2 results
Filters
Settings
Now showing1 - 2 of 2
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Poster abstract: towards distributed machine learning for data acquisition in wireless sensor networks(Association for Computing Machinery (ACM), 2023-05-09); ; Wireless sensor networks (WSNs) use low-cost sensors to monitor various environments, offering accurate and continuous surveillance. WSNs face a significant challenge in managing their limited energy resources due to communication overhead. To address this issue, we present a novel approach that leverages Neural Network (NN) models to predict data and reduce communication in WSNs. Our solution incorporates NN models on both the sensor and the cloud, enabling predictions to be made at the local level. The sensor sends data to the cloud only when the model is no longer able to predict accurately, cloud then fine-tunes the model based on the received data and sends updated weights of the NN to the sensor, reducing the need for communicating each sensed value to the cloud.Publicationtype: Conference Paper19 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , LightEQ: on-device earthquake detection with embedded machine learning(Association for Computing Machinery (ACM), 2023-05-09); ; The detection of earthquakes in seismological time series is central to observational seismology. Generally, seismic sensors passively record data and transmit it to the cloud or edge for integration, storage, and processing. However, transmitting raw data through the network is not an option for sensors deployed in harsh environments like underwater, underground, or in rural areas with limited connectivity. This paper introduces an efficient data processing pipeline and a set of lightweight deep-learning models for seismic event detection deployable on tiny devices such as microcontrollers. We conduct an extensive hyperparameter search and devise three lightweight models. We evaluate our models using the Stanford Earthquake Dataset and compare them with a basic STA/LTA detection algorithm and the state-of-the-art machine learning models, i.e., CRED, EQtransformer, and LCANet. For example, our smallest model consumes 193 kB of RAM and has an F1 score of 0.99 with just 29k parameters. Compared to CRED, which has an F1 score of 0.98 and 293k parameters, we reduce the number of parameters by a factor of 10. Deployed on Cortex M4 microcontrollers, the smallest version of LightEQ-NN has an inference time of 932 ms for 1 minute of raw data, an energy consumption of 5.86 mJ, and a flash requirement of 593 kB. Our results show that resource-efficient, on-device machine learning for seismological time series data is feasible and enables new approaches to seismic monitoring and early warning applications.Publicationtype: Conference Paper11