Current Issue : April-June Volume : 2026 Issue Number : 2 Articles : 5 Articles
Large Language Models (LLMs) demonstrate potential in code generation capabilities, yet their applicability in autonomous vehicle control has not been sufficiently explored. This study verifies whether LLMs can generate executable MATLAB code for softwaredefined vehicle scenarios, comparing five models: GPT-4, Gemini 2.5 Pro, Claude Sonnet 4.0, CodeLlama-13B-Instruct, and StarCoder2. Thirteen standardised prompts were applied across three types of scenarios: programming-based driving scenarios, inertial sensor-based simulations, and vehicle parking scenarios. Multiple automated evaluation metrics—BLEU, ROUGE-L, ChrF, Spec-Compliance, and Runtime-Sanity—were used to assess code executability, accuracy, and completeness. The results showed GPT-4 achieved the highest score 0.54 in the parking scenario with an overall average score of 0.27, followed by Gemini 2.5 Pro as 0.26. Commercial models demonstrated over 60% execution success rates across all scenarios, whereas open-source models like CodeLlama and StarCoder2 were limited to under 20%. Furthermore, the parking scenario yielded the lowest average score of 0.19, confirming that complex tasks involving sensor synchronisation and trajectory control represent a common limitation across all models. This study presents a new benchmark for quantitatively evaluating the quality of SDV control code generated by LLMs, empirically demonstrating that prompt design and task complexity critically influence model reliability and real-world applicability....
1. Passive acoustic monitoring (PAM) coupled with artificial intelligence (AI) is becoming an essential tool for biodiversity monitoring. Traditional PAM systems require manual data offloading and impose substantial demands on data storage and computing infrastructure. The combination of on-device AI processing and network connectivity enables to analyse data locally and transmitting only relevant information, greatly reducing the volume of data requiring storage. However, programming these devices for robust operation is challenging, requiring expertise in embedded systems and software engineering. Despite the increase in AI models for bioacoustics, their full potential remains unrealised without accessible tools to deploy and configure them to meet specific monitoring goals. 2. To address this challenge, we develop acoupi, an open-source Python framework that simplifies the creation and deployment of smart bioacoustic devices. acoupi integrates audio recording, AI data processing, data management and real-time wireless messaging into a unified and configurable framework. By modularising key elements of the bioacoustic monitoring workflow, acoupi allows users to easily customise, extend or select specific components to fit their unique monitoring needs. 3. We demonstrate the flexibility of acoupi by integrating two bioacoustic classifiers: BirdNET, for the classification of bird species and BatDetect2, for the classification of UK bat species. We test the reliability of acoupi over several months, deploying two acoupi-powered devices in a UK urban park. 4. acoupi can be deployed on low-cost hardware such as the Raspberry Pi (RPi) and can be customised for various applications. acoupi standardised framework and simplified tools facilitate the adoption of AI-powered monitoring systems for researchers and conservationists....
The integration of artificial intelligence and machine learning techniques into software engineering practices has fundamentally transformed how we approach software quality assurance, system reliability, and information integrity. This paper surveys recent advances in AI-driven methodologies that address critical challenges in modern software systems, including false information detection in large language models, intelligent agricultural systems, distributed database optimization, and automated software testing. We examine how causal inference, graph neural networks, and multi-modal learning frameworks contribute to building more robust and reliable software systems. Our analysis demonstrates that combining domain-specific knowledge with advanced AI techniques enables the development of smart systems capable of addressing complex software engineering challenges across diverse application domains. This survey highlights emerging trends and provides insights into future research directions for AI-enhanced software engineering....
Software-defined Networking (SDN) has immense potential for network security due to its centralized control and programmability. However, this concentration provides an attractive attack vector for Distributed Denial-of-Service (DDoS), particularly in small and medium-sized enterprises (SMEs) with limited budget and network security resources. This study presents a systematic review of the articles reporting SDN-based DDoS detection and mitigation, focusing on SMEs. Querying eight major databases (2020–2025) resulted in 59 articles (14 reviews, 45 experimental). Two distinct models emerged: (i) lightweight and efficient models and (ii) high-accuracy hybrid deep learning models, with lower resource efficiency. These models were predominantly validated through simulations, raising concerns around their overfitting as SME traffic is heterogeneous and bursty. Mitigation of the attacks leveraged the programmability of SDN but has been rarely evaluated alongside detection models and almost never in live SDN-SME settings. This study’s findings highlighted a lightweight screening solution at the network edge, which is resource-aware and employs a minimal trigger interface to the controller for mitigation rule insertion. This conceptual design aligns well with the constraints of SMEs by minimising the computational load on the central controller while enabling an efficient and rapid response to network security....
This paper presents METAS VNA Tools Version 2.9.0, a metrology software suite designed to support the digital traceability chain in vector network analyzer measurements. Built on the METAS UncLib Version 2.9.0 uncertainty engine, the software enables rigorous modeling of the entire measurement process and comprehensive uncertainty evaluation. By encapsulating values, dependencies, and sensitivities in structured uncertainty objects, the software ensures that traceability and correlation information are preserved and propagated throughout complex calibration chains. This approach allows for seamless, modular uncertainty evaluation and supports the generation of digitally signed calibration certificates with embedded calibration data. The methodology enhances transparency, reproducibility, and interoperability, aligning with the goals of digital transformation in metrology. VNA Tools thus provides a robust foundation for implementing traceable, data-driven workflows across all levels of the metrological infrastructure....
Loading....