ABSTRACT In the current panorama of digital healthcare evolution and the proliferation of electronic health records (EHRs), healthcare systems face many challenges. These challenges encompass data administration and security, data exchange, and ensuring the confidentiality of medical data. Within this context, blockchain technology emerges as a promising way to address the numerous challenges associated with EHRs. Ensuring the security and confidentiality of EHRs remains a pivotal concern encompassing healthcare service recipients and providers. The compromise of a healthcare system leads to the exposure of intricately private health information. Typically stored in centralized databases, this information repository introduces susceptibilities that consequently fuel instances of cyber intrusion. To ensure the safeguarding of medical data, this research paper strategically adopts the Hyperledger Fabric (HLF) blockchain platform, which incorporates attribute‐based access control mechanisms to thwart any malicious attempt at unauthorized access to sensitive information. These apprehensions regarding security have been effectively addressed by leveraging robust cryptographic protocols such as the AES‐256 algorithm. This algorithm encrypts messages, transmitting the encrypted data across the network, thereby restricting visibility solely to intended recipients and providing a secure data exchange. The effectiveness of our proposed solution is rigorously evaluated using the Hyperledger Caliper tool. The evaluation encompasses pivotal performance metrics, average latency, throughput, success rate, resource consumption, and traffic.

ABSTRACT Innovations in technology are revolutionizing healthcare, driving a shift toward patient‐centric smart healthcare systems. Mobile health (mHealth) leverages innovations in wearable sensors, telecommunications, and IoT to establish a novel healthcare model that prioritizes the patient, enabling real‐time monitoring, personalized interventions, and improved access to care, ultimately fostering a proactive approach to health management and enhancing overall patient outcomes. However, safeguarding patient data transparency, security, and privacy within mHealth systems presents significant challenges, particularly concerning personal health records (PHR). Ciphertext‐Policy Attribute‐Based Encryption (CP‐ABE) offers a competent answer to facilitating one‐to‐many data sharing in healthcare environments. Nevertheless, several issues must be addressed before CP‐ABE can be widely deployed. These include the need for timely and effective attribute revocation when user attributes change, resistance to collusion attacks, and ensuring data integrity. This paper proposes a revocable and secure fine‐grained access scheme using blockchain and CP‐ABE. We compare four prominent state‐of‐the‐art schemes through comprehensive experimentation with our proposed approach. Our results demonstrate the relative performance of our scheme, showing a significant reduction in computational costs. Specifically, the key generation cost is reduced by 35% to 67%, and the encryption cost is reduced by 26% to 39%. A detailed analysis of communication, computational, and storage overhead reveals that our suggested solution offers a distinct advantage in terms of efficiency. The Scyther tool is employed to verify the security measures and assess the accuracy of proposed methodologies, subsequently conducting experiments to showcase its efficacy.

ABSTRACT This research endeavor aims to advance the field of Automatic Speech Recognition (ASR) by innovatively integrating multimodal data, specifically textual transcripts and Mel Spectrograms (2D images) obtained from raw audio. This study explores the less‐explored potential of spectrograms and linguistic information in enhancing spoken word recognition accuracy. To elevate ASR performance, we propose two distinct transformer‐based approaches: First, for the audio‐centric approach, we leverage RegNet and ConvNeXt architectures, initially trained on a massive dataset of 14 million annotated images from ImageNet, to process Mel Spectrograms as image inputs. Second, we harness the Speech2Text transformer to decouple text transcript acquisition from raw audio. We pre‐process Mel Spectrogram images, resizing them to 224 × 224 pixels to create two‐dimensional audio representations. ImageNet, RegNet, and ConvNeXt individually categorize these images. The first channel generates the embeddings for visual modalities (RegNet and ConvNeXt) on 2D Mel Spectrograms. Additionally, we employ Sentence‐BERT embeddings via Siamese BERT networks to transform Speech2Text transcripts into vectors. These image embeddings, along with Sentence‐BERT embeddings from speech transcription, are subsequently fine‐tuned within a deep dense model with five layers and batch normalization for spoken word classification. Our experiments focus on the Google Speech Command Dataset (GSCD) version 2, encompassing 35‐word categories. To gauge the impact of spectrograms and linguistic features, we conducted an ablation analysis. Our novel late fusion strategy unites word embeddings and image embeddings, resulting in remarkable test accuracy rates of 95.87% for ConvNeXt, 99.95% for RegNet, and 85.93% for text transcripts across the 35‐word categories, as processed by the deep dense layered model with Batch Normalization. We obtained a test accuracy of 99.96% for 35‐word categories after using the late fusion of ConvNeXt + RegNet + SBERT, demonstrating superior results compared to other state‐of‐the‐art methods.

Summary Wireless sensor and actor networks (WSANs) consist of nodes associated in an ad hoc manner to perform sensing tasks for information gathering and acting functions on the basis of gathered information. Connectivity is an essential requirement of large‐scale wireless networks, and WSANs are supposed to stay connected. The nodes in hostile environments are prone to failures such as battery depletion, physical damage, or hardware malfunction. The failure of some nodes, like cut vertex nodes, can partition the network into multiple network segments. Most of the solutions for network partition recovery proposed in the literature depend on the assumption that the network is obstacle‐free. However, an obstacle‐free environment is not possible in real‐life situations. In the last few decades, UAVs or drones have been engaged in various applications such as industrial inspections, remote sensing, agriculture, military, disaster relief, and so forth, UAVs can be employed to strengthen the connections in wireless networks by coordinating with ground nodes since they can render services in rough areas where ground nodes cannot provide services. Thus, our research is based on using UAVs as relay nodes to reconnect the disjoint partitions. This paper proposes two algorithms: Drone assisted partition recovery algorithm (DAPRA) and drone assisted detection and partition recovery algorithm (DADPRA). In both algorithms, partitions are detected by the sink node. In DAPRA sink node determines the failed cut‐vertex node and sends UAV to the location of the failed cut‐vertex node. In DADPRA algorithm, UAV identifies the failed cut‐vertex node and reconnects the disjoint network segments. DAPRA and DADPRA are analyzed according to the state‐of‐the‐art parameters, that is, recovery and detection time, UAV's travel distance, and the total messages transmitted. The proposed algorithms are compared with similar Distributed Partition Detection and Recovery using UAV (DPDRU) approach. The simulation results show the proposed algorithms detect network partitioning in less time as compared to DPDRU approach.

This paper proposes an efficient algorithm for tracking multiple targets using a network of static and mobile sensors (robots). Multi-target tracking has a broad array of applications, including crowd monitoring, vehicle tracking, warehouse automation, and pedestrian safety, among others. The problem of distributed labeled multi-target tracking comprises constraints on sensing range, communication, label consistency, and motion. Hence, our algorithm strives to minimize label mismatching, communication, movement of robots, and tracking error, which are serious concerns in the existing solutions. The problem is decomposed into two sub-problems: distributed estimation and adaptive movement control. We present a novel track consensus algorithm for estimating the number and tracks of targets, complemented by an efficient label consensus method. This algorithm can effectively identify similar tracks and fuse them in cluttered scenarios. Various movement control strategies are proposed to minimize the moving distance of the robots while keeping the maximum number of targets in the sensing range. The maximum target sensing problem is NP-hard; therefore, we propose and compare heuristic, approximation, and randomized algorithms. We have also verified our proposed solution through extensive simulations and compared the distributed estimation and movement control algorithms with other prominent solutions. We also analyze estimation accuracy using the Optimal Sub-Pattern Assignment (OSPA) metric, asymptotic performance, and communication cost and confirm the real-time computation of our proposed algorithms.

Over the past decade, IoT has gained huge momentum in terms of technological exploration, integration, and its various applications even after having a resource-bound architecture. It is challenging to run any high-end security protocol(s) on Edge devices. These devices are highly vulnerable toward numerous cyber-attacks. IoT network nodes need peer-to-peer security, which is possible if there exists proper mutual authentication among network devices. A secure session key needs to be established among source and destination nodes before sending the sensitive data. To generate these session keys, a strong cryptosystem is required to share parameters securely over a wireless network. In this article, we utilize a Rubik's cube puzzle-based cryptosystem to exchange parameters among peers and generate session key(s). Blockchain technology is incorporated in the proposed model to provide anonymity of token transactions, on the basis of which the network devices exchange services. A session key pool randomizer is used to avoid network probabilistic attacks. Our hybrid model is capable of generating secure session keys that can be used for mutual authentication and reliable data transferring tasks. Cyber-attacks resistance and performance results were verified using standard tools, which gave industry level promising results in terms of efficiency, light weightiness, and practical applications.

Summary Software‐defined networking (SDN) is an emerging paradigm in enterprise networks because of its flexible and cost‐effective nature. By decoupling control and data plane, SDN can provide various defense solutions for securing futuristic networks. However, the architectural design and characteristics of SDN attract several severe attacks. Distributed denial of service (DDoS) is considered as a major destructive cyber attack that makes the services of controller unavailable for its legitimate users. In this research article, an intrusion detection framework is proposed to detect DDoS attacks against SDN. The proposed framework relies on voting‐based ensemble model for the attack detection. Ensemble model is a combination of multiple machine learning classifiers for prediction of final results. In this research article, we propose and analyze three ensemble models named as Voting‐CMN, Voting‐RKM, and Voting‐CKM particularly to benchmarking datasets such as UNSW‐NB15, CICIDS2017, and NSL‐KDD, respectively. For validation of the proposed models, a cross‐validation technique is used with the prediction algorithms. The effectiveness of proposed models is evaluated in terms of prominent metrics (accuracy, precision, recall, and F‐measure). Experimental results indicate that the proposed models achieve better performance in terms of accuracy as compared with other existing models.

In the RPL routing protocol, DODAG Information Solicitation (DIS) control messages are sent by nodes to join the network. In turn, the receiver node replies with DODAG Information Object (DIO) control message after resetting its trickle timer. A malicious node can utilize this RPL protocol behavior to perform the DIS flooding attack by sending illegitimate DIS frequently which forces normal nodes to reset their trickle timers and flood the network with DIO messages. In this study, we show that such attacks can severely degrade the performance of Low Power and Lossy Networks (LLNs) because of the increase in control packet overhead and power consumption. To address DIS flooding attacks, we propose a lightweight mitigation scheme that detects and mitigate such attacks in order to improve LLNs performance.