Search

Hand gesture recognition based on micro‐Doppler (MD) radar has garnered considerable attention from researchers as a potential method for human–computer interaction (HCI). However, two significant challenges, that is, obvious differences in MD map size and lots of redundant information contained in the MD maps, are encountered. Here, a multi‐scale graph‐based hand gesture recognition framework is proposed. First, a MD graph representation method is developed, which is adapted to arbitrary‐sized frames and enables to map the key features in a sparse manner. Then, the multi‐scale information from MD graph is fully extracted for hand gesture recognition. Experimental results show that the proposed framework achieves a state‐of‐the‐art accuracy of 96.73% on the four‐class radar hand gesture dataset, while reducing up to 99% of the redundant information in the MD maps. This framework requires only a little amount of memory storage with good hand gesture recognition capability, demonstrating its high potential in the HCI field. [ABSTRACT FROM AUTHOR]

The issue of ambiguous frequency is a major drawback of the Doppler through-wall radar in target tracking applications. To solve the issue, an additional receiver, which can acquire target information from a different detection perspective, is added to the radar system, and the data fusion process is proposed to identify and track the targets in real time. The linear prediction coding technique is combined with the data fusion process to predict the target initial state for higher positioning accuracy. Compared with the traditional tracking approach, the proposed algorithm can not only improve the target positioning accuracy, suppress the issue of ambiguous frequency, but also help to identify stationary and tangentially moving targets, which is very essential in target tracking applications. A series of experiments are conducted to illustrate an outranking performance of the proposed algorithm in comparison with the traditional approach. [ABSTRACT FROM AUTHOR]

Based on the WRF-3DVar system, different momentum control variables in radar data assimilation are investigated for their effects on the analysis and forecast of strong convective weather systems in the study. The single observation assimilation tests are performed besides three cycling data assimilation experiments based on U-wind/ V -wind (UV) and stream-function and velocity potential (ψχ) that are scalar representations of flow fields for their irrotational and divergence components. The wind field increments for both control variables in single observation tests show that when UV are the control variables, the wind field increments have a smaller range of influence that it better reflects the characteristics of the observed information itself at the convective scale than ψχ. When ψχ are used as control variables, negative increments are generated in the analyzed wind field because ψχ have the property of maintaining the integrated value of the wind field. In addition, radar data from 5 Doppler radars in northeast China are assimilated. The cycling assimilation experiments show that CV_UV produces enhanced forecast than CV_ ψχ in terms of radar reflectivity and precipitation. Another case also proves this conclusion. In particular, CV_UV improves the dynamics field and water vapor convergence, which lead to better intensity and structure prediction of the strong convection process. In contrast, CV_ ψχ yields relative discontinuous wind field structure, which tends to reduce the organization of the forecast strong echoes. Another set of experiments are further designed with only radar reflectivity assimilated with CV_UV. The results show that assimilating radial velocity is helpful to improve the forecasting skill. • Different control variables are studied for their effects on the analysis and forecast of strong convective systems. • CV_UV is relatively more localized in terms of analysis increments and is applicable to the convective scale. • CV_UV improves the dynamics field and water vapor convergence for the strong convection process. • CV_UV produces enhanced forecast than CV_ψχ in terms of radar reflectivity and precipitation. [ABSTRACT FROM AUTHOR]