Abstract--This paper proposes a hardware-aware intrusion detection system (IDS) for Internet of Things (IoT) and Industrial IoT (IIoT) networks; it targets scenarios where classification is essential for fast, privacy-preserving, and resource-efficient threat detection. The goal is to optimize both tree-based machine learning (ML) models and compact deep neural networks (DNNs) within strict edge-device constraints. This allows for a fair comparison and reveals trade-offs between model families. We apply constrained grid search for tree-based classifiers and hardware-aware neural architecture search (HW-NAS) for 1D convolutional neural networks (1D-CNNs). Evaluation on the Edge-IIoTset benchmark shows that selected models meet tight flash, RAM, and compute limits: LightGBM achieves 95.3% accuracy using 75 KB flash and 1.2 K operations, while the HW-NAS-optimized CNN reaches 97.2% with 190 KB flash and 840 K floating-point operations (FLOPs). We deploy the full pipeline on a Raspberry Pi 3 B+, confirming that tree-based models operate within 30 ms and that CNNs remain suitable when accuracy outweighs latency. The widespread deployment of Internet of Things (IoT) systems has expanded the attack surface of modern networks, which now include critical infrastructure and operational environments vulnerable to advanced cyber threats [1], [2].

This study focuses on developing an experimental system for estimating communication channels in a multi-robot mobile system using software-defined radio (SDR) devices. The system consists of two mobile robots programmed for two scenarios: one where the robot remains stationary and another where it follows a predefined trajectory. Communication within the system is conducted through orthogonal frequency-division multiplexing (OFDM) to mitigate the effects of multipath propagation in indoor environments. The system's performance is evaluated using the bit error rate (BER). Connections related to robot motion and communication are implemented using Raspberry Pi 3 and BladeRF x115, respectively. The least squares (LS) technique is employed to estimate the channel with a bit error rate of approximately 10^(-2).

One of the coolest devices for technically minded musicians is the MOD-Duo from MOD-Devices. Not only is it a standalone audio processor with 300 built in audio/MIDI effects, but it also acts as a server that you can interact with through a browser to build any kind of pedalboard imaginable. The sheer number of possibilities on these devices (now upgraded to the MOD-DuoX and most recently MOD-Dwarf), is daunting, but it has the power to revolutionize how people make music. What does this have to do with neural networks? If you are familiar with my work on GuitarML, you know that I develop guitar plugins that use neural networks to mimic analog amplifiers and pedals.

This article looks at the unique challenges introduced by Edge computing for AI/ML workloads, which can have a negative impact on results. It applies available machine learning models to real-world Edge datasets, to show how these challenges can be overcome, while preserving accuracy in the dynamic nature of Edge environments. The field of machine learning has experienced an explosion of innovation over the past 10 years. Although its roots date back more than 70 years when Alan Turing devised the Turing Test, it has not matured significantly until recently. Two primary contributing factors are the exponential growth in both compute power and data that can be used for training. There is now enough data and compute power (some in specialized hardware like GPUs/FPGAs) that new, real-world problems are being solved every day with machine learning.

On November 13, 2015, Google had open-sourced TensorFlow, an end-to-end machine learning platform. Apart from marking five years of being one of the most popular machine learning frameworks, last week was even more significant as TensorFlow crossed the 160 million downloads. This article lists some interesting TensorFlow projects, in no particular order, which enthusiasts can try their hands on. This Handwritten Text Recognition can be implemented using TensorFlow. In this project, the system is trained on the IAM off-line dataset.

This is the 2nd part of a 3-part series on building and deploying a custom object detection model to a Raspberry Pi 3. To get caught up,I'd suggest reading part 1 here: Part 2 will be all about training our object detection network using Google Colab . First and foremost, before training, we'll dig into the network architecture we plan to use. EfficientDet is a neural network architecture for object detection. It's one of the TensorFlow object detection APIs from the various model zoos, like CenterNet, MobileNet, ResNet, and Fast R-CNN. EfficientDets are a family of object detection models that achieve state-of-the-art 55.1mAP (mean average precision) on COCO test-dev, while also being 4x -- 9x smaller and using 13x -- 42x fewer FLOPs than previous detectors.

Getting Started with Raspberry Pi even if you are a beginner, Deep Learning Basics, Object Detection Models - Pros and Cons of each CNN, Setup and Install Movidius Neural Compute Stick (NCS) SDK, Currently, the OpenVINO is available for Raspbian, so the NCS2 is already compatible with the Raspberry Pi, but this course is mainly for the Movidius (NCS version 1).

This article has been contributed by Lin Ma, Software Engineer and KVM Virtualization Specialist at SUSE. With this article, I would like to introduce you to my SUSE Hackweek 19 project. If you worked on similar projects or topics, or if you would like to exchange experiences, please feel free to reach out to me. As a Do-it-Yourself (DIY) enthusiast, I decided to have some fun during Hackweek with music and machine learning based on SUSE Linux Enterprise Server. Or, if you will, you could also think of my project as an Internet of Things (IoT) attempt based on SUSE products.

With the emergence of onboard vision processing for areas such as the internet of things (IoT), edge computing and autonomous robots, there is increasing demand for computationally efficient convolutional neural network (CNN) models to perform real-time object detection on resource constraints hardware devices. Tiny-YOLO is generally considered as one of the faster object detectors for low-end devices and is the basis for our work. Our experiments on this network have shown that Tiny-YOLO can achieve 0.14 frames per second(FPS) on the Raspberry Pi 3 B, which is too slow for soccer playing autonomous humanoid robots detecting goal and ball objects. In this paper we propose an adaptation to the YOLO CNN model named xYOLO, that can achieve object detection at a speed of 9.66 FPS on the Raspberry Pi 3 B. This is achieved by trading an acceptable amount of accuracy, making the network approximately 70 times faster than Tiny-YOLO. Greater inference speed-ups were also achieved on a desktop CPU and GPU. Additionally we contribute an annotated Darknet dataset for goal and ball detection.

This post is the first one of the series. The goal is to share our experience about how to leverage open-source resources to enable deep learning for objection detection on RPi3. As for the first one of the series, the post will talk about why running object detection on RPi3 is difficult. First, deep learning (or to be more specific, CNN) on Raspberry Pi is nothing new. Pete Warden had released DeepBelief SDK for image recognition in 2014 [1], and SqueezeNet [2] was another alternative released in 2015 which aimed to bring lighter solution for embedded systems.