International Journal of Communication Networks and Distributed Systems (18 papers in press)

Regular Issues

• Weight possibilistic fuzzy C-means energy-efficient clustering (WPFCM-EEC) and fuzzy osprey optimisation algorithm (FOOA) for IoT-WSN  
  by T. Kanimozhi , S. Belina V.J. Sara, S.Albert Antony Raj 
  Abstract: The wireless sensor network (WSN) is a key component of IoT systems. Wireless sensing nodes in WSNs have energy limitation issues since replacing or recharging their batteries and upgrading their transmitting and memory capacity is difficult. Thus, clustering algorithms efficiently save node energy and lengthen IoT-based WSN lifespan. These protocols cluster nodes to communicate with a BS at a shorter distance. However, clustering structure concerns hurt current clustering methods. In this research, the weight possibilistic fuzzy C-means-energy efficient clustering (WPFCM-EEC) protocol is presented to extend IoT-based WSN lifetime. The WPFCM-EEC protocol has three parts. First, the optimal cluster count and overlap are established. Next, a weight possibilistic fuzzy C means algorithm creates balanced and static clusters to ensure sensor node energy balance. A unique Cluster Head selection and rotation algorithm that combines the fuzzy osprey optimisation algorithm (FOOA) with a rotation mechanism selects CH in optimal locations by rotating the cluster head function across cluster members. The methods are tested in MATLAB R2023a. Protocol evaluation indicators include first node dies (FND), last node dies (LND), weighted first node dies (WFND), half of the nodes dies (HND), average energy consumption, and throughput.
  Keywords: WSN; wireless sensor network; WPFCM-EEC; weight possibilistic fuzzy C-means-energy efficient clustering; IoT; Internet of Things; FND; first node dies; FOOA; fuzzy osprey optimisation algorithm; LND; last node dies; WFND; weighted first node dies.
  DOI: 10.1504/IJCNDS.2025.10067367
   
  
• Integration of GRU Features with Q-Learning based VARMA for Protocol DDoS Attack Analysis  
  by Meghana Solanki, Sangita Chaudhari 
  Abstract: Network security faces severe threats from attacks on distributed denial of service (DDoS), necessitating attack detection and mitigation. This study introduces a novel approach by integrating Gated Recurrent Unit (GRU) features with a Q-Learning basedVector Autoregressive Moving-Average (VARMA) process for protocol DDoS attack analysis. Traditional models could not deal with complex temporal dependencies in network traffic data, whereas deep learning models lack interpretability and incremental learning. Our model combines Q-Learning based VARMA and GRU, accurately capturing temporal dynamics to detect protocol DDoS attacks. It facilitates incremental learning for adaptation over time. With applicability across diverse contexts, the model offers real-time attack identification, enhancing network security. Experimental results demonstrate superior performance in recall, precision and accuracy compared to existing machine learning and deep learning models.
  Keywords: protocol DDoS attacks; GRU features; Q-learning; VARMA and process; traffic volume; patterns; internet protocol (IP) address; port number; packet size; data rate.
  DOI: 10.1504/IJCNDS.2025.10067549
   
  
• Wireless communication interference signal recognition model based on deep CNN  
  by Bo Liang 
  Abstract: A complex neural network model based on deep convolutional neural networks is proposed to enhance recognition and suppression of interference signals in wireless communications. The model introduces a signal suppression network to address poor reception due to signal interference during transmission, which is a significant challenge in maintaining communication quality in increasingly complex wireless environments. Results show higher recognition accuracy for different interference signals at varied decibels. In complex networks, interference signal recognition accuracy is superior, with noise recognition accuracy surpassing other networks by 5% and 2%. The research method exhibits a lower bit error rate and 0.5dB better amplitude suppression compared to traditional methods. The approach excels in interference signal identification and suppression, improving wireless communication signal recognition performance significantly. This advancement is crucial in ensuring the reliability and security of wireless communication systems, offering a novel solution to the growing challenges posed by interference in modern communication networks.
  Keywords: wireless communication signal; BER; bit error rate; signal interference; CNN; convolutional neural network; accuracy; noise.
  DOI: 10.1504/IJCNDS.2025.10067633
   
  
• A task scheduling technique in the cloud computing environment based on bacterial colony optimisation  
  by R.M. Aravind, R. Pragaladan 
  Abstract: Cloud computing offers high accessibility, scalability, and flexibility in the modern computing era for various useful applications. Distributing and coordinating tasks to get the best resource usage and prevent overload is a difficult problem. In this research, we suggested a load-balancing (LB) method that minimises makespan and increases resource utilisation by employing novel bacterial colony optimisation (BCO) task scheduling to schedule jobs over the available resources. Aiming to balance load across virtual machines (VMs) according to makespan, cost, and resource utilisation all of which are constrained by concurrent considerations the suggested approach also seeks to maximise VM throughput. The CloudSim simulator is used to implement our suggested scheduling strategy. According to the testing results, the algorithms that employed the BCO technique performed better than the other techniques in terms of makespan reduction, low execution time, and average resource usage.
  Keywords: cloud computing; load balancing; BCO; bacterial colony optimisation; task scheduling.
  DOI: 10.1504/IJCNDS.2025.10068708
   
  
• An ad-hoc parallelism approach for accelerating compute-intensive mobile workloads  
  by Yomna M. Abdelmoniem, Islam A.T.F. Taj-Eddin, Nagwa M. Omar, Hosny M. Ibrahim 
  Abstract: Recently, the processing power of mobile devices has increased along with the breadth of applications that can run on them. Running these programs on mobile devices presents a significant barrier due to their processing complexity. In this work, a system for effective task execution in the context of mobile ad hoc parallelism is proposed as a solution to this problem. A new resource allocation scheme for heterogeneous mobile ad hoc cloud (Het-MAHC) is proposed by leveraging information about each mobile device's processing capacity and link lifetime. The proposed approach uses this information in the selection and partitioning process and enables parallel computing and transmission to reduce execution time and improve performance. Using Network Simulator 3, the new scheme is assessed, and its performance is evaluated against the state-of-the-art method. The proposed system is also integrated into the Android OS, and its performance was evaluated on real smartphone devices. Results show substantial performance improvements over existing approaches.
  Keywords: mobile computing; task offloading; resource allocation; parallelism; wireless ad-hoc networks; MAHC; mobile ad-hoc cloud; mobile crowd/volunteer computing; edge/fog/dew computing; IoT; Internet of Things.
  DOI: 10.1504/IJCNDS.2025.10068621
   
  
• TPOT-IDSDN: an AutoML-based model optimisation for intrusion detection system against cyber threats in software defined-networking  
  by D.Sendil Vadivu , Aswin Valsaraj, Ashwin Santhosh, Kaustub Pavagada, Narendran Rajagopalan 
  Abstract: The architectural shift of software defined networks (SDN) creates new security concerns, necessitating the creation of robust intrusion detection systems (IDS) to protect the network infrastructure. This paper focuses on the essential challenge of selecting classifiers for anomaly-based IDS in an SDN environment. An automated machine learning (AutoML) framework called tree-based pipeline optimisation tool (TPOT) was used to speed up this procedure substantially. TPOT automates model selection and hyperparameter optimisation, to decide a best classifier suited for the given dataset. The TPOT framework selected the ExtraTreesClassifier for multiclass and the XGB stacked with the BernoulliNB classifier for binary class with lower execution time (26.91 s, 11.29 s) and 100 % accuracy. A comprehensive examination of standard nine machine learning (ML) classifiers confirmed TPOT has provided the best model. When deployed in the IDS framework of SDN, the selected classifiers showed a 100% detection rate that outperformed other existing approaches.
  Keywords: AutoML; automated machine learning; SDN; software defined network; TPOT; tree-based pipeline optimisation tool; cyber security; intrusion detection systems.
  DOI: 10.1504/IJCNDS.2025.10068807
   
  
• Opposition based learning-lyrebird optimisation approach for optimal path planning in UAV-WSN environment  
  by Nilabh Kumar, Prabhat Kumar 
  Abstract: The rapid advancement in wireless sensor networks (WSNs) has prompted the need for efficient data collection, particularly using unmanned aerial vehicles (UAVs). However, selecting an optimal path for UAVs to collect data from sensor nodes while avoiding obstacles is a significant challenge. Thus, this research introduces a novel meta-heuristic optimisation approach for UAV path planning to address these challenges. Initially, a system model is designed that includes a UAV and a set of sensor nodes randomly deployed within a specified area. The proposed method focuses on UAV path planning using a novel opposition-based learning-lyrebird optimisation approach (OBL-LOA). The proposed approach offers a significant improvement in efficiency and performance for UAV path planning in terms of average flight time (s), network life time (rounds), task completion time, average path length (m), average energy consumption (J), and average data collection efficiency (%) and accomplished 27.5801, 2605.63, 1.03425, 33.716, 0.025, and 0.945 respectively.
  Keywords: path planning; opposition learning; lyrebird optimisation; unmanned automated vehicle; obstacles.
  DOI: 10.1504/IJCNDS.2026.10069337
   
  
• Secure and reliable cooperative spectrum sensing in the presence of massive probabilistic Byzantine attacks  
  by Flavien Donkeng Zemo, Sara Bakkali, Ahmed El Hilali Alaoui  
  Abstract: In cognitive radio networks (CRNs), cooperative spectrum sensing (CSS) improves spectrum sensing performance in radio environments subject to fading and shadowing. However, when some secondary users (SUs) share falsified sensing information in CSS through a falsification attack on sensing data (spectrum sensing data falsification attack, SSDF), the sensing performance degrades significantly. This document proposes a new probabilistic SSDF attack model and a new defence strategy based on a robust identification and suppression mechanism against massive SSDF attacks. Simulation results obtained under various massive probabilistic SSDF attack scenarios show that the detection performance of the proposed defence scheme outperforms that of the weighted sequential probability ratio test (WSPRT), sequential probability ratio test (SPRT), and Majority fusion techniques with which comparisons have been made. The proposed defence scheme guarantees a near-zero probability of error whatever the number of attackers and whatever the SSDF attack strategy, which is not the case for WSPRT, SPRT, and Majority rule.
  Keywords: Byzantine attacks; CR; cognitive radio technology; CSS; cooperative spectrum sensing; data fusion; SS; spectrum sensing; WSPRT; weighted sequential probability ratio test.
  DOI: 10.1504/IJCNDS.2026.10069338
   
  
• A traffic classification-based traffic engineering framework in software-defined networking  
  by Chih-Yu Lin, Chien-Cheng Wu, Hong-Yi Huang 
  Abstract: Traffic engineering is used to optimise network performance. Due to the dynamic nature of the network environment, devising an efficient traffic engineering approach attracts more attention in next-generation networks. However, network performance optimisation is challenging due to stringent requirements and the need for global observation within the dynamic network environment. Fortunately, software-defined networking (SDN) can provide an abstract global view of the complete network environment, so we are motivated to leverage the SDN framework to construct next-generation networks. In this study, we propose a traffic engineering framework for SDN. Our proposed framework optimises transmission paths by employing traffic classification techniques. In addition, we divide this framework into three modules that can operate independently. Therefore, compared with conventional methods, our proposed framework shows greater flexibility. The superiority of our approach has been verified through rigorous testing on the Mininet simulator and the Ryu controller. Finally, we firmly believe that our contribution opens new avenues for more efficient and optimised network management.
  Keywords: network performance optimisation; SDN; software-defined networking; traffic classification; traffic engineering; Mininet; Ryu.
  DOI: 10.1504/IJCNDS.2026.10069766
   
  
• DDoS attack detection in the cloud environment using an optimized long-short-term memory with an improved firefly algorithm  
  by M.C. Malini, N. Chandrakala 
  Abstract: Distributed denial of service (DDoS) attacks are the most dangerous types of attacks on cloud computing. For cloud computing technology to be widely used, defenses against these threats must be developed. Hence, this present research work proposes a new detection scheme based on a long short-term memory (LSTM) optimized by an improved firefly algorithm (IFA) called LSTM-IFA. The IFA uses opposition-based learning (OBL) to increase population diversity and the local search algorithm (LSA) for enhancing its exploitation is the second enhancement. The IFA is used to enhance the performance of LSTM by optimizing hyperparameters that produce high detection accuracy with a fast convergence rate. Experimental findings were done over four distinct datasets to evaluate the proposed LSTM-IFA approach which is obtained 98.67 % of average accuracy. The experiment's findings demonstrated that, compared to previous detection techniques, the suggested enhanced LSTM methodology achieved a greater detection rate and accuracy.
  Keywords: cloud computing; DDoS attack detection; long-short term memory; firefly algorithm; LSA; local search algorithm; OBL; opposition-based learning.
  DOI: 10.1504/IJCNDS.2026.10070246
   
  
• Secure emergency MAC protocol for wireless body area networks  
  by Bhavana Alte, Amarsinh Vidhate 
  Abstract: Wireless body area networks (WBANs) connect many small body sensors for Internet of Things healthcare applications. In vivo and on-body sensor nodes allow WBANs to detect and gather biometric data on bodily changes. Wireless transmission sends observed data. This information can help patients in a critical condition or who cannot reach hospitals due to a physical handicap, or traffic, receive immediate care. Another crucial requirement for WBANs is capacity to provide quality of service (QoS) for different traffic data. Security and privacy are needed for healthcare professionals to use and store patient records securely. It is crucial for WBANs to address privacy and security concerns. The proposed method introduces the emergency MAC (E-MAC) super-frame architecture, enabling QoS. E-MAC accelerates and reliably transmits emergency data via an emergency information management system. To address security concerns, the approach is protected using elliptic curve cryptography. Results show that E-MAC outperforms IEEE 802.15.6.
  Keywords: WBAN; wireless body area network; ECC; elliptic curve cryptography; MAC protocols; emergency traffic; duty-cycle MAC.
  DOI: 10.1504/IJCNDS.2025.10064449
   
  
• A linear regression based prediction model for load distribution and quality of service improvement with different resource utilisation in cloud environment  
  by Gopa Mandal, Santanu Dam, Kousik Dasgupta, Paramartha Dutta 
  Abstract: Cloud computing is a delivery-based consumption model that relies on the internet. The use of cloud enabled devices is increasing rapidly. So, to maintain quality of service (QoS), throughput of the entire system with service level agreements (SLAs) is a major concern between the service providers and the end users. Alternative techniques for virtual machine (VM) consolidation and proper workload allocation may be beneficial. This study proposes a linear regression-based prediction model for load distribution and QoS improvement. The model aims to enhance system throughput and QoS by predicting resource utilisation levels using historical consumption data. Experiments conducted using the CloudSim and CloudAnalyst platforms demonstrate positive results, outperforming existing methodologies. The study also evaluates service level agreement violation (SLAV) and delays to assess the QoS provided by the cloud service provider (CSP). Overall, this research contributes to the enhancement of QoS in cloud and cloud enabled systems like the Internet of Things and the Cloud of Things (CoT) and addresses the challenges of optimising resource utilisation while ensuring QoS.
  Keywords: CoT; Cloud of Things; IoT; Internet of Things; VM consolidation; cloud computing; QoS; quality of service; CloudSim; CloudAnalyst; linear regression.
  DOI: 10.1504/IJCNDS.2025.10065020
   
  
• Harnessing machine learning for dynamic defence in the battle against 5G cybersecurity threats  
  by V. Aanandaram, P. Deepalakshmi 
  Abstract: The evolution mobile network highlighted by means of 5G networks, has caused advanced cyber threats necessitating modern security features. The adaptive multi-layer threat defence machine addresses these threats with machine getting to know (ML), presenting robust resilience. It surpasses traditional strategies via deploying ML algorithms throughout more than one network layers. Network behaviour profiling (NBP) establishes baselines for customers/gadgets, detecting deviations as early malicious signs. Intent prediction (IP) visually anticipates person purpose, while anomaly detection (AD) identifies subtle anomalies. The system's centre, decentralised associative learning (federated learning), keeps confidentiality and model integrity. Continuous Threat Intelligence Integration (TII) permits proactive responses to rising threats. This integrated approach provides better protection for 5G networks, creating an adaptive defence via profiling, prediction and anomaly detection. The adaptive multi-layer threat defence system, combining flexibility, privacy, and scalability, ushers in a generation in which ML supports technological development.
  Keywords: AML-TDS; adaptive multi-layered threat defence system; threat intelligence; anomaly detection; privacy preservation; 5G network security; NBP; network behaviour profiling; cyber threats.
  DOI: 10.1504/IJCNDS.2025.10065033
   
  
• Blockchain-based privacy-preserving technology to secure shared data in vehicular communication  
  by Omessaad Slama, Walid Dhifallah, Salah Zidi, Jaime Lloret, Bechir Alaya, Mounira Tarhouni 
  Abstract: In vehicular ad-hoc networks (VANETs), securely exchanging sensitive data like misbehaviour detection models faces significant security and privacy hurdles. Our solution, machine learning model blockchain-based privacy-preserving (MBPP), combines Blockchain technology and advanced cryptography to tackle this. MBPP ensures data confidentiality and integrity while improving detection model reliability. It involves securely storing ML models on the blockchain using cryptographic hash functions. Our study meticulously evaluates transactional time and computational costs, vital for smooth blockchain transactions. This research not only presents a conceptual framework for blockchain use in VANETs but also offers insights into managing transactions via smart contracts, addressing VANETs' security and privacy challenges effectively.
  Keywords: blockchain; smart contract; computational costs; artificial intelligence algorithms; VANETs; vehicular ad-hoc networks; data sharing; privacy preservation; cryptography hash function.
  DOI: 10.1504/IJCNDS.2025.10064597
   
  
• Enhancing road safety: UAV-enabled warning message dissemination in urban and rural environments for VANETS  
  by Leila Bouchrit, Sajeh Zairi, Ikbal C. Msadaa, Amine Dhraief 
  Abstract: Vehicular ad hoc networks (VANETs) safety applications aim to mitigate the growing frequency of daily road accidents. These applications require the prompt dissemination of warning messages among vehicles, regardless of traffic conditions or environment. The limited terrestrial coverage in sparsely populated rural regions challenges the timely receipt of warnings during accidents. To address these issues, integrating unmanned aerial vehicles (UAVs) into VANETs offers a promising solution. UAVs can function as flying relays, meeting necessary delay limitations and overcoming communication problems. In prior work, we introduced a unified UAVs-VANET architecture, using UAVs to relay safety messages among vehicles in rural settings. This paper extends that work by integrating UAVs in rural and urban environments. Our approach includes determining the minimal residual UAV energy needed to connect to a charging station, ensuring continuous UAV operation. This extended framework ensures all vehicles on a 25 km urban highway receive alerts within 1.95 seconds.
  Keywords: VANET; vehicular ad hoc network; urban environment; rural environment; UAV; unmanned aerial vehicle; LTE/4G D2D; long term evolution device-to-device; IEEE 802.11p; energy optimisation.
  DOI: 10.1504/IJCNDS.2025.10065615
   
  
• Joint coded caching and distributed storage with product matrix non-systematic code  
  by Natasa Paunkoska-Dimoska, Venceslav Kafedziski 
  Abstract: The coded caching technique and data distribution on multiple servers, known as a distributed storage system (DSS), are two separate concepts that bring various improvements to the communication environment. Combining the two techniques enhances the user network performance by decreasing the peak rate, optimising the user memory and decreasing the latency. This paper investigates the benefits of merging coded caching and distributed storage in various multiserver systems. Hence, we propose a new construction called the non-systematic. This approach considers using a multi-server system consisting of only parity servers and encoded linear combinations of popular file segments based on a predefine non-systematic code in the placement and delivery phases making this concept attractive in terms of security. The proposed concept achieves optimised results regarding the peak rate and the user cache memory and tremendously improves the data security compared to other multi-server systems.
  Keywords: coded caching; cache memory; distributed storage; multi-server system; peak rate; security; systematic code; non-systematic code.
  DOI: 10.1504/IJCNDS.2025.10065568
   
  
• Adoption of Internet of Things (IoT) among manufacturing SMEs in developing countries: a TOE framework perspective  
  by Syeda Khadija Mubeen, Ali Vafaei-Zadeh, Azlan Amran, Chang Ruiqi 
  Abstract: This study develops and validates an eight-driver framework by integrating technology-organisation-environment (TOE) theory with institutional theory to enhance understanding of Internet of Things (IoT) adoption intentions. Small and medium enterprises (SMEs) in developing nations face challenges such as inadequate technological infrastructure, limited organisational resources, and external support. Focusing on Pakistani manufacturing SMEs, this research investigates the influence of the TOE framework and institutional theory on IoT adoption intentions, using data from 207 firms analysed through partial least squares structural equation modelling (PLS-SEM). The analysis supported seven of the proposed hypotheses, demonstrating positive impacts of TOE drivers on IoT adoption intentions, except for cost. Limitations include the restricted scope of data collection and the omission of pre-adoption phases. This research provides valuable insights into how IoT can strategically benefit manufacturing SMEs in Pakistan, offering guidance for policymakers, researchers, and practitioners. Recommendations include the need for SMEDA and the Pakistani government to improve technology adoption processes and implementation at the global level.
  Keywords: TOE framework; technology adoption; IoT; Internet of Things; manufacturing SME sector; institutional theory; PLS-SEM; partial least squares structural equation modelling.
  DOI: 10.1504/IJCNDS.2025.10067141
   
  
• Parallel iterative algorithms for Markovian systems on distributed architectures  
  by Mohamed Jarraya 
  Abstract: This paper explores parallel iterative methods for solving Markovian systems, aiming to tackle computational challenges in scientific and industrial contexts. Two strategies for parallelising the Gauss-Seidel iterative scheme in the circuit-switching networks model are deployed and evaluated on both shared memory multiprocessor systems and networks of shared memory multiprocessor machines. The first strategy involves modifying the Gauss-Seidel iterative scheme, while the second employs a colouring technique for components in red and black. An activation message-based termination method is introduced for asynchronous iterations on networks of shared memory multiprocessor machines. Additionally, a novel parallel iterative method for general Markovian systems is proposed and evaluated for both synchronous and asynchronous implementations. This method distributes computational workload differently from the conventional approach used in circuit switching networks.
  Keywords: Markovian systems; asynchronous iterations; synchronous iterations; termination algorithms; shared memory multiprocessor; network of shared memory multiprocessor; relaxation method; over-relaxation method; parallel alternate components relaxation method.
  DOI: 10.1504/IJCNDS.2025.10066711