Time-sensitive networking (TSN) technologies have garnered attention for supporting time-sensitive communication services, with recent interest extending to the wireless domain. However, adapting TSN to wireless areas faces challenges due to the competitive channel utilization in IEEE 802.11, necessitating exclusive channels for low-latency services. Additionally, traditional TSN scheduling algorithms may cause significant transmission delays due to dynamic wireless characteristics, which must be addressed. This paper proposes a wireless TSN model of IEEE 802.11 networks for the exclusive channel access and a novel time-sensitive traffic scheduler, named the wireless intelligent scheduler (WISE), based on deep reinforcement learning. We designed a deep reinforcement learning (DRL) framework to learn the repetitive transmission patterns of time-sensitive traffic and address potential latency issues from changing wireless conditions. Within this framework, we identified the most suitable DRL model, presenting the WISE algorithm with the best performance. Experimental results indicate that the proposed mechanisms meet up to 99.9% under the various wireless communication scenarios. In addition, they show that the processing delay is successfully limited within the specific time requirements and the scalability of TSN streams is guaranteed by the proposed mechanisms.

With the difficulty of collecting desirable training data due to the heterogeneities of IoT sensors in various buildings and the scarcity of fire events, it is time consuming and expensive to apply data-driven deep learning approaches to fire detection systems in specific building environments. Simulation-based learning has been actively researched to mitigate data scarcity problems by reproducing potential fire events. Since simulation-based learning mainly depends on synthetic training data, trained deep learning models may generate erroneous predictions in real-world scenarios that are unlike any of the training samples. In this paper, we propose a trustworthy building fire detection framework based on a multioutput encoder-decoder network, named MEDNet, which is designed for the practical usage of simulation-based learning in building fire detection. The fundamental steps of our approach are (1) modeling and simulating fire events to create realistic synthetic data that reflect data from actual buildings, (2) predicting a fire event and dissimilarities between real input data and synthetic training data based on the trained MEDNet model, and (3) operating a switching mechanism to use a knowledge-based method that does not depend on synthetic training data when dissimilarities exist. Finally, we perform simulation experiments based on a real building compartment where the proposed framework is compared with conventional time-series classification networks on various evaluation datasets. The proposed framework is trustworthy in practical usage because MEDNet with a switching mechanism achieves a 36.65% higher F1-score than conventional time-series classification networks and generates false-positive predictions lower than 0.02% even in unpredictable scenarios.

Deep learning technologies, due to their advanced pattern extraction and recognition of high-dimensional data, have been widely adopted into multisensor-based fire detection systems. Since deep learning approaches can generate erroneous predictions due to incomplete training datasets, a retraining process over unseen observations is needed. However, storing a large amount of data from continuous multisensor streams and labeling them to create a retraining dataset are costly and time-consuming. In this paper, we propose an active learning framework based on an informative experience memory that is populated with meaningful retraining data by assessing the uncertainty of the data. In the proposed framework, the deep learning model predicts fire occurrence and estimates model uncertainty by taking advantage of a Bayesian neural network using Monte Carlo dropout. By storing only higher uncertain data points into the fixed-size informative experience memory and querying them to the system managers, the storage and labeling costs are minimized while improving performance. To evaluate our active learning framework with different neural network structures, we develop three Bayesian neural networks based on conventional classification networks, including the feedforward neural network, fully convolutional network, and long short-term memory. We further investigate various uncertainty assessment scoring methods for classification tasks such as entropy, BALD, variation ratios, and mean STD. Experiments on a real dataset show that the Bayesian FCN using the BALD assessment method has the highest performance gain with an F1 score of 0.95, with an improvement of 24% using only 700 data points.

The explosive demand for wireless communications has intensified the complexity of spectrum dynamics, particularly within unlicensed bands. To promote efficient spectrum utilization and minimize interference during communication, spectrum sensing needs to evolve to a stage capable of detecting multidimensional spectrum states. Signal identification, which identifies each device’s signal source, is a potent method for deriving the spectrum usage characteristics of wireless devices. However, most existing signal identification methods mainly focus on signal classification or modulation classification, thus offering limited spectrum information. In this paper, we propose DSINet, a multitask learning-based deep signal identification network for advanced spectrum sensing systems. DSINet addresses the deep signal identification problem, which involves not only classifying signals but also deriving the spectrum usage characteristics of signals across various spectrum dimensions, including time, frequency, power, and code. Comparative analyses reveal that DSINet outperforms existing shallow signal identification models, with performance improvements of 3.3% for signal classification, 3.3% for hall detection, and 5.7% for modulation classification. In addition, DSINet solves four different tasks with a 65.5% smaller model size and 230% improved computational performance compared to single-task learning model sets, providing meaningful results in terms of practical use.

The proliferation of IoT using heterogeneous wireless technologies within the unlicensed spectrum has intensified cross-technology interference (CTI) in wireless local area networks (WLANs). As WLANs increasingly adopt time-triggered transmission methods to support real-time services, this interference affects throughput, packet loss, and latency. This paper presents a CTI-aware rate adaptation framework designed to mitigate interference in WLANs without direct coordination with heterogeneous wireless devices. The framework includes a CTI identification model and CTI-aware rate selection algorithms. Leveraging short-time Fourier transform, the identification model captures the time–frequency–power characteristics of CTI signals, enabling the estimation of the average power of various heterogeneous wireless technologies employed by interfering devices. The rate selection algorithms predict CTI occurrence times and adjust the transmission rate accordingly, enhancing the performance of existing explicit and implicit interference mitigation methods. Experimental results demonstrated that the lightweight CTI identification model accurately estimated the average power of each type with an error margin of ±1.414 dBm, achieving this in under 1 ms on the target hardware. Additionally, applying the proposed framework to explicit interference mitigation enhanced goodput by 20.67%, reduced packet error rate by 2.38%, and decreased the probability of packets exceeding 1 ms latency by 0.932% compared to conventional methods.