The concept of wireless virtualized networks enables Mobile Virtual Network Operators (MVNOs) to utilize resources made available by multiple Infrastructure Providers (InPs) to set up a service. Nevertheless, existing centralized resource provisioning approaches fail to address such a scenario due to conflicting objectives among InPs and their reluctance to share private information. This paper addresses the problem of resource provisioning from several InPs for services with geographic coverage constraints. When complete information is available, an Integer Linear Program (ILP) formulation is provided, along with a greedy solution. An alternative coalition formation approach is then proposed to build coalitions of InPs that satisfy the constraints imposed by an MVNO, while requiring only limited information sharing. The proposed solution adopts a hedonic game-theoretic approach to coalition formation. For each InP, the decision to join or leave a coalition is made in a decentralized manner, relying on the satisfaction of service requirements and on individual profit. Simulation results demonstrate the applicability and performance of the proposed solution.

Large Language Models (LLMs) capture linguistic structure by operating on sequences of sub-word tokens, yet they often display behaviors that suggest an implicit grasp of high-level concepts.This study probes whether such "ideas" are genuinely encoded in LLM representations and, if so, how faithfully and to what extent.We created a deliberately minimalist LLM (encompassing both tokenizer and transformer architecture)designed to expose internal mechanisms with minimal architectural obscurity.Using a carefully curated toy corpora and probing tasks, we trace how semantically related prompts map onto the model's hidden states.Our findings reveal emergent clustering of conceptually similar inputs even in the stripped-down model.These insights advance our understanding of the representational geometry underpinning modern language models and outline a reproducible framework for future mechanistic studies of semantic abstraction.

With the increase in demand for IoT-based services, the existing infrastructure capabilities of the Fog Providers fell short of satisfying the service requirements, especially for a large geographical area. Therefore, a new concept of fog federation emerged allowing the aggregation of resources from several fog providers available in the area for satisfying the service requirements. The aggregated resources are managed by a third party known as a fog federation manager. It is challenging for the Fog federation manager to decide from whom and how many resources must be acquired. Therefore, in this work, we propose an auction-based resources provisioning approach allowing fog federation to determine the fog providers with their resource provisioning strategies. In addition, a simple example is provided demonstrating the proposed auction-based resources provisioning approach.

In the rapidly evolving field of Machine Learning , selecting the most appropriate model for a given dataset is crucial. Understanding the characteristics of a dataset can significantly influence the outcomes of predictive modeling efforts, making the study of the properties of the dataset an essential component of data science. This study investigates the possibilities of using simulated human data for personalized applications, specifically for testing clustering approaches. In particular, the study focuses on the relationship between dataset characteristics and the selection of the optimal classification model for clusters of datasets. The results of this study provide critical insights for researchers and practitioners in machine learning, emphasizing the importance of dataset characteristics and variability in building and selecting robust models for diverse data conditions. The use of human simulation data provide valuable insights but requires further refinement to capture the full variability of real-world conditions.

Arid and semi-arid regions are characterized by the limited available water. We consider a scenario in which there is one shared water resource among multiple farms as well as a water-sharing policy. It is required to irrigate all or some of these farms currently, but there is not enough water to properly irrigate them. This paper proposes a Multi-farm Irrigation Scheduling System, called MISS. The objective is to save the farms that need urgent irrigation while still satisfying all farm owners according to the water-sharing policy. MISS uses a two-level decision-making method where it considers a priority for the farms that need immediate irrigation as well as a priority for blocks of every farm. MISS is designed based on a realistic scenario where, because of the limited shared water, we sometimes are not able to save all the farms.

The Artificial Intelligence of Things (AIoT), a combination of the Internet of Things (IoT) and Artificial Intelligence (AI), plays an increasingly important role in smart agriculture (SA). AIoT has been adopted in many applications including agriculture, such as crop yield estimation, soil and water conservation, pest and disease detection and supply chain management. While there are plenty of studies on AIoT applications in healthcare, smart cities, manufacturing, and transportation, SA still has a small share of the reported research. This paper presents a comprehensive review of the existing literature in AIoT and Federated Learning (FL) for SA. It identifies current and potential challenges and provides research direction for the future investment in both academia and industry.

Artificial neural networks (ANN) face challenges in the biomedical and health care sectors due to the elastic nature of biomedical data. This data requires a knowledge-centric approach rather than a purely data-centric one. Fuzzy systems efficiently handle the vagueness in medical big data, emulating human perception. These systems provide precise analysis for various medical situations, neutralizing uncertainties like varying disease patterns. They also support ranking populations based on health attributes, aiding in early prognosis and preventive medicine. This special issue is dedicated to focus on the recent advancements and applications of fuzzy systems within the area of healthcare data analysis. It has provided a platform for researchers to share innovative techniques andmethodologiesmore effectively. Through this issue,we aspire to stimulate discussions, foster collaborations and inspire further innovations in leveraging fuzzy systems for more nuanced, human-like interpretations of complex biomedical datasets. As technology evolves, healthcare and diagnostics keeps changing continously. Taking a look at the array of innovative methods, we observe a clear inclination towards deep learning and computational intelligence in diagnostics. For instance, the application of Computational intelligence for analysing CT images for lung cancer detection and the XlmNet, which uses an Extreme Learning Machine Algorithm for classifying lung cancer from histopathological images, both focus on early-stage detection of lung diseases. Their reliance on intricate computational techniques demonstrates a move towards more precise and early diagnostic procedures. On the other hand, we have algorithms like the Residual neural network-assisted one-class classification, specifically tailored for melanoma recognition in imbalanced datasets. It’s evident that there’s a conscious effort to tackle class imbalance issues, which have long been a hurdle in medical image analysis. Mental health and wellbeing are not left behind either. The “Smart Analysis of Anxiety People and Their Activities” and the “Classification Analysis of Burnout People’s Brain Images” both emphasize the growing role of technology in understanding and diagnosing psychological health issues. Similarly, kidney diseases, retinal issues, skin lesions, and other specific conditions are being targeted with specialized models like the Explainable Deep Learning Model for early-stage Chronic Kidney Disease prediction and the modified CNN for retina disease prediction, incorporating the strengths of SVM classifiers. Finally, the integration of ontology-based speculative sense models and hybrid methods like the SVM-ABC for gene expression data classification illustrates a blend of traditional computational methods with modern deep learning, enhancing accuracy and efficiency. We extend our heartfelt appreciation to the Editor-in-Chief of the journal for granting us the opportunity to organise this special issue. We would also like to express our gratitude to the authors and reviewers for their punctual and valuable contributions.We believe that this special issue will provide an additional valuable contribution to the research community.

Graph data is present everywhere and has vast ranging applications from finding the common interests of people to the optimization of road traffic. Due to the interconnectedness of nodes in graphs, training neural networks on graphs can be done in two settings: in transductive learning, the model can have access to the test features in the training phase; in the inductive setting, the test data remains unseen. We explore the differences between inductive and transductive learning on real citation networks when the graphs are converted to undirected graphs. We find that the models achieve better accuracy in the transductive setting than in the inductive setting, but that the gap between validation and test accuracy is also higher, which indicates the models trained in an inductive setting have better generalization capabilities.

Ever since the first isolation of graphene, the semimetal has grown appreciable and has been attracting increasing interest. This interest is reinforced by monolayer graphene's remarkable electronic properties and its usage in revolutionary device developments and applications. However, obtaining monolayer graphene which can be deployed for expansion of experiments in 2D physics comes with its own limitations like high human interventions that requires significant experience, is highly time consuming since it involves repetitive tasks and recognizing graphene crystallites from millions of thicker graphite flakes with other undesired particles is strenuous. Here, we report an approach to detect and discriminate monolayer graphene from other alternating layers of graphene and subsisting substrate impurities. We present a region of interest-based image segmentation process to extricate inapplicable information from the image and extract graphene particles. We, then apply an intensity-based detection model leveraging the characteristic color information to differentiate monolayer graphene from other particles and it is observed that the red color space of the monolayer graphene differs 1.8--6%, green 2.5--8% and blue differ 2.5% to 3% from the surrounding background pixels. We also describe an implementation of our algorithm in a semi-automatic system suitable with our needs.

Small and Medium-sized Enterprises play an essential role in the growth of the global economy. The access to credit for these companies allows them to fund the development of their activities. Artificial Intelligence has emerged as a potential tool to help financial and insurance institutions to assess Small and Medium-sized companies and thus, accelerate their activities. The economic sector in which companies operate is an essential factor when it comes to determining the risk of default. On the other hand, to introduce Artificial Intelligence in a highly regulated industry, the principal actors need to understand the behavior of the models. In this paper, we focus on the development of Artificial Intelligence-based models for different economic sectors Furthermore, we analyze the model behavior using SHapley Additive exPlanations. We compare both the performance and the explanations of the different economic sector models with the global model. Our study shows that there is a slight improvement in terms of performance when creating the different sectorial models. The comparison between the explanations for each model reveals certain disagreements in terms of the most relevant features.

Joint sparse representation (JSR) model has recently emerged as a powerful technique with wide variety of applications. In this paper, the JSR model is extended to error concealment (EC) application, being effective to recover the original image from its corrupted version. This model is based on jointly learning a dictionary pair and two mapping matrices that are trained offline from external training images. Given the trained dictionaries and mappings, the restoration is done by transferring the recovery problem into the sparse representation domain with respect to the trained dictionaries, which is further transformed into a common space using the respective mapping matrices. Then, the reconstructed image is obtained by back projection into the spatial domain. In order to improve the accuracy and stability of the proposed JSR-based EC algorithm and avoid unexpected artifacts, the local and non-local priors are seamlessly integrated into the JSR model. The non-local prior is based on the self-similarity within natural images and helps to find an accurate sparse representation by taking a weighted average of similar areas throughout the image. The local prior is based on learning the local structural regularity of the natural images and helps to regularize the sparse representation, exploiting the strong correlation in the small local areas within the image. Compared with the state-of-the-art EC algorithms, the results show that the proposed method has better reconstruction performance in terms of objective and subjective evaluations.

In this paper, we present a new image sensor architecture for fast and accurate compressive sensing (CS) of natural images. Measurement matrices usually employed in CS CMOS image sensors are recursive pseudo-random binary matrices. We have proved that the restricted isometry property of these matrices is limited by a low sparsity constant. The quality of these matrices is also affected by the non-idealities of pseudo-random number generators (PRNG). To overcome these limitations, we propose a hardware-friendly pseudo-random ternary measurement matrix generated on-chip by means of class III elementary cellular automata (ECA). These ECA present a chaotic behavior that emulates random CS measurement matrices better than other PRNG. We have combined this new architecture with a block-based CS smoothed-projected Landweber reconstruction algorithm. By means of single value decomposition, we have adapted this algorithm to perform fast and precise reconstruction while operating with binary and ternary matrices. Simulations are provided to qualify the approach.

Nowadays, there are still difficulties in the implementation of Compressed Sensing (CS) sensors due to the nature of the measurement matrix. A binary measurement matrix can simplify the CS procedure significantly. However, due to the singularity of this class of measurement matrices, the convergence of some of existing CS reconstruction algorithms, such as the well-known block-based CS with smoothed-projected Landweber reconstruction (BCS-SPL) algorithm, is not guaranteed and can lead to an inaccurate recovery. In this paper we propose a simple, fast and efficient CS recovery algorithm that is able to recover the original image from compressed samples which are obtained using a binary measurement. Singular value decomposition (SVD) is coupled with the BCS-SPL algorithm in order to improve its recovery capability when a binary matrix is employed. The experimental results show that the proposed recovery algorithm has a better performance in terms of reconstruction quality when compared with existing reconstruction algorithm and yields images with quality that matches or exceeds those produced by the BCS-SPL algorithm. Additionally, the proposed algorithm is the most efficient in terms of recovery time, especially at high subrates.

Downsampling based image compression scheme achieves better quality at low bit rates. This paper presents a new scheme in such a paradigm based on adaptive sparse representations with respect to two trained overcomplete dictionaries. The original image is downsampled at the encoder side and an upscaling technique is employed to restore the downsampled image to its original resolution at the decoder side. Due to the downsampling, the high frequency details are removed; therefore, the bit budget of low frequency information is increased, leading to better coding performance at the low bitrates. In order to further improve the coding efficiency, we also propose to encode the residual image as side information. This residual image is obtained by difference between the original image and upscaled image. The low resolution image and the residual image are represented over two dictionaries trained by a bilevel dictionary learning algorithm. Furthermore, the visual salient information is considered into the rate allocation process to improve the rate-distortion performance. The enhanced scheme achieves improvement of the quality at a variety of bitrates at the expense of increasing the system complexity, when compared to the conventional codecs.

Image and video transmission over heterogeneous networks may encounter packet loss due to the channel impairments, leading to quality degradation of the received image. In this paper, a novel robust image transmission system is proposed by casting the error concealment challenge into a sparse recovery framework. To this purpose, a robust encoder is carefully designed in order to mitigate the negative effects of the packet loss. After wavelet decomposition, a quadtree structure of the wavelet coefficients is used to rearrange them into independent partitions. Random linear combinations of coefficients for each partition are then adopted to provide a high error recovery capability. This linear process coupled with a simple packetization process introduces more robustness and error resilience into the transmission system. At the receiver side, the tree-sparse structure of the wavelet coefficients is explicitly exploited in order to model the error recovery problem as a sparse recovery framework. This is achieved by adaptation of a well-known iterative sparse reconstruction algorithm to the defined tree structure built in the wavelet domain. Compared with the state-of-the-art error concealment algorithms, experimental results show that the proposed method has better reconstruction performance in terms of objective and subjective evaluations over a range of packet loss rates, and ensure that a high-quality image can be recovered for the high packet loss scenarios.

Human-machine interaction is becoming a sufficient part of coming future. Being a bright example of HMI, embedded eye-tracking systems allow user to interact with objects by using human's eye-movements. Due to wearable form-factor, developed eye-tracking system has to conform to low-power consumption, low-heat generation and low EM radiation as well as support wireless data transmission and be space efficient. Image capturing, finding of region of interest (ROI), compression of ROI and wireless transmission of compressed data are very beginning steps of the whole algorithm of finding coordinates of pupil. Therefore, area of concentration is to make them as performant as possible from theoretical point of view to further implementation. Combination of hardware powered ROI-finder coupled with well-tuned and optimized image compression system is aimed to speed-up wireless delivery of ROI-images to processing unit. Details of design of such system are presented and discussed.

Sparse representation is a common approach for reducing the spatial redundancy by modelling an image as a linear combination of few atoms taken from an analytic or trained dictionary. This paper introduces a new image codec based on adaptive sparse representations wherein the visual salient information is considered into the rate allocation process. Firstly, the regions of the image that are more conspicuous to the human visual system are extracted using a classical graph-based method. Further, block-based sparse representation over a trained dictionary coupled with an adaptive sparse representation is proposed, such that the adaptivity is achieved by appropriately assigning more atoms of the dictionary to the blocks belonging to the salient regions. Experimental results show that the proposed method outperforms the existing image coding standards, such as JPEG and JPEG2000, which use an analytic dictionary, as well as the state-of-the-art codecs based on trained dictionaries.

This paper introduces a novel image error concealment technique, wherein the correlation among the correctly received pixels is implicitly exploited to recover the missing areas. This correlation is modeled as sparse representations of the correctly received surrounding areas of a lost regions, hence as a linear combination of very few atoms chosen from an over-complete dictionary. Under mild conditions, the sparse representation coefficients of a given zone, including both known and unknown pixels, can be correctly recovered from the sparse representation coefficients of its neighboring area. This linear process coupled with a simple smoothing process introduces a high quality concealed image. Compared with the state-of-the-art error concealment algorithms, experimental results show that the proposed method has better reconstruction performance in terms of objective and subjective evaluations.

This paper propose the use of lapped transforms in a predictive video coding scheme. Our approach involves an inter-frame, block-based prediction based on lapped trans forms and allows rate-distortion optimization, different block sizes and integration with intra-frame prediction. Moreover, this coding scheme strongly reduces the blocking artifacts associated to block-based transforms and permits a better exploitation of the redundancies beyond block borders. The proposed method presents promising results in comparison to the standard implementation of the H.264/AVC, especially for high-definition sequences.

Compressed sensing is applied to multiview image sets and inter-image disparity compensation is incorporated into image reconstruction in order to take advantage of the high degree of inter-image correlation common to multiview scenarios. Instead of recovering images in the set independently from one another, two neighboring images are used to calculate a prediction of a target image, and the difference between the original measurements and the compressed-sensing projection of the prediction is then reconstructed as a residual and added back to the prediction in an iterated fashion. The proposed method shows large gains in performance over straightforward, independent compressed-sensing recovery. Additionally, projection and recovery are block-based to significantly reduce computation time.

In a multiview-imaging setting, image-acquisition costs could be substantially diminished if some of the cameras operate at a reduced quality. Compressed sensing is proposed to effectuate such a reduction in image quality wherein certain images are acquired with random measurements at a reduced sampling rate via projection onto a random basis of lower dimension. To recover such projected images, compressed-sensing recovery incorporating disparity compensation is employed. Based on a recent compressed-sensing recovery algorithm for images that couples an iterative projection-based reconstruction with a smoothing step, the proposed algorithm drives image recovery using the projection-domain residual between the random measurements of the image in question and a disparity-based prediction created from adjacent, high-quality images. Experimental results reveal that the disparity-based reconstruction significantly outperforms direct reconstruction using simply the random measurements of the image alone.

Compressed sensing is applied to multiview image sets and the high degree of correlation between views is exploited to enhance recovery performance over straightforward independent view recovery. This gain in performance is obtained by recovering the difference between a set of acquired measurements and the projection of a prediction of the signal they represent. The recovered difference is then added back to the prediction, and the prediction and recovery procedure is repeated in an iterated fashion for each of the views in the multiview image set. The recovered multiview image set is then used as an initialization to repeat the entire process again to form a multistage refinement. Experimental results reveal substantial performance gains from the multistage reconstruction.

An approach to scalable multi-view video coding with joint best basis wavelet packets is examined in this paper. A 4-D wavelet transform is used to decorrelate the multi-view video data temporally, view-directionally, and spatially for efficient scalable compression. Motion compensated temporal filtering (MCTF) is used for temporal, and disparity compensated view filtering (DCVF) for view- directional decomposition. Adaptive wavelet packets as a generalized wavelet decomposition are presented for spatial decomposition. Two algorithms to find the best basis wavelet packets are evaluated and compared with classical dyadic wavelet transform: a low complexity entropy based best basis search and a search algorithm in a rate-distortion framework. In both cases, the joint best basis is determined for a group of frames rather than for each frame individually. Therefore, the rate to spend for the tree description is minimal while advantage is taken of the similarity of frames within a temporal-view-directional subband.

The geometric features of images, such as edges, are difficult to represent. When a redundant transform is used for their extraction, the compression challenge is even more difficult. In this paper we present a new rate-distortion optimization algorithm based on graph theory that can encode efficiently the coefficients of a critically sampled, non-orthogonal or even redundant transform, like the contourlet decomposition. The basic idea is to construct a specialized graph such that its minimum cut minimizes the energy functional. We propose to apply this technique for rate-distortion Lagrangian optimization in subband image coding. The method yields good compression results compared to the state-of-art JPEG2000 codec, as well as a general improvement in visual quality.

Three-dimensional (t+2D) wavelet coding schemes have been demonstrated to be efficient techniques for video compression applications. However, the separable wavelet transform used for removing the spatial redundancy allows a limited representation of the 2D texture because of spatial isotropy of the wavelet basis functions. In this case, anisotropic transforms, such as fully separable wavelet transforms (FSWT), can represent a solution for spatial decorrelation. FSWT inherits the separability, the computational simplicity and the filter bank characteristics of the standard 2D wavelet transform, but it improves the representation of directional textures, as the ones which can be found in temporal detail frames of t + 2D decompositions. The extension of both classical wavelet and wavelet-packet transforms to fully separable decompositions preserve at the same time the low-complexity and best-bases selection algorithms of these ones. We apply these transforms in t + 2D video coding schemes and compare them with classical decompositions.

A joint source-channel coding scheme for transmission of video over flat Rayleigh fading channels is described. The coding scheme consists of a spatiotemporal motion-compensated wavelet decomposition, a vector quantization of the coefficients through maximum-diversity lattices, and a linear labeling which minimizes simultaneously the source and channel distortion. Modulation diversity via rotated constellations produces the maximum-diversity lattices which increase robustness to channel fading without the addition of redundancy. Experimental results compare the proposed system to a prominent scalable video coder protected by more traditional convolutional codes, and superior performance is observed for high levels of channel noise.

3D video codecs have attracted recently a lot of attention, due to their compression performance comparable with that of state-of-art hybrid codecs and due to their scalability features. In this work, we propose a least mean square (LMS) based adaptive prediction for the temporal prediction step in lifting implementation. This approach improves the overall quality of the coded video, by reducing both the blocking and ghosting artefacts. Experimental results show that the video quality as well as PSNR values are greatly improved with the proposed adaptive method, especially for video sequences with large contrast between the moving objects and the background and for sequences with illumination variations

Three-dimensional (3D) wavelet coding schemes compression applications. The scalability property of such schemes is one of the most important issues for video surveillance systems. In this paper we introduce a new lifting-based method of temporal decomposition which provides a scalability factor of 5 in a motion-compensated subband video coding scheme. Depending on the sequence characteristics, motion model etc., this structure can provide high coding performance. Also, it gives a better leading to an improved temporal scalability. It addresses video surveillance applications, where the motion is very low in most cases

A scalable video coder consisting of motion-compensated temporal filtering coupled with structured vector quantization plus a linear mapping of quantizer indexes that minimizes simultaneously source and channel distortions is presented. The linear index assignment takes the form of either a direct, uncoded mapping or a coded mapping via Reed-Muller codes. Experimental results compare the proposed system to a similar scheme using unstructured vector quantization as well as to a prominent scalable video coder protected by more traditional convolutional codes. The proposed system consistently outperforms the other two schemes by a significant margin for very noisy channel conditions.