Djamel Sadok

Marrone Silvério Melo Dantas, Iago Richard Rodrigues, Assis Tiago Oliveira Filho et al.

IoT devices suffer from resource limitations, such as processor, RAM, and disc storage. These limitations become more evident when handling demanding applications, such as deep learning, well-known for their heavy computational requirements. A case in point is robot pose estimation, an application that predicts the critical points of the desired image object. One way to mitigate processing and storage problems is compressing that deep learning application. This paper proposes a new CNN for the pose estimation while applying the compression techniques of pruning and quantization to reduce his demands and improve the response time. While the pruning process reduces the total number of parameters required for inference, quantization decreases the precision of the floating-point. We run the approach using a pose estimation task for a robotic arm and compare the results in a high-end device and a constrained device. As metrics, we consider the number of Floating-point Operations Per Second(FLOPS), the total of mathematical computations, the calculation of parameters, the inference time, and the number of video frames processed per second. In addition, we undertake a qualitative evaluation where we compare the output image predicted for each pruned network with the corresponding original one. We reduce the originally proposed network to a 70% pruning rate, implying an 88.86% reduction in parameters, 94.45% reduction in FLOPS, and for the disc storage, we reduced the requirement in 70% while increasing error by a mere $1\%$. With regard input image processing, this metric increases from 11.71 FPS to 41.9 FPS for the Desktop case. When using the constrained device, image processing augmented from 2.86 FPS to 10.04 FPS. The higher processing rate of image frames achieved by the proposed approach allows a much shorter response time.

Maria Eduarda Veras, Eduardo Freitas, Assis T. de Oliveira Filho et al.

The demand for ultra-low latency in modern applications, such as cloud gaming and augmented reality, has exposed the limitations of traditional congestion control algorithms regarding bufferbloat. The Low Latency, Low Loss, and Scalable Throughput (L4S) architecture addresses this challenge by combining scalable congestion controls, such as TCP Prague, low-latency queue management with prioritization, and Accurate ECN (AccECN) feedback. Although Linux kernel implementations exist, the research community lacks a complete, high-fidelity model within the Network Simulator 3 (ns-3) for reproducible experiments. This paper presents an implementation of end-host protocols for the L4S architecture in ns-3, focusing on the porting of TCP Prague from the Linux kernel (v6.12) and the integration of AccECN signaling. Significant engineering challenges regarding the adaptation of kernel logic are detailed, particularly the reconciliation of Linux's packet-based arithmetic with ns-3's byte-based architecture for window management and pacing. Simulation results demonstrate that the proposed model faithfully reproduces the congestion response behaviors observed in real-world testbed scenarios, validating the platform's accuracy. Consequently, this work provides the community with a validated toolset for complex L4S performance evaluations in controlled environments.

Pedro R. X. Carmo, Igor de Moura, Assis T. de Oliveira Filho et al.

Modern vehicles are increasingly connected, and in this context, automotive Ethernet is one of the technologies that promise to provide the necessary infrastructure for intra-vehicle communication. However, these systems are subject to attacks that can compromise safety, including flow injection attacks. Deep Learning-based Intrusion Detection Systems (IDS) are often designed to combat this problem, but they require expensive hardware to run in real time. In this work, we propose to evaluate and apply fast neural network inference techniques like Distilling and Prunning for deploying IDS models on low-cost platforms in real time. The results show that these techniques can achieve intrusion detection times of up to 727 μs using a Raspberry Pi 4, with AUCROC values of 0.9890.