International Journal of Ad Hoc and Ubiquitous Computing (7 papers in press)

Regular Issues

• Distributed Denial of Service Attack Detection Using Machine Learning Classifiers  
  by Gautam Rampalli, R. Padmavathy 
  Abstract: Online services risk distributed denial of service attacks due to their availability. These attacks overload system resources and make them unusable by legitimate users. This study aims to analyse publicly available datasets spanning three years. This analysis uses machine learning classifiers to detect and classify the attacks. The experimental results of this approach demonstrate precise attack detection and classification with minimal false-positive rates. This study utilised publicly available datasets and employed machine learning classifiers. Decision tree and random forest classifiers achieved the highest accuracy rates, and the K-nearest neighbours and support vector machine classifiers took longer to execute. Correlation coefficient and recursive feature elimination approaches gave more insights into the features of the utilised datasets. Machine learning models were used to analyse attacks and determine the best accuracies for detection. Machine learning provided favourable detection rates for DDoS attacks, underscoring the importance of algorithm selection.
  Keywords: Denial of Service; Distributed Denial of Service attack detection; Machine Learning Classifiers; Correlation Analysis; Recursive Feature Elimination.
  DOI: 10.1504/IJAHUC.2024.10064418
   
  
• A Hybrid Optimisation Enabled Deep Learning for Object Detection and Multi-Object Tracking  
  by Thirumalai J, Gomathi M, Sindhu T. S, A. Senthil Kumar, Puviarasi R 
  Abstract: The potential of multi-object tracking (MOT) in academia and industry has drawn growing attention. Despite the various methods that have been suggested to address this issue, it continues to be difficult because of things like sudden changes in appearance and severe object occlusions. In this paper, a Jaya political search optimisation (Jaya-PSO) enabled ShuffleNet is developed for object detection (OD) and MOT. Initially, the input video is fed to video frame extraction. The extracted frames are fed into the object segmentation phase, where the segmentation is done by the mask-regional convolutional neural network (Mask-RCNN), trained by tangent squirrel search optimisation (TSSO). Here, TSSO is the integration of the tangent search algorithm (TSA) and squirrel search optimisation (SSO). Then, the object recognition is performed using ShuffleNet trained by Jaya-PSO, where the Jaya-PSO is from the Jaya algorithm, political optimiser (PO) and TSA. Finally, MOT is done by the Henry gas solubility optimised unscented Kalman filtering (HGSO-based UKF). The HGSO-based UKF is the integration of Henry gas solubility optimisation (HGSO) and unscented Kalman filtering (UKF). The measures utilised for analysis are accuracy, sensitivity, specificity and multiple object tracking precision (MOTP). The proposed method attained 92.9% accuracy, 92.1%sensitivity, 92.9% specificity, and 91.0% MOTP.
  Keywords: object detection; multi-object tracking; mask-regional convolutional neural network; Mask-RCNN; tangent search algorithm; TSA; Jaya algorithm; political optimiser.
  DOI: 10.1504/IJAHUC.2024.10064483
   
  
• Throughput Enhancement of RIS and STARRIS Using Adaptive Modulation and Coding for 6G Systems  
  by Nadhir Ben Halima, Sajid M. Sheikh 
  Abstract: In this article, the throughput of reconfigurable intelligent surfaces (RIS) is improved using adaptive modulation and coding (AMC). The best modulation and coding scheme (MCS) is selected at the transmitter using the instantaneous signal to noise ratio (SNR) measured at the receiver and sent back on the feedback channel. RIS is placed between the transmitter and the receiver so that the SNR is maximised at the receiver side as RIS reflections are combined coherently at the receiver and results in significant spatial diversity. All reflections on RIS have a zero phase at the receiver. We show that RIS using AMC offers a larger throughput than conventional RIS with fixed MCS for all SNR range. RIS with AMC offers 48, 42, 36, 30, 24, 18 dB gain for N = 256,128,64,32,16,8 reflectors. We obtained 10 dB gain when RIS uses AMC versus 256-QAM. We also improve the throughput of Simultaneously Transmitting And Reflecting RIS (STARRIS) using AMC. We obtained 13 dB gain when STARRIS uses AMC versus 256-QAM.
  Keywords: AMC; RIS; MCS; throughput enhancement; 6G.
  DOI: 10.1504/IJAHUC.2024.10064800
   
  
• Throughput Optimization of Intelligent Omni-Surfaces (IOS) with Hybrid Solar and RF Energy Harvesting  
  by Faisal Alanazi 
  Abstract: This contribution computes the throughput of intelligent omni-surfaces (IOS) when the source recovers power from the sun and radio frequency (RF) signals to be able to transmit data to two users Ut and Ur. Ut and Ur are located at the transmit and reflect spaces of IOS. We compute the harvested energy as well as its SINR. The harvesting process is optimised to maximise the throughput. We investigate the impact of harvesting time t and the number of reflectors of IOS denoted by N. IOS with N = 64 reflectors offers 48 dB versus N = 8. The results are valid for any size of the PV system. Hybrid solar and RF energy harvesting offers 32 dB gain and 3 dB gain versus RF and solar energy harvesting.
  Keywords: intelligent omni-surfaces; IOS; solar energy; RF energy harvesting; SINR; Rayleigh channels.
  DOI: 10.1504/IJAHUC.2024.10064827
   
  
• Deep Learning Based Detector for Downlink IM-NOMA Systems  
  by Issa Chihaoui, Mohamed Lassaad Ammari 
  Abstract: This work proposes a deep learning (DL) based detector for downlink index modulation non-orthogonal multiple access (IM-NOMA) systems, which we call DL-IM-NOMA. The proposed detector involves a multilayer fully connected deep neural network (DNN) block. The received signal is preprocessed using zero-forcing (ZF) and channel knowledge at the receiver before being fed into the DNN block. Architecture complexity and bit error rate (BER) performance of DL-IM-NOMA have been presented for several configurations, and compared to these of the suboptimal log-likelihood ratio (LLR) detector. Simulation results show that DL-IM-NOMA outperforms LLR detector for high signal-to-noise ratios (SNRs) with lower complexity. It has been shown that DL-IM-NOMA avoids BER floor occurred at strong users using LLR detector. In contrast, for low SNRs, DL-IM-NOMA causes an insignificant performance loss, especially for systems using large constellation size.
  Keywords: Deep learning; NOMA; subcarrier index modulation; OFDM.
  DOI: 10.1504/IJAHUC.2024.10064906
   
  
• A Novel Asanoha-Inspired Method for Optimal Path Planning of Multiple Mobile Sinks  
  by Amit Kumar Keshari, Kumar Nitesh, Bhaskar Karn 
  Abstract: In wireless sensor networks (WSNs), mobile sinks (MSs) act as information collectors, relieving the hot spot problem (HSP) and reducing energy usage in sensor nodes (SNs). To avoid hot spots problem, an MS travels inside a target region (TR) and stops at predetermined rendezvous points (RPs) to collect data from surrounding SNs. By leveraging multiple mobile sinks (MMSs), this study proposes a way for alleviating the hot spot problem to increase the network lifespan. The TR is divided into sub-regions (SRs) using the k-means clustering method, and each SR is assigned an MS. Within each SR, the Asanoha pattern is employed to configure likely RP prediction. The final set of RPs is chosen after optimisation of numerous criteria impacting MS performance. The proposed approach is tested using simulation and compared to existing methods in terms of performance metrics such as the number of RPs, route length, and average waiting time. These analyses are conducted across a range of SNs and communication ranges.
  Keywords: Wireless Sensor Networks; Sensor Nodes; multiple mobile sinks; Connectivity; Rendezvous Points; Trajectory Design; Asanoha pattern.
  DOI: 10.1504/IJAHUC.2024.10065078
   
  
• Data Trading, Power Control and Resource Allocation Algorithms for Metaverse Platform  
  by Sungwook Kim 
  Abstract: Edge computing (EC) has emerged as a cost-effective platform to enhance the computing capability of hardware-constrained IoT devices. Recently, EC-assisted Metaverse system is regarded as the next-generation internet paradigm that allows humans to play, work, and socialise in an alternative virtual world. With the help of ubiquitous wireless connections and powerful EC technologies, the Metaverse system effectively manages the interactions among system agents. In this study, we present a new intelligent Metaverse control scheme based on the reciprocal combination of auction, learning and bargaining methods. Specifically, McAfee double auction is applied to handle the collected data trading between service providers and IoT devices. In addition, learning algorithm and bargaining solution are used to provide a proper resource allocation problem for the devices' wireless communications. To explore the sequential interaction of system agents, we jointly design an integrated control scheme to strike an appropriate Metaverse performance balance. According to the synergy effect, our hybrid protocol is a novel method in the EC-assisted Metaverse infrastructure. Finally, extensive simulations demonstrate that our approach can lead to achieve a mutually desirable solution with a good balance between efficiency and fairness comparing with the currently published Metaverse system control schemes.
  Keywords: Metaverse system; learning algorithm; McAfee double auction model; status quo proportional bargaining; SQPB; resource allocation problem.
  DOI: 10.1504/IJAHUC.2024.10065094