Data Reduction for real-time bridge vibration data on Edge

Chen, Anthony; Liu, Fu-Hsuan; Wang, Sheng‐De

doi:10.1109/dsaa.2019.00077

Cited by 11 publications

(7 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Second, it dynamically adjusts the rate at which features are computed from the original signal. Chen et al [16] proposed to extract the data features based on fast Fourier transform (FFT) and apply K-means to generate a set of patterns to represent the time-series data in the application of reducing real-time bridge vibration data. Wang et al [17] proposed an energy-efficient load balancing tree-based data aggregation scheme (LB-TBDAS) for grid-based WSNs.…”

Section: Data Aggregationmentioning

confidence: 99%

Autonomous Internet of Things (IoT) Data Reduction Based on Adaptive Threshold

Zhang,

Na,

Zhang

2023

Sensors

View full text Add to dashboard Cite

With the development of intelligent IoT applications, vast amounts of data are generated by various volume sensors. These sensor data need to be reduced at the sensor and then reconstructed later to save bandwidth and energy. As the reduced data increase, the reconstructed data become less accurate. Usually, the trade-off between reduction rate and reconstruction accuracy is controlled by the reduction threshold, which is calculated by experiments based on historical data. Considering the dynamic nature of IoT, a fixed threshold cannot balance the reduction rate with the reconstruction accuracy adaptively. Aiming to dynamically balance the reduction rate with the reconstruction accuracy, an autonomous IoT data reduction method based on an adaptive threshold is proposed. During data reduction, concept drift detection is performed to capture IoT dynamic changes and trigger threshold adjustment. During data reconstruction, a data trend is added to improve reconstruction accuracy. The effectiveness of the proposed method is demonstrated by comparing the proposed method with the basic Kalman filtering algorithm, LMS algorithm, and PIP algorithm on stationary and nonstationary datasets. Compared with not applying the adaptive threshold, on average, there is an 11.7% improvement in accuracy for the same reduction rate or a 17.3% improvement in reduction rate for the same accuracy.

show abstract

Section: Data Aggregationmentioning

confidence: 99%

Autonomous Internet of Things (IoT) Data Reduction Based on Adaptive Threshold

Zhang,

Na,

Zhang

2023

Sensors

View full text Add to dashboard Cite

show abstract

“…The terminology used in Fig. 2 is as following: a (0) is a vector of N inputs a (2) is a vector of M outputs w (0) is a [PxN] matrix of weights values between input and hidden layer w (1) is a [MxP] matrix of weights values between hidden and output layer N, P and M is a number of the neurons in input, hidden and output layer respectively Since the first (input) layer does not do any computations (serves only for input signals propagation to other layers), the output of hidden layer can be simply expressed (using formula (1)) in a matrix form as:…”

Section: Multilayer Perceptronmentioning

confidence: 99%

“…where a (1) is the vector of outputs from the hidden layer (or inputs to the output layer). Full matrix equation for hidden layer output can be written in the following form (see formula 5).…”

Section: Multilayer Perceptronmentioning

confidence: 99%

“…Artificial intelligence (AI) is nowadays very expanding area of interest for most of the researchers as well as technicians in the industrial area. Connection of AI and technical diagnostics, especially in the predictive maintenance of machines in the industry [4], is very common and popular topic also thanks to Industry 4.0 and Internet of Things (IoT) [1]. E.g.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Matlab Implementation Of Multilayer Perceptron For Bearing Faults Classification

Doseděl¹

2021

Proceedings II of the 27st Conference STUDENT EEICT 2021

View full text Add to dashboard Cite

This paper deals with implementation of multilayer perceptron neural network (NN) for bearing faults classification. Neural network has been created from scratch as an M-script with back propagation learning algorithm also, but without using advanced MATLAB packages. Public available bearing dataset from Case Western Reserve University has been used for both training and testing phase, as well as for the final classification process. Problem with sparse input data for training the network has also been addressed. This relatively simple and small neural network is capable to classify the failures of a bearing with very low error rate.

show abstract

“…However, this approach may significantly increase the cost of AIoT applications, as AIoT networks often deploy devices on a large scale. Alternatively, computational tasks can be optimized by reducing the amount of data through internal optimization, such as pre-processing [7]. By reducing the data volume, the number of data points that need to be processed decrease, which in turn reduces the size of the dataset used.…”

Section: Introductionmentioning

confidence: 99%

Two‐step attribute reduction for AIoT networks

Ren,

Lyu,

Wang

et al. 2024

IET Communications

View full text Add to dashboard Cite

The evolution of Artificial Intelligence of Things (AIoT) pushes connectivity from human‐to‐things and things‐to‐things, to AI‐to‐things, has resulted in more complex physical networks and logical associations. This has driven the demand for Internet of Things (IoT) devices with powerful edge data processing capabilities, leading to exponential growth in device quantity and data generation. However, conventional data preprocessing methods, such as data compression and encoding, often require edge devices to allocate computational resources for decoding. Additionally, some lossy compression methods, like JPEG, may result in the loss of important information, which has negative impact on the AI training. To address these challenges, this paper proposes a two‐step attribute reduction approach, targeting devices and dimensions, to reduce the massive amount of data in the AIoT network while avoiding unnecessary utilization of edge device resources for decoding. The device‐oriented and dimension‐oriented attribute reductions identify important devices and dimensions, respectively, to mitigate the multimodal interference caused by the large‐scale devices in the AIoT network and the curse of dimensionality associated with high‐dimensional AIoT data. Numerical results and analysis show that this approach effectively eliminates redundant devices and numerous dimensions in the AIoT network while maintaining the basic data correlation.

show abstract

Data Reduction for real-time bridge vibration data on Edge

Cited by 11 publications

References 8 publications

Autonomous Internet of Things (IoT) Data Reduction Based on Adaptive Threshold

Autonomous Internet of Things (IoT) Data Reduction Based on Adaptive Threshold

Matlab Implementation Of Multilayer Perceptron For Bearing Faults Classification

Two‐step attribute reduction for AIoT networks

Contact Info

Product

Resources

About