Intelligent IoT Systems

A special issue of Electronics (ISSN 2079-9292). This special issue belongs to the section "Computer Science & Engineering".

Deadline for manuscript submissions: closed (30 April 2022) | Viewed by 3838
The review of a manuscript by experts starts right after the submission and the review outcome will be communicated to the authors as soon as possible. The publication process starts immediately for those papers that are accepted.

Special Issue Editors


E-Mail Website
Guest Editor
Department of Mathematics and Computer Science, Eindhoven University of Technology, Eindhoven, The Netherlands
Interests: lifecycles of resource-constrained embedded networks; architectures for resource management; QoS management and data analytics; intelligent environments (smart spaces); Internet of Things (IoT); IoT intelligence
Special Issues, Collections and Topics in MDPI journals

E-Mail Website
Guest Editor
Department of Computer Science and Engineering, University of Nebraska-Lincoln, Lincoln, NE, USA
Interests: wireless underground communications; agricultural internet of things; dynamic spectrum access in 5G networks; wearable embedded systems; connected autonomous systems; cyber-physical networking

Special Issue Information

Dear Colleagues,

Internet of Things (IoT) systems are typically resource-constrained, networked, embedded systems that extend the reach of IP communication into the physical world, to billions of resource-constrained endpoints that are called ‘Things’. Things organize themselves into Internet-enabled networks and cooperate to realize a complex task. On top of their own resource limitations, Things are connected into resource-constrained access networks, with low power, lossy, low bitrate asymmetric links and limited group communication primitives.

The rise of the IoT as a technology is also timely and complementary to the realization of ubiquitous smart interaction of humans with networked embedded systems in “intelligent environments”, also known as “smart spaces”, which can do complex tasks for humans, e.g., as in intelligent lighting, smart homes, smart cities, smart buildings, smart healthcare, intelligent transportation systems, and Industry 4.0. Existing implementations of such intelligent IoT systems, however, are mainly limited to the scope of specific application silos, and technology experts play a significant part in the whole process. IoT networks are complex by nature due to distributed services on many heterogeneous devices collaboratively fulfilling common goals defined by applications. Therefore, it is crucial for society that (intelligent) IoT systems and the processes on them be well-understood, easily controlled, and optimized. Unfortunately, most of the current guidelines, architecture designs, protocols, and algorithms for IoT systems are limited in scope.

This must be addressed by building up theory and knowledge on IoT lifecycles (i.e., the series of stages IoT devices, services, and applications go through, from inception to decline) and by testing them in real settings. The objective of this Special Issue is to bring visibility to ongoing studies in the development of IoT systems (lifecycle management, architectures, protocols) and IoT intelligence (algorithms specifically tailored to and deployable on IoT systems). Researchers who are active in this field are invited to submit their manuscripts to this Special Issue.

Dr. Tanir Ozcelebi
Prof. Dr. Mehmet Can Vuran
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. Electronics 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 2400 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

  • Internet of Things
  • Intelligent IoT systems
  • Lifecycles of IoT devices, services, and applications
  • Distributed data analytics
  • Analysis and engineering of IoT systems
  • Low-power (IP) networking
  • Networked embedded systems
  • Security and privacy
  • System architecture
  • System design
  • System performance and dependability
  • Programming and service composition
  • Steering IoT application behavior
  • Quality aspects
  • Edge AI

Published Papers (1 paper)

Order results
Result details
Select all
Export citation of selected articles as:

Research

25 pages, 1091 KiB  
Article
Integer-Only CNNs with 4 Bit Weights and Bit-Shift Quantization Scales at Full-Precision Accuracy
by Maarten Vandersteegen, Kristof Van Beeck and Toon Goedemé
Electronics 2021, 10(22), 2823; https://doi.org/10.3390/electronics10222823 - 17 Nov 2021
Cited by 5 | Viewed by 3471
Abstract
Quantization of neural networks has been one of the most popular techniques to compress models for embedded (IoT) hardware platforms with highly constrained latency, storage, memory-bandwidth, and energy specifications. Limiting the number of bits per weight and activation has been the main focus [...] Read more.
Quantization of neural networks has been one of the most popular techniques to compress models for embedded (IoT) hardware platforms with highly constrained latency, storage, memory-bandwidth, and energy specifications. Limiting the number of bits per weight and activation has been the main focus in the literature. To avoid major degradation of accuracy, common quantization methods introduce additional scale factors to adapt the quantized values to the diverse data ranges, present in full-precision (floating-point) neural networks. These scales are usually kept in high precision, requiring the target compute engine to support a few high-precision multiplications, which is not desirable due to the larger hardware cost. Little effort has yet been invested in trying to avoid high-precision multipliers altogether, especially in combination with 4 bit weights. This work proposes a new quantization scheme, based on power-of-two quantization scales, that works on-par compared to uniform per-channel quantization with full-precision 32 bit quantization scales when using only 4 bit weights. This is done through the addition of a low-precision lookup-table that translates stored 4 bit weights into nonuniformly distributed 8 bit weights for internal computation. All our quantized ImageNet CNNs achieved or even exceeded the Top-1 accuracy of their full-precision counterparts, with ResNet18 exceeding its full-precision model by 0.35%. Our MobileNetV2 model achieved state-of-the-art performance with only a slight drop in accuracy of 0.51%. Full article
(This article belongs to the Special Issue Intelligent IoT Systems)
Show Figures

Figure 1

Back to TopTop