Fast and highly accurate fire detection algorithms are crucial for production and daily life. However, detecting fires in the early stages is challenging due to the lack of distinct features and fixed shapes of flames and smoke. To address this issue, we propose a fire detection algorithm, DCGC(Dual Channel Group Convolution)-YOLO, based on an improved YOLOv5 model. The DCGC-YOLO model introduces a new layer structure and anchor algorithm to optimize original YOLOv5 model. Firstly, we introduce a Cross Stage Partial (CSP) structure with a cascade of large convolutional kernels in the bottleneck layer. This structure increases network’s receptive field, enhances feature extraction capabilities, and employs a channel cleansing mechanism that combines various channels separated by group convolution, promoting information exchange in the channel dimension and enabling better information encoding. Next, we integrate the Effective Squeeze and Extraction (eSE) mechanism into the new layer structure, enhancing the model’s ability for long-range modeling and focusing more on target areas. Finally, we utilize an Intersection over Union (IoU)-based anchor generation algorithm to adjust the anchor sizes on custom fire dataset, enhancing the model’s robustness and improving detection accuracy. Experimental results on our custom fire dataset demonstrate that the proposed DCGC-YOLO algorithm effectively detects targets with mAP of 41.1%, which is 2.9% higher than YOLOv5s, while reducing network parameters and computational complexity. To further validate the effectiveness of our proposed algorithm, we conduct experiments on the COCO2017 dataset. The results show that DCGC-YOLO achieves mAP of 38.9%, demonstrating strong generalization and competitiveness compared to state-of-the-art detectors.
Being fascinated by the small‐area radiation, the operators with strong capability of resource planning tend to superpose the device‐to‐device (D2D) communications upon licensed frequency bands for effectively excavating wireless resources. However, this inevitably causes mutual interference in such a hybrid network. A solution to address the interference issue is to appropriately choose reusing resources for D2D communications combined with power control, while ensuring the quality of service (QoS) of prioritised cellular users. The authors formulate the resource sharing strategy as a joint channel selection and power control optimisation under general sum power/QoS constraints. An alternating optimising method is proposed to deal with such a non‐convex problem. The performance of proposed resource sharing algorithm is illustrated by simulation results.
This letter presents a YOLO-based detection model called DS-YOLO (Diverse Branch Block and SlimNeck), which enables fast and accurate identification of smoke and fire. Firstly, DP-ELAN module in the backbone is constructed based on Diverse Branch Block (DBB) and Partial Convolution (PConv). It is possible to enhance detection capability while reducing the number of parameters. Secondly, utilizing the SlimNeck based on Ghost Shuffle Convolution (GSConv) to reduce model complexity. Finally, we use Normalized Gaussian Wasserstein Distance (NWD) instead of the standard IoU to improve the detection capability for small-sized objects.The experimental results on our custom smoke and fire dataset indicate that the proposed DS-YOLO achieves a mAP of 70.1%, while reducing computational complexity. This represents an 1.3% improvement over the baseline model.
As the problem of information overload becomes more severe, it has become increasingly difficult for users to browse news that they are interested in. News recommendation is an effective method to alleviate this problem. In news recommendation, accurately obtaining rich semantic news representations and modeling user’s historical interests are long-standing challenges. Based on these two issues, we propose a News recommendation approach with Word-related joint Topic prediction(NWT). The core of our approach is the topic perceptron and the news encoder that emphasizes word relevance. The topic information and word relevance information are fused through joint training, enhancing the abilities of topic prediction and news representation. In the news encoder, we employ multi-head self-attention network to emphasize word relevance, capturing the semantic relationships between any two words in the news title and enhancing the representation of the news. In the topic perceptron, we use the enhanced word relevance-based news representation as input and learn the topic information in the news by assigning weights to different topics. During joint training, the overall news recommendation module obtains the news topic information. Additionally, the user encoder learns the user’s topic preferences and utilizes attention networks to highlight news articles that are more representative of the user’s interests, thereby acquiring a more accurate user representation. We conduct extensive experiments on the MIND dataset, and the results demonstrate that NWT outperforms most existing baseline methods across various evaluation metrics.
Emerging terahertz (THz) communication will become the main way of wireless communication in the future, it will also use hybrid precoding which combines analog and digital hybrid precoding like millimeter wave communication systems, to reduce the system energy consumption. Indoor as a main scenario for communication, to better meet the demands for data transmission. In this paper, an improved hybrid precoding method based on successive interference cancelation and singular value decomposition (SIC-SVD) is proposed for indoor Sub-THz communication system with sub-connected structures. However, the introduction of phase shifters imposes constant modulus constraints, making the problem of hybrid precoding difficult to solve directly. The proposed method first decomposes the optimization problem of achievable rates for the system, based on the special structure of sub-connected, into a series of equivalent sub-problems. The algorithm first designs the analog precoding matrix according to the SIC method, after obtaining the designed analog precoding matrix, the algorithm defines an equivalent channel and employs singular value decomposition (SVD) to obtain the digital precoding matrix. The simulation results show that compared with the existing method based on SIC, the proposed algorithm has slightly higher achievable rate and is closer to the optimal full-digital precoding method. At the same time, this paper also presents a simple analysis of the proposed algorithm for imperfect channel state information (CSI) with different accuracy. Simulation results show that the spectral efficiency is also very near to the perfect CSI case when the accuracy of the CSI is 0.9.
12 June 2025 in IEEE Wireless Communications Letters
Millimeter-wave (mmWave) integrated sensing and communications (ISAC) with hybrid architecture has been seen as a potential candidate for next generation wireless systems. However, simultaneously achieving high-resolution sensing and high throughout communication under multipath-sensitive signal interference-plus-noise-ratio (SINR) constraints remains challenging. To address this issue, we propose a hybrid beamforming solution with rank-1 optimization for balancing sensing and communication, in which sensing resolution is maximized by minimizing Cramér-Rao bound (CRB) with multi-users’ rate requirement represented by corresponding SINRs. Furthermore, for hybrid beamforming, the aforementioned problem is addressed by jointly designing digital and analog beamformers, where digital beamforming is solved with semi-definite relaxation (SDR) and second-order cone programming (SOCP), and analog beamformer is obtained by employing singular value decomposition (SVD) and successive interference cancellation (SIC), respectively. Simulation results highlight the effectiveness of the proposed algorithm.