Evaluating wildfire risk systems using standard machine-learning metrics such as F1-score or IoU is fundamentally flawed: these metrics assess event prediction accuracy, not the operational coherence of a continuous risk signal. This work proposes a novel monotonic evaluation framework that measures whether increases in a predicted risk score consistently correspond to increases in observed operational load, such as number of fires, intervention time, and deployed resources. Moreover, we compare three structurally different approaches on the French Alpes-Maritimes department: the expert-based DFE index, GRU- based predictive models, and FARS, a hybrid multi-agent system combining predictive AI with LLM-based reasoning. Experimental results reveal that the DFE, despite poor classification metrics, exhibits the most balanced monotonic behavior across the full risk scale. GRU models achieve strong local monotonicity but fail to produce well-distributed risk levels. FARS inherits and reveals the structural limitations of upstream signals rather than correcting them. The central finding is a paradigm shift: a good risk model does not predict fires accurately, but one whose ordinal scale meaningfully explains operational dynamics, as proved in this paper. Code of the monotonic framework is available on github.

Forest-fire prediction using Artificial Intelligence (AI) continues to face major challenges, including (i) the ability to generalize across regions with very different risk profiles, (ii) managing the inherent daily variability and randomness of fire occurrences (including extreme fire days). These factors together have hindered the deployment of dependable prediction systems in operational settings. In this work, we introduce a novel multi-risk modeling framework specifically designed to tackle all two challenges simultaneously. The proposed approach is applied to daily forest-fire prediction across mainland France. We develop a voting-based system that combines the outputs of multiple models trained on signals smoothed with a range of convolutional kernels, capturing both local and seasonal variations. The proposed solution achieves superior performance compared to conventional models, demonstrating improved cross-regional transferability and robustness to daily fluctuations. Notably, it significantly enhances prediction skill for the rare but damaging extreme-fire days, where traditional models often fail. Our experiments reveal that using an ensemble of multiple risk models can better capture the complex dynamics of fire risk and provide more reliable guidance for decision-makers. Supplementary materials are available here.

Recently, detecting dust and damage on solar panels has become feasible with deep learning models. However, existing solutions often struggle with either model performance or practical implementation challenges. To address these limitations, this paper presents a novel lightweight and robust framework for dust detection on solar panels, leveraging the DINOv2 transformer model enhanced with a multi-head self-attention mechanism. The proposed approach integrates advanced data augmentation and fine-tuning strategies to optimize performance, achieving state-of-the-art results with an accuracy of 94.1% and an F1-score of 93.2%, significantly outperforming traditional CNN-based methodologies. Additionally, the trained model is lightweight, requiring fewer parameters, which reduces memory, computation, and resource demands, leading to lower inference time. The success of this approach highlights the potential of transformer architectures in computer vision tasks, particularly for capturing long-range dependencies and contextual relationships. By striking a balance between model efficiency and real-world application constraints, the proposed solution offers a practical and scalable approach for automated dust detection, enabling proactive maintenance and optimizing the performance of solar power installations.

The increasing demand for real-time, resource-efficient deep learning models has driven research into frugal object detection techniques, enabling deployment in edge computing, IoT, and mobile applications. This paper comprehensively reviews lightweight object detection architectures, including CNN-based, transformer-based, and hybrid models. It explores optimization techniques such as knowledge distillation, quantization, pruning, and neural architecture search (NAS). We analyze trade-offs between accuracy, computational efficiency, and power consumption and providing insights into their impact on various edge AI hardware platforms, including Jetson Nano, Google Edge TPU, and FPGA-based deployments. A comparative evaluation highlights the strengths and limitations of different frugal optimization strategies, emphasizing hybrid approaches that balance detection performance and efficiency. We also discuss challenges such as cross-hardware generalization, real-time adaptability, adversarial robustness, and explainability. Finally, we outline promising future research directions, including advancements in energy-efficient vision transformers, on-device continual learning, adaptive compression techniques, and sustainable AI solutions. This study serves as a roadmap for researchers and practitioners developing efficient object detection models for resource-constrained environments.

The current Field-Programmable Gate Array (FPGA) implementation of cryptographic algorithms faces performance and security challenges because these algorithms were not originally designed to take FPGA features into account. One significant performance limitation arises from the iteration of a round function for a high round number, given the fixed structures like static substitution and diffusion primitives throughout the process. This paper introduces a new framework for a key-dependent, flexible one-round stream cipher scheme specifically designed to benefit from FPGA features. It is called RFCA. Security and performance analyses validate the effectiveness and robustness of the proposed solution, ensuring the desired cryptographic properties. In comparison with an AES implementation, RFCA is 34 times faster.

In modern society, power consumption plays a crucial role, since it is capable of influencing multiple sectors including residential, commercial, and industrial domains. It covers the electrical energy amount used by various devices, appliances, machinery, and systems within a specific time frame. The accurate prediction of power consumption is imperative for effective energy management, resource allocation, infrastructure planning, and cost optimization. In this study, we focused on predicting power consumption within DAEWOO Steel CO. Ltd, located in South Korea. We acquired, preprocessed, and analyzed a dataset containing information about the daily operations of the industrial facility, with data being sampled at 15-minute intervals. By leveraging machine learning techniques, we employed six tree-based algorithms and three ensemble learners to forecast the target variable. Our comparative analysis examined the performance of these regressors across three forecasting horizons: 15 minutes, 1 hour, and 1 day. We discovered that the efficacy of the regressors is complex and is linked to the forecasting horizon. Notably, the stacking ensemble learner outperformed others for the 15-minute horizon, achieving impressive metrics of 98.5% for D2, 99.9% for R2, 0.81 for MSE, and 0.37 for MAE. For 1-hour ahead predictions, XGBoost emerged as the most accurate model, attaining metrics of 96.3% for D2, 99.7% for R2, 53.07 for MSE, and 3.5 for MAE. Finally, for 1-day ahead forecasting, the Extra Tree regressor surpassed its counterparts, achieving metrics of 93.1% for D2, 99.3% for R2, 13944 for MSE, and 74.46 for MAE. These findings underscore the importance of tailoring predictive models to specific forecasting horizons and highlight the efficacy of ensemble learning techniques in enhancing power consumption predictions across varying time frames.

With the advent of the Multimedia Internet of Things (MIoT), many image compression techniques have been proposed to address the network’s considerable challenges related to performance and security. However, many MIoT devices, such as the nRF52832 SoC with 64Kb RAM or even less, have significant memory constraints, making conventional methods unsuitable. MIoT networks face considerable challenges related to performance and security due to limitations in the power, computation, and memory of MIoT devices. These limitations result in difficulties in handling high image volumes. Multimedia compression is a potential solution to reduce data size. As MIoT devices often rely on wireless connections, they are also vulnerable to diverse security attacks (passive and active). This work introduces a secure and efficient image crypto-compression technique dedicated to devices having limited memory. It also proposes using denoising and a super-resolution deep learning model to reduce the overhead of the compression process and a lightweight cipher scheme that requires a single round of simple operations to reduce the overhead of the encryption process. The proposed approach effectively addresses the mentioned challenges with minimal overhead on the MIoT device, especially in terms of computational and communication delays, and extensive experimentation underscores its suitability in both effectiveness and robustness.

Solar panels, the primary components of solar photovoltaic systems, play a pivotal role in converting sunlight into electricity. However, the efficiency and performance of solar panels can be significantly influenced by environmental factors, notably the accumulation of dust and debris on their surfaces. This paper focuses on the investigation of deep learning image classification techniques to detect dust periodically, utilizing solar panel images collected by drones or robots. This approach aims to reduce the impact of dust on solar panels and help identify effective cleaning methods for each case. This work proposes the development of a deep learning binary image classifier model specifically designed to differentiate between “dusty” and “clean” solar panels. The proposed system is based on pre-trained deep learning models fine-tuned for dusty solar panel detection. The results demonstrate that fine-tuning the weights of the pre-trained model enhances performance, with the EfficientNetB7 model yielding the best outcome.

This work delves into the significant impact of Machine Learning (ML) on the advancement and improvement of Energy Management Systems (EMS), focusing on the incorporation of renewable energy sources, smart grids, and the general enhancement of energy efficiency, reliability, and sustainability. The main aim of this work is to offer a detailed summary of Machine Learning technologies that can be used in modern energy systems. It explains how these technologies can improve certain tasks like load forecasting, energy optimization, predictive maintenance, fault detection and diagnosis, and incorporating renewable energy systems supported by relevant approaches and application areas. Moreover, the work examines the benefits and opportunities presented by machine learning in boosting efficiency, enhancing system stability and resilience, and contributing to environmental sustainability. In addition, it identifies challenges and outlines future research needed to facilitate the adoption of ML in energy systems. In conclusion, the study underlines the critical role of machine learning in the evolution of energy systems and underscores the importance of collaborative efforts to overcome existing challenges and fully leverage machine learning’s potential in the smart energy management systems domain.

Federated Learning (FL) is widely recognized as one of the most effective collaborative learning methodologies for training various local models using private datasets. However, model inversion or membership inference may be used by an eavesdropper or by the central server, which is frequently a reliable but curious entity, to know details about local client datasets. To prevent these attacks during communication or on the server side, several security solutions are presented. However, these existing solutions suffer from high overhead in terms of computation in addition to storage and communication overhead, especially for real-time applications or for limited client devices. Therefore, in this work, a secure scheme that consists of two layers of the Symmetric Encryption Algorithm (SCA) is proposed. The first layer uses a novel lightweight additive homomorphic cipher scheme that is adapted and modified, while the second layer uses a traditional SCA. This work focuses on the effect of employing symmetric HE in an FL model instead of using an asymmetric approach, which is the base of existing HE cipher schemes. Experimental results indicate a significant gain in terms of computation, resources, storage, communication, delay overhead, and immunity against a wide variety of attacks, especially with the usage of a dynamic key approach with variable cryptographic primitives per block or iteration.

Recently, several security schemes for Random Linear Network Coding (RLNC) have been proposed to increase the immunity of the RLNC technology against security attacks. One of the presented security schemes aims at securing the Global Encoding Vectors (GEV) that use other vectors to maintain the proper encoding process of RLNC at intermediate nodes. However, this approach introduces overhead in terms of computational complexity (block cipher with multiple rounds and operations) and communication (2xn elements for each packet instead of n). To that end, this paper proposes a new scheme, $S E R S$, that overcomes the disadvantages and limitations of the existing security schemes by relying on a single GEV instead of two, which is the case of the original RLNC. The proposed scheme reduces the required computational complexity by eliminating AES encryption and by keeping the source RLNC encoding step as a secret. SERS is based on a dynamic key structure, and the introduced modifications result in a modern lightweight, and secure RLNC while achieving higher efficiency and minimizing the space of vulnerabilities. SERS exhibits minimal computational complexity and communication overhead, and it ensures message confidentiality and availability, in addition to source authentication when a homomorphic keyed hash function is employed. A second variant of the proposed scheme is also presented. The main advantages of the proposed scheme are that 1) it operates at the sub-generation level, 2) it can be implemented in parallel, and 3) it increases the security level by using different RLNC encoding matrices instead of just one.

Cybersecurity-enabled smart renewable energy system security requires a multifaceted strategy that includes proactive risk assessment, strong defenses, stakeholder cooperation, and the combination of cutting-edge technology such as Artificial Intelligence (AI) and blockchain. To safeguard vital infrastructure and guarantee a continuous supply of energy, defense-in-depth strategies such as encryption protocols, and access controls can be put into place in conjunction with cybersecurity awareness campaigns and regular security audits. In the face of changing cyber threats, these initiatives are crucial to protect data privacy and system resilience for the development of sustainable energy. This study aims to provide important insights into current practices, identify areas for improvement, and provide insight for further studies and regulations intended to improve the energy infrastructure’s security and resilience in the digital era.

Recent IoT emerging systems and applications require processing over encrypted data to preserve confidentiality. However, designing efficient Homomorphic Encryption (HE) algorithm to respond better to real-time requirements of IoT applications and tiny IoT devices is primordial. Unfortunately, existing homomorphic asymmetric schemes do not provide the efficiency implementation. While, on the other hand, homomorphic symmetric approaches suffer from a low level of security. In this paper, an efficient flexible additive homomorphic symmetric cipher scheme is proposed and it is called “ACiS”. Moreover, ACiS is based on the dynamic key approach to reach a high level of security, where different cryptographic primitives are used for different sessions. In addition, it can achieve a good performance and robustness against confidentiality attacks. It is based on a simple round function that should be iterated for nr iterations and$nr > 1$. Furthermore, the proposed solution is analyzed and evaluated in terms of security and performance levels. Experimental results prove that the proposed solution reaches a good balance between performance and security level compared to the Paillier cipher scheme (a well-known asymmetric additive HE scheme).

Due to the volume of multimedia sensed data, a network of Multimedia Internet of Things (MIoT) devices faces various challenging constraints, most notably in terms of communication overhead, power consumption, and memory usage. A set of these MIoT devices is unable to overcome the large data-size challenge via the use of the Lossy Multimedia Compression (LMC) such as JPEG and BPG since they are limited in memory and computation. Instead, in this paper, we propose to down-scale images at MIoT devices with a factor of 2, 3 or ≥ 4, which reduces the memory consumption, computation, and communicated data size and consequently the latency and energy consumption. To recuperate high-quality images, we apply a Deep Learning (DL) denoising/super-resolution model at the server-side. On the other hand, as MIoT devices use a wireless connection, there is a higher risk of transmission packets loss compared to a wired connection. Almost, packets loss are managed through costly data re-transmissions or data redundancy. However, these solutions with intrinsically voluminous data such as the multimedia one are costly, especially for limited MIoT devices. To overcome this challenge, the denoising/super-resolution model did also undergo a training model to retrieve high-quality images from down-scaled erroneous ones. The obtained results show how effective this proposed solution is, especially when it comes to the enhancement of visual quality of down-scaled and erroneous images with minimum communication, latency, and consequently resource overhead.

In this work, we propose three efficient variants of a message authentication encryption (MAE) algorithm, which is based on the dynamic key-dependent concept and dynamic operation mode to reach a high level of security. These variants consist of a single pass and a single round, in addition to the use of common operations for the encryption and authentication processes to reduce the required execution time and resources. Accordingly, the proposed scheme outperforms the existing solutions that are based on the static approach with multiple rounds. Furthermore, to reduce the overhead associated with the regeneration of the dynamic key and the corresponding cryptographic primitives, we propose a simple, yet effective update process. In such a scheme, even when the same plaintext is processed, it will be encrypted and authenticated using different cryptographic primitives (substitution and permutation tables in addition to round keys), which guards against the existing cryptanalysis techniques. The experimental results show that the proposed MAE variants are more efficient than the counter with cipher block chaining message authentication code (CCM), Galois message authentication code (GMAC), offset codebook mode (OCB), and the Chacha20-poly1305. The best performance is achieved with the third MAE variant that presents a high throughput with an enhancement of at least 373% compared to CCM, 90% compared to GCM, 23% compared to OCB, and 22% compared to Chacha20-poly1305.

In this paper, we propose a new dynamic key and message-dependent hash function based on Artificial Neural Networks (ANN) that satisfies the necessary security requirements with low computational complexity. It requires only two rounds of confusion and diffusion operations. Moreover, a dynamic non-invertible construction technique of a synaptic weight matrix is defined in order to ensure the one-way property. The proposed message authentication algorithm is evaluated in terms of desirable cryptographic properties. The results show that the proposed solution is robust due to the dynamic cryptographic primitives approach, which also results in reduced latency and required resources. The initial weight matrix is changed at regular time intervals depending on the application requirements. Recent enhancement in ANN hardware implementations enables the practical application of the proposed solution within wireless and mobile networks.

Protection on end users’ data stored in Cloud servers becomes an important issue in today’s Cloud environments. In this paper, we present a novel data protection method combining Selective Encryption (SE) concept with fragmentation and dispersion on storage. Our method is based on the invertible Discrete Wavelet Transform (DWT) to divide agnostic data into three fragments with three different levels of protection. Then, these three fragments can be dispersed over different storage areas with different levels of trustworthiness to protect end users’ data by resisting possible leaks in Clouds. Thus, our method optimizes the storage cost by saving expensive, private, and secure storage spaces and utilizing cheap but low trustworthy storage space. We have intensive security analysis performed to verify the high protection level of our method. Additionally, the efficiency is proved by implementation of deploying tasks between CPU and General Purpose Graphic Processing Unit (GPGPU) in an optimized manner.

High number of rounds is needed for the existing message authentication algorithms, such as keyed hash functions like Hash-based Message Authentication Code (HMAC) or block cipher based functions like Cipher-based Message Authentication Code (CMAC) and Galois Message Authentication Code (GMAC). Moreover, the employed compression functions consist of several operations to achieve two main properties: confusion and diffusion. This large number of rounds introduces high overhead for resource-limited systems like Internet of Things (IoT) or delay-sensitive systems that have real-time requirements like Intelligent Transparent Systems. In this paper, a new lightweight message authentication algorithm is proposed to reduce the number of rounds to one. The proposed compression function is based on the RC4 stream cipher to reduce the required overhead in terms of latency and resources. Finally, the security and performance analysis shows that the proposed keyed hash function is resistant towards existing security attacks with low resources overhead.

Visible Light Communication (VLC) is a wireless technology that exploits Light Emitting Diodes (LEDs) for both, illumination and data communication. A major challenge is that this system is vulnerable to passive attacks due to the broadcast nature of wireless networks. In this paper, an efficient and lightweight cipher scheme for VLC systems is proposed at the physical layer. Unlike previous schemes in the literature, the proposed one-round scheme utilizes simple substitution and phase shuffling operations to secure the underlying Orthogonal Frequency Division Multiplexing (OFDM) symbols. A dynamic key derivation scheme that benefits from the dynamic properties of VLC channels is also proposed. Experimental simulations and cryptanalysis show that the proposed solution strikes a good balance between performance and security robustness.

Device-to-Device (D2D) communication is one of the key components of 4G/5G mobile networks since it enhances network capacity and enables support for public applications. On the other hand, device authentication in D2D is an inherent problem since devices connect and leave the network frequently and freely. Recently, "3rd Generation Partnership Project (3GPP)" has adopted the Authentication Key Agreement (AKA) protocol for 5G New Radio (NR) networks, where the Home Network (HN) first authenticates the User Equipment (UE) and then produces sessions keys. However, in D2D communication, data is directly conveyed between communicating entities (independently from the HN), which makes the AKA protocol not very suited for this technology. In this paper, we propose a new framework based on Physical Layer Security (PLS) that targets device authentication and key establishment in D2D/5G communication systems. The proposed protocol uses common channel characteristics and asymmetric cryptography to ensure legitimate authentication, without relying on the core network. Finally, security and performance analysis are presented to prove the proposed scheme's efficiency and immunity against different types of authentication attacks.

Authentication is critical for any digital system as it represents the first step towards accessing data and resources. Authentication of entities, especially devices in the Internet-of-Things (IoT) system, is one of the most important security challenges that needs to be addressed; otherwise, it will hinder the deployment of IoT applications. The most widely used authentication mechanisms in IoT are based on one-factor cryptographic techniques. These techniques are often not sufficient in the context of IoT due to the limited computational power of IoT devices and the severity of security concerns, especially that these devices are physically not well protected. Consequently, any weakness in the identification/authentication schemes would allow a compromised entity to perform dangerous attacks. To overcome the above-mentioned limitations and achieve high authentication accuracy, we propose an efficient two-factor lightweight mutual authentication scheme for IoT entities, which can be deployed at various levels; device, control, aggregation node, gateway, and server. The first factor is based on a cryptographic protocol which employs a configurable Physically Unclonable Function (PUF) along with a nonce extracted from the physical channel. The second factor is an entity-based fingerprint that uses specific information (i.e., features that can be extracted from various layers of the communication protocol) to construct a unique fingerprint for each entity. The proposed scheme is designed to require the minimum possible overhead in terms of computation and communication overhead, and ensure maximum security resilience against authentication attacks.

Data Confidentiality (DC) is considered one of the most important security services. Currently, a set of existing cipher algorithms is being used to ensure DC. However still, designing and implementing a more efficient cipher scheme is always being sought. Moreover, various vulnerabilities are constantly being targeted by new kinds of attacks such as the physical ones. In addition, some cipher algorithms exhibit limitations in terms of latency and required resources, and hence cannot be preferred to constrained devices. This leads to a trade-off between system performance and security level. Towards solving these challenges, we propose a lightweight cipher scheme that ensures a high level of security with minimal latency and resource requirements compared to existing standards such as AES. Specifically, the proposed cipher scheme is based on the original RC4 but adapted and extended by introducing two different round functions that consist of substitutions, addition, and non-invertible diffusion operations to achieve the aforementioned goals. Experimental results indicate that the proposed cipher is a strong and promising stream cipher candidate that has high key sensitivity, high randomness degree, as well as periodicity properties.

Security attacks against Internet of Things (IoT) are on the rise and they lead to drastic consequences. Data confidentiality is typically based on a strong symmetric-key algorithm to guard against confidentiality attacks. However, there is a need to design an efficient lightweight cipher scheme for a number of applications for IoT systems. Recently, a set of lightweight cryptographic algorithms have been presented and they are based on the dynamic key approach, requiring a small number of rounds to minimize the computation and resource overhead, without degrading the security level. This paper follows this logic and provides a new flexible lightweight cipher, with or without chaining operation mode, with a simple round function and a dynamic key for each input message. Consequently, the proposed cipher scheme can be utilized for real-time applications and/or devices with limited resources such as Multimedia Internet of Things (MIoT) systems. The importance of the proposed solution is that it produces dynamic cryptographic primitives and it performs the mixing of selected blocks in a dynamic pseudo-random manner. Accordingly, different plaintext messages are encrypted differently, and the avalanche effect is also preserved. Finally, security and performance analysis are presented to validate the efficiency and robustness of the proposed cipher variants.

The Internet of Things (IoT) systems are vulnerable to many security threats that may have drastic impacts. Existing cryptographic solutions do not cater for the limitations of resource-constrained IoT devices, nor for real-time requirements of some IoT applications. Therefore, it is essential to design new efficient cipher schemes with low overhead in terms of delay and resource requirements. In this paper, we propose a lightweight stream cipher scheme, which is based, on one hand, on the dynamic key-dependent approach to achieve a high security level, and on the other hand, the scheme involves few simple operations to minimize the overhead. In our approach, cryptographic primitives change in a dynamic lightweight manner for each input block. Security and performance study as well as experimentation are performed to validate that the proposed cipher achieves a high level of efficiency and robustness, making it suitable for resource-constrained IoT devices.

Physical layer (PHY) security has emerged as a promising methodology for securing current and future networks that employ orthogonal frequency-division multiplexing (OFDM) technology. OFDM is the basic building block for multicarrier modulation in most contemporary networks such as vehicular ad hoc networks, Internet of Things (IoT), as well as 4G/5G systems. Most existing OFDM-based security solutions lack the notion of secrecy and dynamicity when combining a secret key with random information extracted from the physical channel. Yet, some solutions perform encryption preinverse fast Fourier transform and some postinverse fast Fourier transform, without clear guidelines concerning the impact on performance and security. In this paper, OFDM-based encryption schemes at the PHY are investigated, analyzed, and weaknesses are identified. It is shown that encryption in the frequency domain slightly mitigates the effects of channel fading and improves the bit error-rate performance. On the other hand, time-domain encryption is shown to be more secure. Furthermore, a dynamic secret key approach that enhances the security level of OFDM-based encryption schemes, in addition to a new technique for updating cipher primitives for input OFDM symbols or frames, are proposed. These schemes are shown to strike a good balance between performance and security robustness as demonstrated through experimental simulations.

In an mHealth remote patient monitoring scenario, usually control units/data aggregators receive data from the body area network (BAN) sensors then send it to the network or “cloud”. The control unit would have to transmit the measurement data to the home access point (AP) using WiFi for example, or directly to a cellular base station (BS), e.g., using the long-term evolution (LTE) technology, or both (e.g., using multi- homing to transmit over multiple radio access technologies (Multi-RATs). Fast encryption or physical layer security techniques are needed to secure the data. In fact, during normal conditions, monitoring data can be transmitted using best effort transmission. However, when real-time processing detects an emergency situation, the current monitoring data should be transmitted real-time to the appropriate medical personnel in emergency response teams. In this paper, a fast and secure approach for transmitting monitoring data over multi-RATs is proposed. The presented approach consists of benefiting of the presence of multi-RATs in order to exchange the secrecy information more efficiently while optimizing the transmission time.

Recently, physical layer security has emerged as a promising security scheme for wireless networks, in contrast to traditional solutions that mainly rely on upper network layers. As such, several physical layer encryption algorithms that benefit from the random characteristics of physical channels have appeared in the literature. However, the majority of these schemes lack the notion of secrecy and dynamicity. In this paper, we focus on enhancing the physical layer encryption for wireless machine-to-machine devices, which share the same channel, with the aim of striking a good balance between performance and security robustness. The main idea is to perform encryption at the physical layer after symbol modulation. The cipher scheme is based on one round and one operation that reduces the encryption overhead in terms of latency and required resources. Furthermore, we propose a dynamic key approach that combines a pre-shared/stored secret key with a dynamic nonce extracted from the channel information to generate a dynamic key. The main advantage of the dynamic key approach is that it achieves a high-security level with minimal overhead. The dynamic key can be changed frequently upon any change in channel parameters or upon starting a new session. In addition to data encryption, a preamble encryption scheme is also proposed to prevent unauthorized synchronization or channel estimation by illegitimate users. Finally, security and performance analyses are performed to demonstrate the validity, efficiency and robustness of the proposed approach.

In this paper, an efficient secure storage scheme is presented which aims to provide security to end-user's data while mostly storing it to public clouds. This proposed scheme is based on the invertible Discrete Wavelet Transform (DWT) to fragment data into two or three fragments with different levels of importance and protected accordingly. As a matter of fact, the most important fragment takes the smallest amount of storage space and can be stored in a user trusted area while the less important fragments take most of the storage space and are uploaded to public clouds. In order to reduce the required execution time, General Purpose Graphic Processing Unit (GPGPU) is employed for accelerating computation. Additionally, a benchmark was realized to compare between the proposed scheme and AES algorithm applied to the entire data.

The increasing usage of personal multimedia devices such as mobiles phones, smart glasses, etc. has pointed out the need for securing the captured/exchanged images content. Nevertheless, because of the limited resources of these novel platforms on one hand and the voluminous nature of multimedia content on the other hand, preserving multimedia content security remains a research challenge. In this paper, we tackle this issue by presenting a fast format-compliant selective encryption algorithm for JPEG 2000 images. It is based on selectively choosing data from the JPEG 2000 code-stream in a uniformly dynamic-key dependent manner to apply the proposed encryption algorithm. The encryption algorithm consists of two rounds of substitution-diffusion processes, based on a dynamic key, that is changed for every input image. Extensive security analysis has been conducted to evaluate the effectiveness of the proposed scheme. The obtained results have demonstrated the robustness of our algorithm against the most known types of attacks and have shown a significant improvement in term of execution time reduction compared to a similar existing JPEG 2000 images encryption scheme.

Although symmetric ciphers may provide strong computational security, a key leakage makes the encrypted data vulnerable. In a distributed storage environment, reinforcement of data protection consists of dispersing data over multiple servers in a way that no information can be obtained from data fragments until a defined threshold of them has been collected. A secure fragmentation is usually enabled by secret sharing, information dispersal algorithms or data shredding. However, these solutions suffer from various limitations, like additional storage requirement or performance burden. This poster presents a novel flexible keyless fragmentation scheme, balancing memory use and performance with security. It could be applied in many different contexts, such as dispersal of outsourced data over one or multiple clouds or in resource-restrained environments like sensor networks. The scheme has been implemented in JAVA and Matlab. Preliminary analysis shows good performance and data protection.

The security mechanisms for Long Term Evolution (LTE) networks are essential between User Equipment and eNodeB to prevent diverse threats and attacks. Data confidentiality and Data integrity are the main inevitable features for any secure communication system. Thus the 3GPP has standardized till now three pairs of algorithms (EEA1, EIA1), (EEA2, EIA2) and (EEA3, EIA3) for LTE security. In this paper, a new efficient DI algorithm based on a keyed hash function called ERDIA is introduced. Accordingly, the use of key and message dependent diffusion layers ensure the key sensibility and the avalanche effect using only one processing round. The experimental results show that the proposed hash function is immune against most possible known attacks. Besides, a lower computational time is attained compared to EIA2. Furthermore, our proposition would be similarly well suited for other networks and applications such as wireless networks.

Abstract The idea of employing chaotic maps in building encryption schemes has attracted the attention of many researchers since the late 1980s. In 1998, M.S. Baptista proposed an elegant encryption scheme based on a one–dimensional chaotic map. Many variants of this scheme have been proposed. Baptista's scheme and some of its variants have been subjected to cryptanalytic attacks such as the one‐time pad attack and the entropy attack. We propose a variant of Baptista's encryption scheme based on two coupled one–dimensional chaotic maps, which also employs mixing, that is, every character in the ciphertext depends on all preceding plaintext characters. Our proposed scheme overcomes the aforementioned attacks. Empirical results show that this idea improves the performance of the scheme. Baptista's approach generates a ciphertext, which is generally larger than the plaintext, but the distribution of the ciphertext symbols allows compression. Simulation results verify that our proposed scheme accommodates better compression rates than the original scheme. Copyright © 2015 John Wiley & Sons, Ltd.

Wireless Multimedia Sensor Networks (WMSNs) have to deal with two main opposite constraints: (a) the voluminous nature of the sensed data, almost of megabytes order on one hand and (b) the limited resources, characterizing specifically WMSNs platforms (limited energy provision and CPU power, wireless communications, etc.) on the other hand. Additionally, as these platforms are open infrastructures, they are hence vulnerable to several types of attacks. In this paper, we propose a new and novel approach, specifically tailored to significantly reduce the transmitted multimedia data whilst allowing a high level of security, in particular data confidentiality. In fact, rather than using traditional cryptography systems which applied on multimedia data, inquiring a huge communication and hence overhead hence are not suitable within WMSNs, our approach begins first by transforming the input images using Voronoi tessellation, that reduces significantly their volume while preserving at the same time their perceived quality. As this transformation is performed on a random fashion basis, this randomness ensures then the content confidentiality. The proposed approach achieves hence both: (a) reducing noticeably the amount of the data sent by the source nodes, prolonging hence the overall network lifetime and (b) makes the network more robust against attacks. We study the parameters for setting up our approach by incorporating two schemes (basic one and more elaborated one). Finally, we demonstrate, through extensive experiments, the robustness (i.e., secure against several types of attacks) and the effectiveness of our approaches over the current state-of-the-art techniques.

International audience

Recently, several encryption schemes have been presented to Random Linear Network Coding (RLNC). The recent proposed lightweight security system for Network Coding is based upon protecting the Global Encoding Vectors (GEV) and using other vector to ensure the encoding process of RLNC at intermediate nodes. However, the current lightweight security scheme, presents several practical challenges to be deployed in real applications. Furthermore, achieving a high security level results in high computational complexity and adds some communication overhead. In this paper, a new scheme is proposed to overcome the drawbacks of the lightweight security scheme and that can be used for RLNC real-time data exchange. First, the cryptographic primitive (AES in CTR mode) is replaced by another approach that is based on the utilization of a new flexible key-dependent invertible matrix (dynamic diffusion layer). Then, we show that this approach reduces the size of communication overhead of GEV from 2 × h to h elements. In addition to that, we also demonstrate that besides the information confidentially, both the packet integrity and the source authentication are attained with minimum computational complexity and memory overhead. Indeed, cryptographic strength of this scheme shows that the proposed scheme has sufficient security strength and good performance characteristics to ensure an efficient and simple implementation thus, facilitating the integration of this system in many applications that consider security as a principal requirement.

The security of Wireless Sensor Networks (WSN) is essential for effective deployment in various areas and applications such as military and business. The existing security solutions of WSN are based on multi-round function, which requires high computing complexity and energy consumption. WSN have, however, limited resources that prevent their efficient deployment for a long period. In this paper, a new kind of security system based on a cipher and authentication algorithm called EDCA is presented to ensure the necessary security requirements with low computation complexity. Furthermore, the proposed cipher is based on a dynamic binary diffusion layer. The contents of packets is divided into many blocks, which are mixed together to produce the cipher blocks. Likewise, an enhanced version of cipher is presented to attain a better statistical properties. Additionally, EDCA is evaluated by comparing it with AES, which is considered reliable and robust in several standards of sensor networks. The results show that the proposed algorithm has a reduced computation complexity, and is robust and could be adopted for different types of wireless or mobile networks.

Random Linear Network Coding (RLNC) is a promising technology of Network coding (NC) {that is} verified to be both sufficient and efficient. In this paper, we propose an efficient implementation of coding process, ensuring the security against active and passive attacks, in order to deploy RLNC in real networks, especially on battery constrained mobile devices with low computation capabilities such as mobile phones or sensors. We first present our flexible secure solution that can achieve simultaneously the information confidentiality, the packet integrity and the source authentication. It contains a new scheme of generation of invertible key dependent binary Global Encoding Matrix (GEM) with a complexity, a memory consumption and a decoding delay lower than the traditional $RLNC$. The effectiveness of coding process is proved by modifying the Galois field of calculation from integer (int8, int16) to binary in order to ensure low computational requirements that lead to high throughput and low energy consumption. Furthermore, theoretical and numerical results reveal that the proposed methods give an effectiveness of coding and a higher level of security compared to many recent works in this field.

In this paper, a new ciphering algorithm is proposed for Long Term Evolution (LTE) data confidentiality. The proposed cipher scheme is based on a novel stream cipher framework which uses Substitution-Diffusion (SD) structure to provide key-streams that possess acceptable cryptographic performance (avalanche effect and key sensibility). Standards have been already adopted by 3GPP for LTE data confidentiality; EEA1, EEA2, and EEA3 which are based on Snow 3G, Advanced Encryption Standard (AES), and ZUC, respectively. Although the above mentioned algorithms have sufficient security strength against attacks, our solution is constructed to ensure less complexity with similar security strength. The proposed algorithm consists of an addition layer and a modified RC6 substitution layer which could require no memory and has stronger cryptographic properties compared to RC6. Furthermore, a new dynamic non-invertible diffusion layer technique is introduced, which is constructed from the output of the substitution layer. Theoretical and simulation results show that our algorithm is immune against liner, differential, chosen/known-plain-text, brute force and statistical attacks. Equally important to note, our proposed cipher algorithm has a lower computational time compared to AES, and could be adapted for other kinds of wireless networks.