Next-Generation Machine Learning Models in Signal Processing Applications
A program that can identify patterns or draw conclusions from a dataset that has never been seen before is called a machine learning model. Machine learning models, for instance, can analyse and accurately identify the meaning of previously unsaid sentences or word combinations in natural language processing. Unsupervised and supervised learning are frequently combined in deep learning. Thus, similar to semi-supervised learning, deep learning algorithms use both unlabelled and annotated data instances. As a regularize and optimization tool, unsupervised learning facilitates training. Automating repetitive activities is another important application of machine learning for signal processing. Machine learning models can be trained to recognize and adjust to patterns in the data, eliminating the need to manually create and apply signal processing methods for particular jobs. Supervised, unsupervised, semi-supervised, self-supervised, and reinforcement learning are the five main categories into which machine learning algorithms can be divided. One way to characterize the field of engineering science that studies the relative motion of a machine's many parts and the forces acting on them is as the Theory of Machines.
Businesses can benefit from machine learning in a number of ways, including data analysis, transcription and translation, automated customer service via chatbots, fraud detection, security threat identification, personalization, and recommendations. Computers can now see, comprehend, and translate spoken and written language, analyse data, make recommendations, and do a wide range of other sophisticated tasks thanks to a collection of technologies known as artificial intelligence (AI). Generally speaking, generative models are able to convert one image to another while improving or changing the input image's style. Large datasets of photos and videos, as well as text, are used to train GenAI models in order to identify patterns and correlations among pixels. The discipline of creating effective algorithms and models that may identify and reveal a potential hidden structure in signals, hence retrieving desired information, is known as machine learning for signal processing (MLSP). Granular information like source and destination, service, protocol, and port numbers can be obtained by using machine learning to evaluate traffic flows from endpoint groups. Training computers to become more proficient at tasks without explicit programming is the aim of machine learning. In order to do this, a number of actions must be taken. First, information must be gathered and prepared. Next, an algorithm, also known as a training model, must be chosen. Adaptive, individually designed customer experiences, such customized promotions, can be produced by machine learning algorithms.
Generative AI is the term for unsupervised and semi-supervised machine learning methods that let computers create new content from preexisting content, such as text, audio and video files, photos, and even code. The primary goal is to create entirely unique artifacts that resemble the authentic thing. Finding patterns in user data and using these complex patterns to inform predictions is the main goal of machine learning, which is used to address business issues and provide answers to queries. Machine learning facilitates both data analysis and trend identification. Because it necessitates a thorough understanding of computer science and mathematics, machine learning can be challenging to acquire. A wide range of contributions is encouraged, embracing various disciplines and viewpoints without restriction : Next-Generation Machine Learning Models in Signal Processing Applications.
Potential topics include but are not limited to the following:
-
Deep Learning Frameworks for Complex Signal Processing Activities.
-
Processing signals methods for explainable artificial intelligence.
-
Application of Federate Learning in Automated Signal Processing.
-
Reinforcement Learning in Networks of Adaptive Signal Processors.
-
Transformer applications in communication and signal processing systems.
-
Self-Supervised Learning Models for Enhancement and Denoising Signals.
-
Processing signals using multi-modal predictive techniques.
-
Connected Signal Processing and Analysis using Graph Neural Networks.
-
Processing signals scenarios including few-shot and zero-shot learning.
-
Challenges of Solidity and Privacy in Adversarial Machine Learning Strategies in Signal Processing.
-
Energy-Saving Machine Learning Models for Applications in Real-Time Signal Processing.
-
Bioinspired Artificial intelligence Methods for Identifying Signal Patterns.
-
Edge AI in Signal Processing: Facilitating Applications with Low Latency.
-
Methods of Meta-Learning for Signal Processing in Changing Conditions.
-
Signal Processing with Quantum Machine Learning Models: Prospects and Difficulties.
Important Dates:
Paper Submission Deadline - 10.10.2025
Author Notification - 20.01.2026
Revised Papers Submission - 10.04.2026
Guest Editor Details:
Dr. Makhlouf Derdour
Full-Professor,
Computer Science Department,
University of Oum el Bouaghi, Algeria.
E-mail: derdour.makhlouf@univ-oeb.dz, makhloufderdour@hotmail.com
Google Scholar Link: https://scholar.google.com/citations?user=CRkhvzgAAAAJ&hl=en
IEEE Link: https://ieeexplore.ieee.org/abstract/document/10614597
Orcid Link: https://orcid.org/0000-0001-6622-4355
Scopus Link: https://www.scopus.com/authid/detail.uri?authorId=36550031000
Biography: Makhlouf Derdour received his Engineering degree in computer sciences from the University of Constantine, Algeria, in 2004, his Magister degree in computer sciences from the University of Tebessa, and his PhD degree in computer networks from the University of Pau and Pays de l’Adour (UPPA), France, in 2012. He is currently a full professor at the Computer Science Department of the University of Oum el Bouaghi, Algeria. His research interests include software architecture, multimedia applications, adaptation and self-adaptation of applications, design and modelling of systems, and systems security. He is the general chair of the International Conference on Pattern Recognition and Intelligent Systems (PAIS). Currently, he is a director of Artificial Intelligence and Autonomous Things Laboratory.
Prof. Gregorio Díaz-Descalzo
Associate-Professor,
Department of Computer Science,
Universidad de Castilla-La Mancha,
Albacete, 02071, Castilla-La Mancha, Spain
E-mail: gregorio.diaz@uclm.es
Google Scholar Link: https://scholar.google.com/citations?user=9QICiOAAAAAJ&hl=en
IEEE Link: https://ieeexplore.ieee.org/author/37297156200
Orcid Link: https://orcid.org/0000-0002-9116-9535
Scopus Link: https://www.scopus.com/authid/detail.uri?authorId=57541643100
Biography: Gregorio Díaz is an associated Professor at the University of Castilla-La Mancha within the ReTiCS research group with tenure distinction (2011), published more than 26 journal papers, from which the JCR index indexes 23, participated in 48 international and national conferences, main researcher of 4 ERDF projects. His research aims to make software more reliable, secure, and easier to design. He has supervised over 29 master theses, including 4 in research areas and 4 PhD theses. He has taught in undergraduate and postgraduate studies and was awarded the quality award Euro-Inf Bachelor by EQANIE.
Dr. Philippe Roose
Full-Professor,
Université de Pau et des Pays de l’Adour,
64600 Bayonne, France
E-mail: philippe.roose@iutbayonne.univ-pau.fr
Google Scholar Link: https://scholar.google.fr/citations?user=sILWKAMAAAAJ&hl=fr
IEEE Link: https://ieeexplore.ieee.org/author/37072567700
Orcid Link: https://orcid.org/0000-0002-2227-3283
Scopus Link: https://www.scopus.com/authid/detail.uri?authorId=55843683500
Biography: Philippe Roose is Full Professor at LIUPPA research lab – E2S, University of Pau et des Pays de l’Adour (France). He is currently head of a Bachelor in Advances Computing. He is the co-inventor of both KalimuchoTM & PISCO platforms (patented x4). He supervised 20 PhD theses (2 current) and was involved in over 40 PhD defences. He published many national and international articles, journals and books and organized several national and international conferences. His works mainly focus on green computing, middleware, software architecture, context-awareness, and autonomy. He strongly believes in multi and plural-disciplinary research with other computer scientists from the semantic web, interactions and interfaces, HPC, design methods, etc., but also with other scientists from other disciplines: historians, anthropologists, geographers, etc.