlation from among the multi-parameter water quality based on the computation result; generating a total phosphorus prediction model through a regression analysis between the upper parameters and the total phosphorus; measuring the multi-parameter water quality; and predicting the total phosphorus by replacing the total phosphorus prediction model with the measured multi-parameter water quality.

Abstract: A method and apparatus for determining a length of a transmission frame based on a Link Quality Indicator (LQI) are provided. A target transmitter may request an LQI from a target receiver, and the target receiver may transmit information about the LQI to the target transmitter. The target transmitter may determine the length of the transmission frame based on the LQI. The target transmitter) may determine the length of the transmission frame proportional to the LQI.

Abstract: A maintenance method for a subscriber line of a B-NT in an ATM PVC switching system which is capable of more expanding a maintenance range of an OAM cell loopback function disclosed in the article 2. The method includes the steps of a first step in which a loopback OAM cell is transmitted to a B-NT when the ATM PVC switching system receives a request of a state information of subscriber links of a B-NT from an operator, or a loopback operation timer becomes time-out, a second step in which the B-NT, which receives a loopback OAM cell from the ATM PVC switching system, carries the statement information of subscriber lines of B-NT apparatus on a loopback OAM cell and then transmits the information to the ATM PVC switching system, and a third step in which the ATM PVC switching system, which receives a loopback OAM cell from the B-NT apparatus, stores a state information of subscriber lines into a state management table, reports a state information of subscriber lines to the system operator.