This study used micro-Doppler radar (MDR) measurements to investigate the significance of associations between cognitive functions and gait features of elderly persons. The aim of the work was to develop a system that would enable the risks of developing dementia and related diseases to be monitored remotely on a daily basis. Study participants were adults aged 75 years and older. Gait velocity parameters corresponding to the walking speed and leg and foot velocities were remotely extracted via a simple 24-GHz MDR system in real time. The relationships between the extracted gait velocity parameters and the global cognition and cognitive functions in various cognitive domains (processing speed, memory, executive function, and language domains) that were assessed by conventional paper- and question-based tests were statistically analyzed. Our results revealed that, apart from the walking speed, which was mainly considered in a previous study, other parameters reflecting the leg and foot velocities are effective for the detection and classification of elderly participants with lower cognitive functions in the various cognitive domains. In particular, the statistical significance of the association of the leg velocity in the swing phase with the results of all the cognitive function tests is larger than that of the walking speed. Another important finding is that different gait velocity parameters are associated with each cognitive domain and this means that the MDR-based gait measurement can be used to determine which cognitive domain has deteriorated.

This chapter presents Kalman filters for tracking moving objects and their efficient design strategy based on steady-state performance analysis. First, a dynamic/measurement model is defined for the tracking systems, assuming both position-only and position-velocity measurements. Then, problems with the Kalman filter design in tracking systems are summarized, and an efficient steady-state performance index proposed by the author [termed the root-mean-squared error index (the RMS index)] is introduced to resolve these concerns. The analytical relationship between the proposed RMS index and the covariance matrix of the process noise is shown, leading to a proposed design strategy that is based on this relationship. Theoretical performance analysis is conducted using the performance indices to show the optimality of the design strategy. Numerical simulations show the validity of the theoretical analyses and effectiveness of the proposed strategy in realistic situations. In addition, the optimal performance of the position-only-measured and position-velocity-measured systems is analyzed and compared. This comparison shows that the position-velocity-measured Kalman filter tracking is accurate when compared with the position-only-measured filter.

This letter presents a gait classification technique for the identification of individuals with different gait patterns using simulated micro-Doppler radar remote sensing data. Proposed feature parameters for the classification are principal components of velocities extracted via micro-Doppler radar signals generated using motion capture-based kinematic data. Distinct differences were found in the proposed parameters among three groups of subjects with different gait patterns: healthy young and elderly adults, and elderly adults with a history of falls (elderly fallers).

We present an automatic parameter setting method to achieve an accurate second-order Kalman filter tracker based on a steady-state performance index. First, we propose an efficient steady-state performance index that corresponds to the root-mean-square (rms) prediction error in tracking. We then derive an analytical relationship between the proposed performance index and the generalized error covariance matrix of the process noise, for which the automatic determination using the derived relationship is presented. The model calculated by the proposed method achieves better accuracy than the conventional empirical model of process noise. Numerical analysis and simulations demonstrate the effectiveness of the proposed method for targets with accelerating motion. The rms prediction error of the tracker designed by the proposed method is 63.8% of that with the conventional empirically selected model for a target accelerating at 10 m/s ² .

This study presents a radar-based remote measurement system for classification of human behaviors and falls in restrooms without privacy invasion. Our system uses a dual Doppler radar mounted onto a restroom ceiling and wall. Machine learning methods, including the convolutional neural network (CNN), long short-term memory, support vector machine, and random forest methods, are applied to the Doppler radar data to verify the model’s efficiency and features. Experimental results from 21 participants demonstrated the accurate classification of eight realistic behaviors, including falling. Using the Doppler spectrograms (time–velocity distribution) as the inputs, CNN showed the best results with an overall classification accuracy of 95.6% and 100% fall classification accuracy. We confirmed that these accuracies were better than those achieved by conventional restroom monitoring techniques using thermal sensors and radars. Furthermore, the comparison results of various machine learning methods and cases using each radar’s data show that the higher-order derivative parameters of acceleration and jerk, and the motion information in the horizontal direction are the efficient features for behavior classification in a restroom. These findings indicate that daily restroom monitoring using the proposed radar system accurately recognizes human behaviors and allows early detection of fall accidents.

High-resolution, short-range sensors that can be applied in optically challenging environments (e.g., in the presence of clouds, fog, and/or dark smog) are in high demand. Ultrawideband (UWB) millimeter-wave radars are one of the most promising devices for the above-mentioned applications. For target recognition using sensors, it is necessary to convert observational data into full 3-D images with both time efficiency and high accuracy. For such conversion algorithm, we have already proposed the range points migration (RPM) method. However, in the existence of multiple separated objects, this method suffers from inaccuracy and high computational cost due to dealing with many observed RPs. To address this issue, this letter introduces Doppler-based RPs clustering into the RPM method. The results from numerical simulations, assuming 140-GHz band millimeter radars, show that the addition of Doppler velocity into the RPM method results in more accurate 3-D images with reducing computational costs.