The intelligent use of deep learning (DL) techniques can assist in overcoming noise and uncertainty during fingerprinting-based localization. With the rise in the available computational power on mobile devices, it is now possible to employ DL techniques, such as convolutional neural networks (CNNs), for smartphones. In this paper, we introduce a CNN model based on received signal strength indicator (RSSI) fingerprint datasets and compare it with different CNN application models, such as AlexNet, ResNet, ZFNet, Inception v3, and MobileNet v2, for indoor localization. The experimental results show that the proposed CNN model can achieve a test accuracy of 94.45% and an average location error as low as 1.44 m. Therefore, our CNN model outperforms conventional CNN applications for RSSI-based indoor positioning.

In this paper, we propose two data augmentation schemes for deep learning architecture that can be used to directly estimate user location in an indoor environment using mobile phone tracking and electronic fingerprints based on reference points and access points. Using a pretrained model, the deep learning approach can significantly reduce data collection time, while the runtime is also significantly reduced. Numerical results indicate that an augmented training database containing seven days’ worth of measurements is sufficient to generate acceptable performance using a pretrained model. Experimental results find that the proposed augmentation schemes can achieve a test accuracy of 89.73% and an average location error that is as low as 2.54 m. Therefore, the proposed schemes demonstrate the feasibility of data augmentation using a deep neural network (DNN)-based indoor localization system that lowers the complexity required for use on mobile devices.

Recently, deep-learning-based indoor localisation systems have attracted attention owing to their higher performance compared with traditional indoor localization systems. However, to achieve satisfactory performance, the former systems require large amounts of data to train deep learning models. Since obtaining the data is usually a tedious task, this requirement deters the use of deep learning approaches. To address this problem, we propose an improved data augmentation technique based on received signal strength indication (RSSI) values for fingerprint indoor positioning systems. The technique is implemented using available RSSI values at one reference point, and unlike existing techniques, it mimics the constantly varying RSSI signals. With this technique, the proposed method achieves a test accuracy of 95.26% in the laboratory simulation and 94.59% in a real-time environment, and the average location error is as low as 1.45 and 1.60 m, respectively. The method exhibits higher performance compared with an existing augmentation method. In particular, the data augmentation technique can be applied irrespective of the positioning algorithm used.

To accommodate the rapidly increasing demand for connected infrastructure, automation for industrial sites and building smart cities, the development of Internet of Things (IoT)-based solutions is considered one of the major trends in modern day industrial revolution. In particular, providing high precision indoor positioning services for such applications is a key challenge. Wi-Fi fingerprint-based indoor positioning systems have been adapted as promising candidates for such applications. The performance of such indoor positioning systems degrade drastically due to several impairments like noisy datasets, high variation in Wi-Fi signals over time, fading of Wi-Fi signals due to multipath propagation caused by hurdles, people walking in the area under consideration and the addition/removal of Wi-Fi access points (APs). In this paper, we propose data pre- and post-processing algorithms with deep learning classifiers for Wi-Fi fingerprint-based indoor positioning, in order to provide immunity against limitations in the database and the indoor environment. In addition, we investigate the performance of the proposed system through simulation as well as extensive experiments. The results demonstrate that the pre-processing algorithm can efficiently fill in the missing Wi-Fi received signal strength fingerprints in the database, resulting in a success rate of 88.96% in simulation and 86.61% in a real-time experiment. The post-processing algorithm can improve the results from 9.05–10.94% for the conducted experiments, providing the highest success rate of 95.94% with a precision of 4 m for Wi-Fi fingerprint-based indoor positioning.

No abstract

For cellular-based vehicle-to-everything (C-V2X) communication, vital information about status and intention is periodically broadcasted by each vehicle using the cooperative awareness message (CAM) service. In C-V2X, the task of resource allocation can either be carried out in a centralized manner by the network, termed Mode 3, or by the vehicles themselves in a distributed manner without any core network support, termed Mode 4. Mode 4 scheduling is accomplished by employing sensing-based semi-persistent scheduling (SB-SPS), where the vehicles sense the medium and identify the best time-frequency resource combination for transmission of the CAM. Focusing on Mode 4 in this paper, we present a comprehensive analysis of the impact of variations in the transmit power of the vehicle on the performance of SB-SPS for C-V2X communications in various traffic scenarios through simulations. An adaptive-transmit power control (A-TPC) algorithm is presented to improve the quality of service for various large-scale traffic scenarios, where each vehicle uses real-time channel-sensing information to adjust the transmit power in order to avoid interference with neighbouring vehicles. The results demonstrate that our proposed algorithm outperforms the conventional TPC schemes.

No abstract

No abstract