International Journal of Digital Electronics and Microprocessor Technology

Low-Latency Embedded AI Device for Vehicle-to-Vehicle Communication

Kukuchuku Shadrach — 2026-10-09

Vehicle-to-Vehicle (V2V) communication has emerged as a fundamental technology for enabling intelligent transportation systems, autonomous driving, and cooperative vehicular safety applications. However, conventional cloud-centric vehicular communication architectures often suffer from high communication latency, excessive bandwidth consumption, limited scalability, and delayed response times, which can adversely affect safety-critical driving operations. To address these challenges, this paper proposes a low-latency embedded Artificial Intelligence (AI) device for intelligent V2V communication that integrates embedded edge intelligence, real-time AI inference, adaptive wireless communication, cooperative perception, and lightweight cybersecurity mechanisms within a unified framework. The proposed architecture employs local sensor fusion and edge-based AI processing to perform real-time object detection, collision prediction, trajectory estimation, and intelligent decision-making directly within vehicles, thereby reducing dependence on remote cloud infrastructures. Furthermore, an adaptive V2V communication module utilizing intelligent packet prioritization and resource allocation is incorporated to enhance communication reliability and responsiveness in dynamic vehicular environments. The proposed system was evaluated using MATLAB simulations under varying vehicular mobility and communication conditions. Simulation results demonstrate an average end-to-end communication latency of approximately 12.3 ms, an average network throughput of 63.7 Mbps, and a packet delivery ratio of 97.4%, confirming the suitability of the framework for ultra-reliable low-latency vehicular communications. Additionally, the embedded AI framework achieved an average energy consumption of 1.9 J while reducing the collision risk score from approximately 0.9 during initial operation to below 0.25 under steady-state conditions. The system also maintained stable performance under increasing network load, supporting up to 28 active communication links without significant degradation in communication quality. These results demonstrate that the proposed embedded AI-enabled V2V architecture provides an effective, scalable, and energy-efficient solution for next-generation autonomous vehicles, smart transportation infrastructures, and future 5G/6G vehicular communication ecosystems.

Deep Learning for Early Plant Disease Detection: A Review and Deployment Framework for Lightweight CNNs (MobileNetV3) on Edge Devices

Kalla Venkata Subrahmanyam — 2026-11-05

Crop diseases threaten yield and food security, while timely expert diagnosis and reliable internet access remain limited in many farming communities. This narrative review examines lightweight convolutional neural networks (CNNs), with emphasis on MobileNetV3, and the model-compression and deployment practices relevant to early plant disease classification on edge devices. A targeted literature search and comparison of published studies, datasets, backbone specifications, and compression methods were used; numerical results are reported from the cited publications rather than from new experiments. The reviewed evidence shows that MobileNetV3-Small achieved approximately 99.50% test accuracy on the controlled PlantVillage dataset in one published study, retaining that accuracy after post-training quantization while reducing the reported parameter count from about 1.5 million to 0.93 million. Other benchmark studies report very high performance on PlantVillage but lower results on more realistic PlantDoc imagery, illustrating that dataset and evaluation conditions materially affect the apparent ranking of architectures. ImageNet accuracy and computational cost are useful for describing pretrained backbones, but they are not substitutes for crop- and disease-specific evaluation. The review therefore outlines a workflow covering field-aware data splitting, transfer learning, quantization, pruning, distillation and conversion to mobile runtimes, and recommends reporting per-class recall and macro-F1 alongside latency, memory, and energy on the target device. The study concludes that reliable deployment depends not only on compact architectures but also on diverse field data, independent field-level testing, uncertainty-aware predictions, and transparent on-device benchmarking. The main obstacle to field use is not model size but the gap between laboratory and field images. Diverse field data, robust evaluation, and explainability are the priority, enabling reliable offline diagnostics for resource-limited farming communities.

Design and Security Analysis of a Dynamic Key-Based Lightweight Encryption Algorithm for Resource-Constrained Embedded Systems

Neha Saravanan — 2026-07-01

This paper focuses on improving security in resource-constrained embedded systems and Internet of Things (IoT) networks by developing an efficient, lightweight encryption technique with low computational requirements. The proposed approach introduces a dynamic key-based encryption algorithm that combines XOR operations with an iterative key evolution process and a feedback mechanism to strengthen data security. The methodology involves implementing the encryption algorithm in the C programming language and evaluating its performance through statistical analysis using Python tools. Security assessment is carried out using various parameters such as entropy, histogram distribution, correlation coefficient, and avalanche effect to measure the effectiveness of the encryption process. The results demonstrate increased randomness in encrypted data and reduced correlation between plaintext and ciphertext, indicating stronger security characteristics. The main contribution of the work lies in integrating dynamic key evolution with lightweight encryption methods to provide improved protection while maintaining computational efficiency for embedded and IoT applications.