Image recognition is an important task in computer vision with broad applications. In recent years, with the advent of deep learning, lightweight convolutional neural network (CNN) has brought new opportunities for image recognition, which allows high-performance recognition algorithms to run on resource-constrained devices with strong representation and generalization capabilities. This paper first presents an overview of several classical lightweight CNN models. Then, a comprehensive review is provided on recent image recognition techniques using lightweight CNN. According to the strategies applied to optimize image recognition performance, existing methods are classified into three categories: (1) model compression, (2) optimization of lightweight network, and (3) combining Transformer with lightweight network. In addition, some representative methods are tested on three commonly used datasets for performance comparison. Finally, technical challenges and future research trends in this field are discussed.

Federated learning (FL) trains a global learning model by using a central server to collaborate with multiple decentralized clients. In a wireless network, the data transmission latency between a client and the FL server is substantially affected by signal quality dynamics and bandwidth allocation. FL clients require synchronized communication at each round to update their models simultaneously, which makes bandwidth allocation methods for conventional wireless tasks infeasible to use. Existing bandwidth allocation studies for FL mainly focused on allocating bandwidth of one bandwidth provider without cost. In this paper, we consider a more practical and challenging problem: how to assign the bandwidth to clients under multiple wireless providers to minimize the FL round length (i.e., the latency that FL finishes one round of model training and updating) with bandwidth capability and cost constraints? We propose a model that maps the problem into a new variant of the knapsack problem, called multi-dimensional max-min multiple knapsacks (MDM$^{\,3}$KP). Based on MDM$^{\,3}$KP, we create an iterative solution to find the client assignment and bandwidth allocation that minimizes the FL round length. Comprehensive simulation results show that the solution reduces the FL round length by up to 70.8% compared with other benchmarks.

New network architectures, such as the Internet of Things (IoT), 5G, and next-generation (NextG) cellular systems, put forward emerging challenges to the design of future wireless networks toward ultra-high data rate, massive data processing, smart designs, low-cost deployment, reliability and security in dynamic environments. As one of the most promising techniques today, artificial intelligence (AI) is advocated to enable a data-driven paradigm for wireless network design. In this paper, we are motivated to review existing AI techniques and their applications for the full wireless network protocol stack toward improving network performance and security. Our goal is to summarize the current motivation, challenges, and methodology of using AI to enhance wireless networking from the physical to the application layer, and shed light on creating new AI-enabled algorithms, mechanisms, protocols, and system designs for future data-driven wireless networking.

Tire pattern image classification is an important computer vision problem in pubic security, which can guide policeman to detect criminal cases. It remains challenge due to the small diversity within different classes. Generally, a tire pattern image classification system may require two characteristics: high accuracy and low computation. In this paper, we first assume that capturing rich feature representation will benefits tire classification and learning through a lightweight network will improve computing efficiency. We then propose a simple yet efficient two-stage training mechanism: 1) We learn a feature extractor using a V ariational Auto-Encoder framework constrained by contrastive learning, projecting images to latent space owing rich feature representation. 2) We train a single-layer linear classification network depend on the features extracted by the previous trained encoder. The Top-1 and Top-5 accuracy on tire pattern dataset is 89.8% and 96.6% respectively, validating the effectiveness of our strategy.