IET Wireless Sensor Systems最新文献

To acquire an accurate location on the occasions, such as in an indoor, tunnel, and valley, where satellite navigation signals fail. The paper designs a pedestrian navigation system by using the zero velocity update procedure technology (ZUPT) and Kalman filter to reduce the location error. The measurement noise characteristic (mean and variance) of the micro electro mechanical systems gyros is unknown and time variant, but in traditional studies, it is usually thought and calculated as a constant. So the optimality of the error estimation of the Kalman filter cannot be reached. To address this question, this paper proposes the improved Sage–Husa Adaptive Kalman Filter (SHAKF) based on the index fading memory factor to realise the state estimation of the Kalman filter and navigation error correction. The advantage of improved SHAKF is it can accurately estimate the state vector when the measurement noise is unknown and time variant. To verify the validity of novel navigation methods, walking experiments under outdoor environments and indoor environments are carried out. The results of actual walking experiments demonstrate that the proposed method can effectively reduce the pedestrian location error compared with the traditional ZUPT method. The mean location error is reduced by more than 10%, and the variance of the location error is reduced by more than 5%.

Nowadays, time-varying and high-density data of wireless sensor systems and communication networks compel us to propose and realise low-complexity and time-efficient algorithms for searching, clustering, and sorting. A novel threshold-based sorting algorithm applicable to dense and ultra-dense networks is proposed in this study. Instead of sorting whole data in a large data set and selecting a certain number of them, the proposed algorithm sorts a specific number of elements that are larger or smaller than a threshold level or located between two threshold values. First, based on the mean value and standard deviation of the data, a theoretical analysis to find the exact and approximate thresholds, respectively for known (Gaussian, uniform, Rayleigh, and negative exponential) and unknown probability distributions is presented. Then, an algorithm to sort a predefined number of data is realised. Finally, the effectiveness of the proposed algorithm is shown in the view of the time complexity order, the running time, and the similarity measure. To do this, theoretical and numerical analyses are used to show the superiority of the proposed algorithm in known and unknown distributions to the well-known conventional and gradual conventional versions of Merge, Quick, and K-S mean-based sorting algorithms.

In order to improve the rationality of landscape architecture layout planning and reduce the energy consumption in the process of data fusion, a landscape architecture layout planning modelling method based on multi-sensor data fusion is proposed in this study. Firstly, the spatial layer of landscape plants is divided, and the quantitative index value of landscape architecture is calculated by Simpson diversity index. Secondly, the weight of plant index is determined by analytic hierarchy process, and the error of garden image is analysed by unsupervised classification of garden image based on multi-sensor data fusion. Then, the garden image is classified according to the existing template, and the training samples of each category of garden are extracted. Finally, the modelling of landscape architecture layout planning is constructed, and the optimal planning method of landscape architecture layout based on maximum likelihood method is proposed. The experimental results show that this method can effectively reduce the energy consumption in the data fusion process, reduce the probability of damaged nodes becoming cluster heads, improve the stability of landscape forest layout planning method, match the plants according to the plant species, and improve the rationality of landscape architecture layout planning.

This paper studies the performance improvement of the IEEE 802.15.4 non-beacon-enabled mode originated by the inclusion of the Request-To-Send/Clear-To-Send (RTS/CTS) handshake mechanism resulting in frame concatenation. Under IEEE 802.15.4 employing RTS/CTS, the backoff procedure is not repeated for each data frame sent but only for each RTS/CTS set. The maximum throughput and minimum delay performance are mathematically derived for both the Chirp Spread Spectrum and Direct Sequence Spread Spectrum Physical layers for the 2.4 GHz band. Results show that the utilisation of RTS/CTS significantly enhances the performance of IEEE 802.15.4 applied to healthcare in terms of bandwidth efficiency.

One of the most important challenges of wireless sensor networks is controlling network congestion and transmitting data in a way that improves the quality of service (QoS) parameters. Thus, it increases network performance and reduces energy consumption. Energy consumption increases due to various reasons, such as unsuccessful delivery of packets to the receiver, congestion in the network, retransmission of packets, delay in delivering packets to the base station, and so on. Given the importance of some data in the field of health, congestion should be avoided and secure data transmission should be ensured. This study divides the collected data into two groups based on their intrinsic characteristics by presenting a congestion management protocol: (1) critical data and (2) non-critical data. The proposed protocol provides a dynamic routing algorithm based on the TOPSIS model for non-critical data transmission. In addition, an algorithm for transmitting critical data through the shortest possible path is also provided based on support vector machines (SVMs). This improves the network performance through using multi-classification that is obtained from SVMs. The simulation results indicate that the proposed method works better than other methods and leads to better performance in delay, network performance, and power consumption.

The pervasiveness of Wi-Fi signals provides significant opportunities for human sensing and activity recognition in fields such as healthcare. The sensors most commonly used for passive Wi-Fi sensing are based on passive Wi-Fi radar (PWR) and channel state information (CSI) data, however current systems do not effectively exploit the information acquired through multiple sensors to recognise the different activities. In this study, new properties of the Transformer architecture for multimodal sensor fusion are explored. Different signal processing techniques are used to extract multiple image-based features from PWR and CSI data such as spectrograms, scalograms and Markov transition field (MTF). The Fusion Transformer, an attention-based model for multimodal and multi-sensor fusion is first proposed. Experimental results show that the Fusion Transformer approach can achieve competitive results compared to a ResNet architecture but with much fewer resources. To further improve the model, a simple and effective framework for multimodal and multi-sensor self-supervised learning (SSL) is proposed. The self-supervised Fusion Transformer outperforms the baselines, achieving a macro F1-score of 95.9%. Finally, this study shows how this approach significantly outperforms the others when trained with as little as 1% (2 min) of labelled training data to 20% (40 min) of labelled training data.