sensors-logo

Journal Browser

Journal Browser

Recent Developments and Challenges in Artificial Intelligence and Deep Learning in Advanced Sensing Systems

A special issue of Sensors (ISSN 1424-8220). This special issue belongs to the section "Communications".

Deadline for manuscript submissions: 25 January 2025 | Viewed by 759

Special Issue Editors

Special Issue Information

Dear Colleagues,

Sensing technologies can benefit sensing processes in many research fields and applications. Significant advances in artificial intelligence (AI) have given rise to advanced intelligent sensor technologies and sensing systems. The emergence of high-end AI sensor technology is a cornerstone of the new era of technological advancements and transformation in satellites, military applications, artificial intelligent Internet of Things, neuromorphic computing systems, and the next generation of robots and Industry 5.0. These sensing systems have advanced capabilities as they can use multiple sensors capable of detecting and processing multidimensional information. Moreover, low-end AI sensor technology can also provide cost-effective solutions for a wide range of applications such as smart buildings, smart homes, and individual healthcare.

Furthermore, recent advances in deep learning techniques have also led to significant progress in sensing systems. Due to improvements in computation power and speed and graphics processing units (GPUs), the design and development of new deep learning algorithms has been accelerated. Several fundamental deep learning frameworks, such as convolutional neural networks (CNNs), deep neural networks (DNNs), recurrent neural networks (RNNs), and transformer architectures, have been successfully applied in the fields of image processing, natural language processing, automatic speech recognition, and healthcare informatics. They have also contributed to great achievements in image, speech, and text modalities and have continuously emerged in different architectures to provide better accuracy for targeted tasks.

This Special Issue focuses on AI sensors, advanced AI sensing systems, and new or improved deep learning architectures or techniques to address a broad range of sensor- and sensing-related issues in intelligent systems. Contributions addressing the design and development of novel deep learning frameworks, specifically focusing on their effectiveness, intelligence, and reliability in achieving superior performance, are also encouraged. This Special Issue is dedicated to theoretical innovations, research, and developments for real-world applications. The potential topics of this Special Issue include, but are not limited to, the following:

  • Emerging AI sensors or advanced sensing systems;
  • New deep learning techniques and applications;
  • AI sensing technology for smart cities, health monitoring and medical applications, industrial sensing, and urban environmental monitoring and sensing;
  • AI sensing technology in marine sensing, oceans, coastal zones, and inland waters;
  • Wearable AI sensors and self-powered sensor systems;
  • AI and deep learning techniques in sensing, smart cities, and remote sensing applications;
  • AI Internet of Things and Internet of Vehicles for sensing and smart cities;
  • AI and situational awareness for natural disaster monitoring;
  • AI sensors or sensing for robots and autonomous systems;
  • Big data processing using deep learning and sensing;
  • New deep learning and image processing techniques and applications;
  • AI and deep learning in embedded systems.

Prof. Dr. Kah Phooi Seng
Prof. Dr. Li-Minn Ang
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

  • AI sensors
  • intelligent sensing
  • advanced sensing
  • deep learming
  • AI internet of things
  • AI wearables

Published Papers (1 paper)

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

Research

21 pages, 2219 KiB  
Article
Deep Learning Based Over-the-Air Training of Wireless Communication Systems without Feedback
by Christopher P. Davey, Ismail Shakeel, Ravinesh C. Deo and Sancho Salcedo-Sanz
Sensors 2024, 24(10), 2993; https://doi.org/10.3390/s24102993 - 8 May 2024
Viewed by 526
Abstract
In trainable wireless communications systems, the use of deep learning for over-the-air training aims to address the discontinuity in backpropagation learning caused by the channel environment. The primary methods supporting this learning procedure either directly approximate the backpropagation gradients using techniques derived from [...] Read more.
In trainable wireless communications systems, the use of deep learning for over-the-air training aims to address the discontinuity in backpropagation learning caused by the channel environment. The primary methods supporting this learning procedure either directly approximate the backpropagation gradients using techniques derived from reinforcement learning, or explicitly model the channel environment by training a generative channel model. In both cases, over-the-air training of transmitter and receiver requires a feedback channel to sound the channel environment and obtain measurements of the learning objective. The use of continuous feedback not only demands extra system resources but also makes the training process more susceptible to adversarial attacks. Conversely, opting for a feedback-free approach to train the models over the forward link, exclusively on the receiver side, could pose challenges to reliably end the training process without intermittent testing over the actual channel environment. In this article, we propose a novel method for the over-the-air training of wireless communication systems that does not require a feedback channel to train the transmitter and receiver. Random samples are transmitted through the channel environment to train a mixture density network to approximate the channel distribution on the receiver side of the network. The transmitter and receiver models are trained with the resulting channel model, and the transmitter can be deployed after training. We show that the block error rate measurements obtained with the simulated channel are suitable for monitoring as a stopping criterion during the training process. The resulting method is demonstrated to have equivalent performance to the end-to-end autoencoder training on small message sequences. Full article
Show Figures

Figure 1

Back to TopTop