Convolution Neural Networks (CNNs) are gaining ground in deep learning and Artificial Intelligence (AI) domains, and they can benefit from rapid prototyping in order to produce efficient and low-power hardware designs. The inference process of a Deep Neural Network (DNN) is considered a computationally intensive process that requires hardware accelerators to operate in real-world scenarios due to the low latency requirements of real-time applications. As a result, High-Level Synthesis (HLS) tools are gaining popularity since they provide attractive ways to reduce design time complexity directly in register transfer level (RTL). In this paper, we implement a MobileNetV2 model using a state-of-the-art HLS tool in order to conduct a design space exploration and to provide insights on complex hardware designs which are tailored for DNN inference. Our goal is to combine design methodologies with sparsification techniques to produce hardware accelerators that achieve comparable error metrics within the same order of magnitude with the corresponding state-of-the-art systems while also significantly reducing the inference latency and resource utilization. Toward this end, we apply sparse matrix techniques on a MobileNetV2 model for efficient data representation, and we evaluate our designs in two different weight pruning approaches. Experimental results are evaluated with respect to the CIFAR-10 data set using several different design methodologies in order to fully explore their effects on the performance of the model under examination.

In this article, we provide a set of federated learning (FL) protocols for future Internet architectures, which integrate the edge computing with the Internet of Things (IoT) known as “IoT edge computing.” The proposed protocols aim to the efficient implementation of the FL, i.e., distributed intelligence, in future IoT networks, where edge computing will leverage the overall procedure at the edge of the network. We first provide a list of application requirements for such an FL implementation, which result in the architecture of constrained and nonconstrained IoT devices based on a set of Internet engineering task force (IETF) standards. The FL protocols consist of three stages as follows: 1) initial configuration; 2) distributed training; and 3) cloud updates. The specified FL protocols are tested using an experimental IoT platform, which is used to obtain experimental results that provide the performance evaluation of the FL protocols in terms of accuracy, time, and latency. We propose those FL protocols for the next-generation Internet (NGI), where IoT, edge computing, and FL will be blended efficiently for future Internet applications.

The early detection of plant diseases is critical to improving agricultural productivity and ensuring food security. However, conventional centralized deep learning approaches are often unsuitable for large-scale agricultural deployments, as they rely on continuous data transmission to cloud servers and require high computational resources that are impractical for Internet of Things (IoT)-based field environments. In this article, we present a distributed deep learning framework based on Federated Learning (FL) for the diagnosis of plant diseases in IoT sensor networks. The proposed architecture integrates multiple IoT nodes and an edge computing node that collaboratively train an EfficientNet B0 model using the Federated Averaging (FedAvg) algorithm without transferring local data. Two training pipelines are evaluated: a standard single-model pipeline and a hierarchical pipeline that combines a crop classifier with crop-specific disease models. Experimental results on a multicrop leaf image dataset under realistic augmentation scenarios demonstrate that the hierarchical FL approach improves per-crop classification accuracy and robustness to environmental variations, while the standard pipeline offers lower latency and energy consumption.

Mobile multimedia Internet services, such as mobile virtual reality (VR), request for advanced quality of experience (QoE) provision. In the future Internet, such a QoE provision will be leveraged through edge computing, which can be facilitated through advanced programmable frameworks, such as service function chaining (SFC). To this direction, we provide in this work, the design, development, and experimentation of a QoE-aware edge computing framework using SFC. In particular, the QoE-aware edge computing is provided by a network function virtualization-enabled SFC architecture. The SFC architecture provides adaptive media services through programmable service functions (SFs). Our solution provides multimedia dynamic adaptive streaming HTTP services with QoE under certain network conditions. The proposed QoE-aware edge computing is guaranteed with both data and computation offloading techniques, including a required QoE-aware edge resource management. Specific use cases are designed in order to test our solution that shows better QoE in comparison with the baseline use case, where neither data nor computational offloading is considered.