2024. 4th Issue
Volume XVI, Number 4
MESSAGE FROM THE EDITOR-IN-CHIEF
Pal Varga
Advances in Speech Recognition, Musical Anamnesis, Inter-ISP communication, MEC offloading and Software-Defined Radio 
I nfocommunications is a vast domain, and our journal tries to capture the latest advances. 2024 was a turbulent year in many ways – full of technological advances and societal challenges. Regarding the Infocommunications Journal, 2024 was a very special year, as we handled a record number of 265 papers that has been submitted. This is the last issue, vith quite diverse topics – from speech recognition and transfer learning, through enhancing QoS of IoT devices by MEC offloading, Software-Defined Radios and inter-ISP communications. Let us briefly capture the topics in the December 2024 issue of the Infocommunications Journal.
PAPERS FROM OPEN CALL
Yan Meng and Péter Mihajlik
Assessing the Efficacy of Adapters in Cross-Language Transfer Learning For Low-Resource Automatic Speech Recognition
In recent years, the application of adapter modules in large language models proved to be successful in reducing computing and memory costs during fine-tuning. In our paper, we apply adapters to the field of automatic speech recognition. Specifically, we add adapters to different pre-trained speech recognition models to evaluate their efficiency in cross-language transfer learning. In this study, the evaluations are extended to GPU memory consumption, training duration, and recognition accuracy. By comparing the effects of adapters added to different models, we further explore the impact of whether the foundational model was (pre-) trained in the target language.
Reference
DOI: 10.36244/ICJ.2024.4.1
Download
Marouane Myyara, Oussama Lagnfdi, Anouar Darif, and Abderrazak Farchane
Ehancing QoS for IoT Devices through Heuristics-based Computation Offloading in Multi-access Edge Computing
Multi-access Edge Computing (MEC) networks, particularly with the advent of 5G, aim to reduce latency and increase speed to meet the demands of resource-intensive applications in the Internet of Things (IoT), such as private wireless networks, online gaming, industry, and remote healthcare. These applications require guaranteed performance. However, while Quality of Service (QoS) management is well established in the Cloud, improving it remains a challenge in MEC environments. This study addresses this challenge by proposing heuristic computation offloading algorithms for IoT-intensive devices in MEC networks. These algorithms aim to minimize service time while maximizing the QoS, taking into account tasks and resource characteristics to determine the optimal execution location for IoT device applications. We evaluated our approach using the EdgeCloudSim simulator, and the results demonstrate its superiority over existing solutions. Our approach significantly improves QoS by reducing the service time of IoT application tasks. This research fills a gap in efficient QoS improvement and contributes to advances in computation offloading strategies in MEC environments. It paves the way for enhanced performance of IoT applications in these networks.
Reference
DOI: 10.36244/ICJ.2024.4.2
Download
Hari Krishnan S, and Mr. S. Sadiqvali
Enhancing Signal Processing Efficiency in Software-Defined Radio Using Distributed Arithmetic and Look-Up Table-Based FIR Filters
To meet the requirements of the wireless communication industry, digital communication systems require increasingly advanced coding and modulation technologies. Software-Defined Radio enables these advanced ideas to be easily adopted by such systems. The Finite Impulse Response filter is frequently used in wireless communication to pre-process detected signals to reduce noise by utilizing delay elements, multipliers, and adders. Traditional multiplier-based finite impulse response filter designs result in hardwareintensive multipliers that use a lot of space and energy and pose poor calculation speeds and low performance in throughput and latency. To overcome the existing issues, a novel Distributed Arithmetic with a Look Up Table-based FIR filter is proposed, which reduces the Bit Error Rate and latency and improves throughput by optimizing the channel equalizer as a crucial part of Software Defined Radio applications. Further, a key feature named the decimation factor is incorporated to dynamically alter the filter's output frequency response without altering the filter coefficients. Moreover, the worst-case critical route latency of partial product accumulation is reduced using a highly adaptable Parallel Prefix Adder. Additionally, the finite impulse response filters are integrated to decrease the number of Look-Up Tables, thereby saving time and memory. It also investigates the filter efficiency using faster multipliers and adders and validates it on an Artix-7 FPGA. As a result, the proposed model improved the filter’s performance over the other existing designs by achieving an operating speed of 260 MHz, delay of 190 ps, power dissipation of 1 m mW and throughput of 938.12 Mbps with the number of Look-Up Tables being 16504.
Reference
DOI: 10.36244/ICJ.2024.4.3
Download
Hamid Garmani, Mohamed El Amrani, Driss Ait Omar, Mohamed Baslam, and Hicham Zougagh
Analysis of Interactions Among ISPs in Information Centric Network with Advertiser Involvement
In response to the escalating volume of Internet traffic, scalability challenges have emerged in content delivery. Information-Centric Networking (ICN) has emerged as a solution to accommodate this surge in traffic by leveraging caching. Collaborative caching within ICN is pivotal for enhancing network performance and reducing content distribution costs. However, current pricing strategies on the Internet do not align with ICN interconnection incentives. This paper delves into the economic incentive caching of free content among various types of ICN providers, including advertisers and Internet service providers (ISPs). Specifically, we employ game-theoretic models to analyze the interaction between providers within an ICN framework, where providers are incentivized to cache and share content. Content popularity is modeled using a generalized Zipf distribution. We formulate the interactions among ISPs as a non-cooperative game and, through mathematical analysis, establish the existence and uniqueness of the Nash equilibrium under certain conditions. Additionally, we propose an iterative and distributed algorithm based on best response dynamics to converge towards the equilibrium point. Numerical simulations demonstrate that our proposed game models yield a winwin solution, showcasing the effectiveness of our approach in incentivizing collaborative caching of free content within ICN.
Reference
DOI: 10.36244/ICJ.2024.4.4
Download
Fabian Simmank, Katarzyna Grebosz-Haring, Thomas Ballhausen, Christian Thomay, Martin Biallas, and Markus Tauber
Towards Automated Musical Anamnesis for Music-based Intervention in Dementia Patients
Dementia is a neurodegenerative disease affecting millions worldwide, leading to cognitive decline and difficulties in daily activities. Music-based interventions offer a promising, cost-effective, non-pharmacological approach to improving quality of life for people with dementia. However, understanding both preferred and familiar music, as well as individual music affinity, is crucial to avoid overstimulation and ensure meaningful engagement. Developing a protocol for musical anamnesis, which gathers a patient’s musical history and hearing health, demands significant manual effort and expertise, limiting its scalability. An automated approach could enhance the sustainability of music-based therapy by reducing therapist time while maintaining relevance and preference evaluation. Here, we introduce Automated Musical Anamnesis (AMA), a personalized, scalable intervention combining interdisciplinary methods to support people with dementia
Reference
DOI: 10.36244/ICJ.2024.4.5
Download
Technical Co-Sponsors
|
|
| |
Supporter
|
National Cooperation Fund, Hungary |
