Search

Long-Range networks are increasingly used in smart spaces due to their ability to provide longer communication range while consuming low energy. To facilitate communication among different Long-Range nodes, a gateway is used. For instance, in smart buildings such as airports, railway stations, indoor stadiums, and auditoriums, sensory data from multiple sites are transferred to a base station through a Long-Range gateway. However, when multiple nodes transmit data simultaneously to the gateway, it generates network interference, especially in high-ceiling smart buildings where Long-Range nodes with sensors are attached to monitor the building’s health. In this paper, we present a new method for identifying interfering Long-Range Nodes (LNs) in high-ceiling smart buildings using a classification model. Our approach involves gathering and analyzing network parameters, such as signal-to-noise ratio and received signal strength indicator, from the signals to extract features that the classifier uses for interference classification. The approach categorizes interference based on the number of interfering LNs, with each class representing a distinct number of interfering LNs. We also introduce a push-based mechanism to detect and adjust the power levels of faulty LNs, reducing interference. Our method is cost-effective as it is hardware-independent, making it feasible to implement on the LG platform. Finally, we present a dataset of network interference generated by varying the number of nodes, obstacles, and other parameters. We train the model on the generated dataset and evaluate its effectiveness using a test bed. The experimental results demonstrate that the approach can successfully identify interference nodes in a complex network.

Scene text spotting aims at simultaneously localizing and recognizing text instances, symbols, and logos in natural scene images. Scene text detection and recognition approaches have received immense attention in computer vision research community. The presence of partial occlusion or truncation artifact due to the cluttered background of scene images creates an obstacle in perceiving the text instances, which makes the process of spotting very complex. In this paper, we propose a light-weight scene text spotter that can address the issue of cluttered environment of scene images. It is an end-to-end trainable deep neural network that uses local part information, global structural features, and context cue information of oriented region proposals for spotting text instances. It helps to localize in scene images with background clutters, where partially occluded text parts, truncation artifacts, and perspective distortions are present. We mitigate the problem of misclassification caused by inter-class interference by exploring inter-class separability and intra-class compactness. We also incorporate multi-language character segmentation and word-level recognition in a light-weight recognition module. We have used six publicly available benchmark datasets in different smart devices to illustrate the efficacy of the network.

Wireless Sensor Networks (WSNs) consist of several battery powered sensor nodes. The sensing coverage of the Field of Interest (FoI) is an important function of the sensor nodes in connected WSNs. A FoI is said to be covered if each point in the FoI is monitored by at least one sensor node. Due to small size, battery power supply, simple architecture, and light weight Operating System of the sensor nodes, maintaining the desired coverage of the FoI consists various issues and challenges in a connected WSN. This paper surveys the existing work done to address various issues and challenges for solving the coverage and connectivity problems in WSNs. Our discussion emphasis on sensing models, classification of coverage, research issues in WSNs and practical challenges in deployment of WSNs. We review a brief but complete overview of the various solutions of coverage problems in connected WSNs and describing insights into issues and challenges for research in this area.