Tiny Machine Learning-Based Time Series Processing
A special issue of Sensors (ISSN 1424-8220). This special issue belongs to the section "Communications".
Deadline for manuscript submissions: 10 December 2024 | Viewed by 74
Special Issue Editors
Interests: infomobility systems; human–computerinteraction; cyber-physical system engineering; artificial intelligence; technology-enhanced learning
Special Issues, Collections and Topics in MDPI journals
Interests: embedded systems; sensors and sensory systems; hardware accelerators; embedded machine learning and neural networks; deep learning; object detection
Special Issue Information
Dear Colleagues,
Time series data, which comprise sequences of observations collected over time by various types of sensors, hold a huge value across several application domains.
Machine learning extracts insights for precise predictions and informed decisions in diverse fields.
Bringing intelligence close to the information source is key to achieving the benefits of edge computing, in terms of latency, bandwidth, and energy. Advancements in TinyML technologies have enabled the execution of powerful deep learning models also on extremely constrained devices, opening unprecedented perspectives for application field deployment.
In this Special Issue, we aim to investigate the latest developments in the area of time series processing based on TinyML. Topics of interest include, but are not limited to, the following:
- Enhancing sensors with TinyML;
- Energy-efficient circuits and system architectures for time series TinyML;
- Real-time time series applications on resource-limited devices;
- Software/hardware co-design for efficient low-power embedded systems;
- TinyML-based processing for time series forecasting, classification, anomaly detection;
- Performance and system assessment in TinyML time series processing for field deployment;
- Energy harvesting and power management in embedded time series processing;
- Quantization/compression methods for efficient embedded deployment;
- Binary models for time series processing;
- Neural architecture search (NAS) methods for embedded time series processing;
- Explainability of time series
Dr. Francesco Bellotti
Dr. Ali Dabbous
Dr. Paolo Pasini
Guest Editors
Manuscript Submission Information
Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.
Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Sensors is an international peer-reviewed open access semimonthly journal published by MDPI.
Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.
Keywords
- timeseries processing
- embedded electronic systems
- embedded computing
- hardware/software co-design
- embedded machine learning and neural networks
- efficient hardware implementation
