Trained Long Short-term Memory Research Articles

Deep‐DRX: A framework for deep learning–based discontinuous reception in 5G wireless networks

AbstractThe tremendous advancement in various types of mobile devices with distinct services demands emerging 5G networks to deal with different kinds of traffic. The high data rates along with beam searching operation in 5G increase user equipment (UE) energy expenses. Discontinuous reception (DRX) is used in long‐term evolution network to save the power of UE. The DRX conserves the energy at the cost of increased latency. On the other hand, long short‐term memory (LSTM), a deep neural network, has shown incredible results in learning time‐varying sequences from data sets. In this article, we propose a novel idea to train LSTM network for prediction of next packet arrival time based on real wireless traffic trace (dataset). Then, we use the trained LSTM model to predict dynamic sleep time in DRX for 5G networks. Our proposed algorithm, deep learning–based DRX (Deep‐DRX) is able to make dynamic sleep cycle in 5G communications. Deep‐DRX achieves 10% and 30% of power saving with a mean delay of 1.006 ms and 1.05 ms for Trace 1 and Trace 2, respectively.

Rolling Bearing Initial Fault Detection Using Long Short-Term Memory Recurrent Network

The complete failure of the rolling bearing is a deterioration process from the initial minor fault to the serious fault, it is meaningless for guiding maintenance when the serious fault is alarmed. This work presents a novel initial fault diagnosis framework based on sliding window stacked denoising auto-encoder (SDAE) and long short-term memory (LSTM) model. In this approach, multiple vibration value of the rolling bearings are entered into SDAE by sliding window processing. Then, multiple vibration value of the rolling bearings of the next period is predicted from the signal reconstructed by the trained SDAE in the previous period using LSTM. For the given input data, the reconstruction errors between the next period data and the output data generated by trained LSTM are used to detect initial anomalous conditions. The proposed method not only utilizes the ability of SDAE to learn the inherent distribution of data, but also ensures that LSTM can extract timing relationships between data cycles, and the model is built using only normal data. The initial fault detection as a key difficulty in the operating condition monitoring and performance degradation assessment of the rolling bearing is effectively solved. Experimental and classic rotating machinery datasets have been employed to testify the effectiveness of the proposed method and its preponderance over some state-of-the-art methods. The experiment results indicate that the proposed method can effectively detect the initial anomalies of the rolling bearing and accurately describe the deterioration trend with strong robustness, and have high significance for maintenance guiding.

Model-Based Prediction of Dissolved Oxygen Content in Fish Production

Abstract. Dissolved oxygen (DO) plays a vital role in fish production, but DO sensors are expensive and often malfunction as a result of corrosion and deposition of debris. In this work, polynomial regression, back-propagation neural network (BPNN), and long short-term memory (LSTM) were applied to develop models for estimation of DO from non-DO measurements in fish production. The root mean square error (RMSE), mean absolute percentage error (MAPE), coefficient of determination (R2), index of agreement (d), and Nash-Sutcliffe efficiency coefficient (NSE) were used to measure the performance of the three models. The results showed that the LSTM model achieved the best performance. A trained LSTM model can be used to estimate DO from water pH, temperature, and turbidity without using DO sensors. The estimated DO values can be used as a warning signal for DO sensor maintenance and as estimates of compromised DO measurements in fish production. HighlightsPrediction models were developed to estimate dissolved oxygen (DO) and their performance.The LSTM model achieved the best performance and can be used to estimate DO without using DO sensors.The estimated DO can be used as a warning signal for DO sensor malfunction and maintenance.The estimated DO can also be used to correct compromised DO measurements. Keywords: BP neural network, Dissolved oxygen, Long short-term memory, Polynomial regression, Principal component analysis.

Bidirectional handshaking LSTM for remaining useful life prediction

Abstract Unpredictable failures and unscheduled maintenance of physical systems increases production resources, produces more harmful waste for the environment, and increases system life cycle costs. Efficient remaining useful life (RUL) estimation can alleviate such an issue. The RUL is predicted by making use of the data collected from several types of sensors that continuously record different indicators about a working asset, such as vibration intensity or exerted pressure. This type of continuous monitoring data is sequential in time, as it is collected at a certain rate from the sensors during the asset's work. Long Short-Term Memory (LSTM) neural network models have been demonstrated to be efficient throughout the literature when dealing with sequential data because of their ability to retain a lot of information over time about previous states of the system. This paper proposes using a new LSTM architecture for predicting the RUL when given short sequences of monitored observations with random initial wear. By using LSTM, this paper proposes a new objective function that is suitable for the RUL estimation problem, as well as a new target generation approach for training LSTM networks, which requires making lesser assumptions about the actual degradation of the system.

Forecasting Root-Zone Electrical Conductivity of Nutrient Solutions in Closed-Loop Soilless Cultures via a Recurrent Neural Network Using Environmental and Cultivation Information.

In existing closed-loop soilless cultures, nutrient solutions are controlled by the electrical conductivity (EC) of the solution. However, the EC of nutrient solutions is affected by both growth environments and crop growth, so it is hard to predict the EC of nutrient solution. The objective of this study was to predict the EC of root-zone nutrient solutions in closed-loop soilless cultures using recurrent neural network (RNN). In a test greenhouse with sweet peppers (Capsicum annuum L.), data were measured every 10 s from October 15 to December 31, 2014. Mean values for every hour were analyzed. Validation accuracy (R2) of a single-layer long short-term memory (LSTM) was 0.92 and root-mean-square error (RMSE) was 0.07, which were the best results among the different RNNs. The trained LSTM predicted the substrate EC accurately at all ranges. Test accuracy (R2) was 0.72 and RMSE was 0.08, which were lower than values for the validation. Deep learning algorithms were more accurate when more data were added for training. The addition of other environmental factors or plant growth data would improve model robustness. A trained LSTM can control the nutrient solutions in closed-loop soilless cultures based on predicted future EC. Therefore, the algorithm can make a planned management of nutrient solutions possible, reducing resource waste.

Prediction of a New Kind of MR Data

In the field of TD-LTE network problem analysis, compared with traditional methods such as DT(Driver Test) and CQT(Call Quality Test), MR (Measurement Report) has the advantages of comprehensive information and high efficiency, begins to get more and more attention and application. In order to solve the problem of limited open time of measurement data acquisition system, a data prediction method based on LSTM (Long Short-Term Memory) model is proposed. Selecting part of the MR parameters as the experimental object, training LSTM Model with measurement data in a district of Beijing. Experimental results show that the proposed method can predict MR data accurately. Compared with the traditional prediction model ARIMA (Autoregressive Integrated Moving Average Model), this method has lower prediction error and more stable generalization ability.