Enhancing Human Action Recognition with Adaptive Hybrid Deep Attentive Networks and Archerfish Optimization

In recent years,wearable devices-based Human Activity Recognition(HAR)models have received significant attention.Previously developed HAR models use hand-crafted features to recognize human activities,leading to the extraction of basic features.The images captured by wearable sensors contain advanced features,allowing them to be analyzed by deep learning algorithms to enhance the detection and recognition of human actions.Poor lighting and limited sensor capabilities can impact data quality,making the recognition of human actions a challenging task.The unimodal-based HAR approaches are not suitable in a real-time environment.Therefore,an updated HAR model is developed using multiple types of data and an advanced deep-learning approach.Firstly,the required signals and sensor data are accumulated from the standard databases.From these signals,the wave features are retrieved.Then the extracted wave features and sensor data are given as the input to recognize the human activity.An Adaptive Hybrid Deep Attentive Network(AHDAN)is developed by incorporating a“1D Convolutional Neural Network(1DCNN)”with a“Gated Recurrent Unit(GRU)”for the human activity recognition process.Additionally,the Enhanced Archerfish Hunting Optimizer(EAHO)is suggested to fine-tune the network parameters for enhancing the recognition process.An experimental evaluation is performed on various deep learning networks and heuristic algorithms to confirm the effectiveness of the proposed HAR model.The EAHO-based HAR model outperforms traditional deep learning networks with an accuracy of 95.36,95.25 for recall,95.48 for specificity,and 95.47 for precision,respectively.The result proved that the developed model is effective in recognizing human action by taking less time.Additionally,it reduces the computation complexity and overfitting issue through using an optimization approach.

Signal classification method based on data mining formulti-mode radar

For the multi-mode radar working in the modern electronicbattlefield, different working states of one single radar areprone to being classified as multiple emitters when adoptingtraditional classification methods to process intercepted signals,which has a negative effect on signal classification. A classificationmethod based on spatial data mining is presented to address theabove challenge. Inspired by the idea of spatial data mining, theclassification method applies nuclear field to depicting the distributioninformation of pulse samples in feature space, and digs out thehidden cluster information by analyzing distribution characteristics.In addition, a membership-degree criterion to quantify the correlationamong all classes is established, which ensures classificationaccuracy of signal samples. Numerical experiments show that thepresented method can effectively prevent different working statesof multi-mode emitter from being classified as several emitters,and achieve higher classification accuracy.

Anomalous signals before 2011 Tohoku-oki Mw9.1 earthquake,detected by superconducting gravimeters and broadband seismometers

The 2011 Tohoku-oki earthquake,occurred on 11 March,2011,is a great earthquake with a seismic magnitude Mw9. 1,before which an Mw7. 5 earthquake occurred. Focusing on this great earthquake event,we applied Hilbert-Huang transform（ HHT） analysis method to the one-second interval records at seven superconducting gravimeter（ SG） stations and seven broadband seismic（ BS） stations to carry out spectrum analysis and compute the energy-frequency-time distribution. Tidal effects are removed from SG data by T-soft software before the data series are transformed by HHT method. Based on HHT spectra and the marginal spectra from the records at selected seven SG stations and seven BS stations we found anomalous signals in terms of energy. The dominant frequencies of the anomalous signals are respectively about 0. 13 Hz in SG records and 0. 2 Hz in seismic data,and the anomalous signals occurred one week or two to three days prior to the event. Taking into account that in this period no typhoon event occurred,we may conclude that these anomalous signals might be related to the great earthquake event.

Data signal processing via manchester coding-decoding method using chaotic signals generated by PANDA ring resonator

We investigate the nonlinear behaviors of light recognized as chaos during the propagation of Gaussian laser beam inside a nonlinear polarization maintaining and absorption reducing （PANDA） ring resonator system. It aims to generate the nonlinear behavior of light to obtain data in binary logic codes for transmission in fiber optics communication. Effective parameters, such as refractive indices of a silicon waveguide, coupling coefficients （~）, and ring radius ring （R）, can be properly selected to operate the nonlinear behavior. Therefore, the binary coded data generated by the PANDA ring resonator system can be decoded and converted to Manchester codes, where the decoding process of the transmitted codes occurs at the end of the transmission link. The simulation results show that the original codes can be recovered with a high security of signal transmission using the Manchester method.

Study on the performance of data fusion system for sonar signal detection

The mathematical model and fusion algorithm for multisensor data fusion are presented, and applied to integrate the decisions obtained by multiple sonars in a distributed detection system. Assuming that all the sonars and the fusion system operate at the same false alarm probability, the expression for the detection probability of the fusion system is obtained. Computer simulations reveals that the detection probability and detection range of the fusion system are significantly improved compared to the original distributed detection system.

Metropolitan spatial structure analysis based on the identification of commuting zones with Nanjing City as an example

Commuting zone research is critical to the understanding of the operational rules of the metropolitan spatial structure and improving spatial performance.This study aims to identify the main commuting centers and zones by using cellular data with Nanjing City as the example.This study analyzes the operational features of the internal spatial structures of the city from two dimensions by merging multi-source data,namely,commuting centers and zones,thus achieving an understanding of the existing problems with the urban spatial structures and their internal causes.Results showed that the commuting zones of Nanjing are distributed in a pattern of“multiple commuting centers”,with XinjiekoueHunan Road and Hongwu RoadeChaotiangongeShuangtang as the core,Mochou Lake as the main commuting area,and Dongshan and Jiangpu as the secondary commuting zones.Significant differences and similarities are discovered in our comparisons along the two dimensions of commuting zones and centers in terms of spatial structural factors,such as land use,transportation,and commuting in the city.The similarity is shown as a common declining trend in the values of all our indicators with the increase in the distance of commuting zones from the city center.However,the differences are significant in terms of the clustering features of the various parameters concerning commuting centers and zones.Specifically,four clustering patterns are discovered,namely,“monocentric clustering”,“circular monocentric clustering”,“polycentric clustering”,and“sparsely dotted distribution”.This study sheds light on the existing problems with the city’s spatial structure and proposes some overall suggestions toward urban spatial structure improvement on the basis of these findings.

Driver mental load identification model Adapting to Urban Road Traffic Scenarios

load model established in a road traffic scene is difficult to adapt to the changes of the surrounding road environment during the actual Objective:At present,most research on driver mental load identification is based on a single driving scene.However,the driver mental driving process.We proposed a driver mental load identification model which adapts to urban road traffie scenarios.scene discrimination sub-model can quickly and accurately determine the road traffic scene.The driver load identification sub-model Methods:The model includes a driving scene discrimination sub-model and driver load identification sub-model,in which the driving sub-model.selects the best feature subset and the best model algorithm in the scene based on the judgement of the driving scene classification Results:The results show that the driving scene discrimination sub-model using five vehicle features as feature subsets has the best performance.The driver load identification sub-model based on the best feature subset reduces the feature noise,and the recognition tends to be consistent,and the support vector machine(5VM)algorithm is better than the K-nearest neighbors(KNN)algorithm.effect is better than the feature set using a single source signal and all data.The best recognition algorithm in different scenarios Conclusion:The proposed driver mental load identificution model can discriminate the driving scene quickly and accurately,and then identify the driver mental load.In this way,our model can be more suitable for actual driving and improve the effect of driver mental load identification.

Spatial patterns of residents’ daily activity space and its influencing factors based on the CatBoost model: A case study of Nanjing, China

The complexity and fragmentation of people’s activity space are challenging to planners.However,the relevant studies are mostly concerned on the relationship between the social attributes and the activity space of residents in a single or several communities,or the spatiotemporal laws of activity space on a macro scale.The research on the spatial characteristics of residents’activity space still needs to be strengthened.The present study analyses the spatial patterns of residents’activity space based on mobile phone signaling data to fill the gap of previous studies that assessed residents’activity space across small geographic areas.First,according to the spatial scope and direction of an activity space and residents’activity coverage rate,spatial patterns can be divided into three types:compact,extended,and directional extension patterns.The CatBoost method is then used to statistically analyze the influencing variables of spatial patterns,and the order of importance of the following influencing factors is determined:the built environment is more influential than social and economic situations.This study aims to strengthen the understanding of residents’activity space at the spatial level and provide a basis for the optimization of communities with different spatial patterns.

Development of Suburban New Towns in Shanghai: Jobs-Housing Spatial Relationship Analysis

This paper identifies the employment and housing locations of residents in Shanghai based on mobile phone signaling data, so as to obtain the employment density and commuting data and analyze the development of nine suburban new towns from the perspective of jobs-housing spatial relationship. Firstly, the paper defines employment-intensive areas and gets the average employment density of each new town according to the employment density data. Then it marks out the scope of the employment influence through analyzing the sources of workers in each new town in accordance with the commuting data. Finally, it analyzes the jobs-housing balance of each new town using independence index, finding that suburban new towns in Shanghai have become main clusters of economic activities, while the scope of employment influence in each new town is still concentrated in its administrative area, with less attraction to residents in other areas. The independence index demonstrates a law that the suburban new town which is farther from the central city sees a higher degree of jobs-housing balance. Among them, new towns located in the outer suburbs with a low independence index indicate their special development situation, the reason of which is worth further study.