Automatic fall detection using radar technology significantly advances assisted living and smarter healthcare solutions. In this paper, a novel unsupervised method for detecting fall incidents in human daily activities is proposed. By analyzing ultra-wideband radar returns, a radar time series is obtained and used to find the time-frequency signatures of different activities. These signatures are subsequently binarized and used as input to a deep stacked auto-encoder network for latent feature extraction. The latent features are then clustered through a clustering layer that leverages an auxiliary distribution function to fine-tune the samples clustered into fall and non-fall groups. The proposed fall clustering method is compared against several clustering approaches in terms of accuracy of clustered samples, normalized mutual information and adjusted rand index metrics. It is shown that the proposed method realizes automatic latent feature learning from the radar data and can distinguish fall from non-fall classes by uncovering patterns in a data set with no pre-existing labels. The results show that the proposed unsupervised fall detection method outperforms the other approaches in terms of providing higher accuracy of the clustered data samples. The advantage of the proposed method is that it does not need a large dataset to achieve distinctiveness between clusters.

Dual polarimetric (DP) and compact polarimetric (CP) SAR modes are preferred over quad-channel fully polarimetric (FP) modes for wide-area maritime domain surveillance applications because they provide twice the swath width. Recently, the multilook complex (MLC) product was introduced for RADARSAT Constellation Mission (RCM) imagery, which preserves the polarimetric phase information at a considerably lower data volume over the traditional phase-preserved single-look complex (SLC) product. From statistical theory, the optimal detector for polarimetric data has been previously derived and is known as the optimal polarimetric detector (OPD). However, for a deterministic target model, this detector cannot be applied to MLC data and is also impractical, because it requires complete a priori knowledge of the target. Instead, a suboptimal detector, known as the polarimetric whitening filter (PWF), is often used in practice. This letter proposes a new detector called “quadratic matched filter (QMF)” for CP and DP data that can be applied to MLC products for improved vessel detection over the PWF. A technique to estimate target parameters at processing time is also proposed, which can be used to estimate target parameters for both OPD and QMF. The feasibility and improved performance of the QMF detector are demonstrated through simulated receiver operating characteristic (ROC) performance analysis, and by demonstrating detection performance on an image acquired by RCM. It is shown that the QMF provides approximately 1.5 and 3 dB improvement in signal-to-clutter-plus-noise ratio (SCNR) and peak-signal-to-clutter-plus-noise ratio (PSCNR), respectively, over PWF.

Two-dimensional (2D) simultaneous localization and mapping (SLAM) using a LIDAR is a method used to track the position and orientation of a moving platform. 2D-SLAM assumes that the platform translates in a 2D plane and can only rotate about an axis perpendicular to that plane. However, the assumption of no out-of-plane (OOP) motion does not hold true for platforms experiencing motion in six degrees-of-freedom (6-DOF), such as wearable technologies that have no 3D LIDAR. This paper proposes a new algorithm, called the Level Plane for SLAM (LPS) for removing OOP motion from 2D-LIDAR scans generated on platforms experiencing 6-DOF without requiring scan-matching in 3D. Like other existing methods, an IMU is combined with a 2D-LIDAR to determine the platform's orientation, capture OOP motion, and generate a scan in$3\mathrm{D}$. Unlike other methods, OOP motion is removed by projecting scans onto a globally stabilized coordinate frame in$2\mathbf{D}$where both scan matching and map alignment take place. The proposed algorithm is validated over a series of experiments with different levels of induced and observed OOP motion. Experimental results show that LPS is able to handle more OOP motion than other algorithms and run in real-time.

Pneumothorax (PTX) occurs when air is introduced within the pleural space, separating the chest wall and the lung. Lung ultrasound imaging can be used to rapidly assess for PTX through the presence or absence of the lung sliding artifact. The dynamic, B-mode based lung sliding leads to a characteristic M-mode pattern when sliding is present - the “seashore sign” - (ruling out PTX) and when it is absent - the “barcode” sign (suggestive of PTX). Given its low cost, flexibility, and ability to be worn continuously for long durations, wearable single-element ultrasound sensors may be a feasible option for assessing lung sliding using M-mode ultrasound. In this work, we constructed a thoracic phantom to simulate normal lung function and PTX in order to investigate the feasibility of wearable ultrasound sensor (WUS) technology for the assessment of lung sliding in M-mode. Our preliminary results show that the presence and absence of lung sliding can be visually detected in phantom M-mode images collected using a WUS, and may be analytically differentiated using mean local Shannon entropy.

Standard noise radars, as well as noise-type radars such as quantum two-mode squeezing radar, are characterized by a covariance matrix with a very specific structure. This matrix has four independent parameters: the amplitude of the received signal, the amplitude of the internal signal used for matched filtering, the correlation between the two signals, and the relative phase between them. In this paper, we derive estimators for these four parameters using two techniques. The first is based on minimizing the Frobenius norm between the structured covariance matrix and the sample covariance matrix; the second is maximum likelihood parameter estimation. The two techniques yield the same estimators. We then give probability density functions (PDFs) for all four estimators. Because some of these PDFs are quite complicated, we also provide approximate PDFs. Finally, we apply our results to the problem of target detection and derive expressions for the receiver operating characteristic curves of two different noise radar detectors. In summary, our work gives a broad overview of the basic statistical behavior of noise-type radars.

Automatic detection of a falling person based on noncontact sensing is a challenging problem with applications in smart homes for elderly care. In this article, we propose a radar-based fall detection technique based on time-frequency analysis and convolutional neural networks. The time-frequency analysis is performed by applying the short-time Fourier transform to each radar return signal. The resulting spectrograms are converted into binary images, which are fed into the convolutional neural network. The network is trained using labeled examples of fall and nonfall activities. Our method employs high-level feature learning, which distinguishes it from previously studied methods that use heuristic feature extraction. The performance of the proposed method is evaluated by conducting several experiments on a set of radar return signals. We show that our method distinguishes falls from nonfalls with 98.37% precision and 97.82% specificity, while maintaining a low false-alarm rate, which is superior to existing methods. We also show that our proposed method is robust in that it successfully distinguishes falls from nonfalls when trained on subjects in one room, but tested on different subjects in a different room. In the proposed convolutional neural network, the hierarchical features extracted from the radar return signals are the key to understand the fundamental composition of human activities and determine whether or not a fall has occurred during human daily activities. Our method may be extended to other radar-based applications such as apnea detection and gesture detection.

This paper analyzes the structure of the feature space of radar data collected from real subjects either still or in motion and provides dimensionality reduction and modeling through genetic programming. Three movement classes: sedentary and still, sedentary with movements, and walking contained in the returns obtained from a single channel continuous wave phase-modulated radar are considered. Unsupervised methods are used for finding the intrinsic dimensionality of the space of the original features and nonlinear mappings are used to obtain lower dimensional representations. The classification results for the original and the reduced dimension data are similar, thus, indicating the redundancy of the eliminated features. The white-box models obtained through genetic programming is then compared with the conventional black-box models obtained through supervised classification using random trees, extreme learning machines and multilayer perceptron. For this problem, the explicit white-box models obtained with genetic programming produced equal or better classification accuracies than those obtained with black-box approaches. In addition to explainability, the genetic programming models found have the additional advantage of involving only a few relevant predictors, exhibiting good feature selection capabilities.

Detecting falls using radar has many applications in smart health care. In this paper, a novel method for fall detection in human daily activities using an ultra wideband radar technology is proposed. A time series derived from the radar scattering matrix is used as input to the the residual network for automatic feature extraction. In contrast to other existing methods, the proposed method relies on multi-level feature learning directly from the radar time series signals. In particular, the proposed method utilizes a deep residual neural network for automating feature learning and enhancing model discriminability. The performance of the proposed method is compared with that of the other methods such as support vector machine, K-nearest neighbors, multi-layer perceptron and dynamic time warping techniques. The results show that the proposed fall detection method outperforms the other methods in terms of accuracy and sensitivity values.

Classification is presented as a pre‐processing step in this study. The state of the subject is classified as the unmoving state with normal breathing (normal breathing class), unmoving state with no breathing (stop breathing class) or the state when the subject is moving (erratic signal class) before breathing estimation algorithms are applied. Estimation algorithms may be applied to obtain breathing rate if normal breathing class is detected or alarms may be generated if stop breathing is detected, and fine‐grained classification of activities may be pursued if the erratic signal is detected. Experiments were performed using a single‐channel pulse‐modulated continuous wave radar with three subjects for a total of 135 min. In each experiment, the subject was continuously monitored for 15 min and the subject performed activities that resulted in a signal that belonged to one of the three classes. Besides extracting a feature that assessed the distribution of energy of the signal in the frequency domain, a novel nonlinear time series feature extraction method based on the higher‐dimensional embedding technique was applied to ascertain periodicity of the reflected signal. Bayes classifier was used to classify each 5‐s segment of radar returns. A 30‐fold cross validation resulted in 97% of overall classification accuracy.

In real life applications, it is crucial to monitor the different kinds of human activity without interfering with their regular occupations. Contactless physiological monitoring using radars is a valuable tool, but typically it is performed when the human subjects are immobile. This paper analyzes single channel Continuous Wave (CW) Doppler radar signals in relation to three levels of human activity: i) Sedentary and still, ii) Sedentary and moving and iii) Walking. A combination of computational intelligence techniques (GammaTest, neural networks, random forest and genetic algorithms) was used for assessing the predictive ability of 43 features derived from the radar return signal, as well as of subsets of them, which were composed of highly predictive attributes. It is shown that with about one half the number of attributes it is possible to achieve high levels of classification accuracy, in some cases improving false negative ratios. While several attributes were completely irrelevant and noisy, others were required by discriminating each of the classes. There are attributes required by certain classes in particular and there are others associated to the distinction of classes with subtle differences.

This paper presents a novel algorithm for radar-based estimation of vital signs in a noncontact, privacy friendly manner while subjects are in motion. Unlike the traditional methods that merely use the Fourier spectrum of the output of the radar receiver to obtain estimates of breathing and heart rates, the proposed algorithm uses time-frequency approach. From the Time-Frequency Representation (TFR) of the output of a pseudo-random binary Phase Modulated Continuous Wave (PMCW) radar, frequency of the maximum amplitude at every time instant is estimated and a timeseries of dominant frequencies is formed. MUSIC algorithm is then applied to estimate the vital signs from this series. The proposed algorithm is demonstrated using simulated and real data. Simulated data is obtained through modeling the output of a PMCW radar. Real data is obtained by monitoring a walking subject for 10 minutes in a realistic setting with a 24.125 GHz PMCW radar. The vital sign estimates obtained using the proposed method are found to match closely the estimates from wearable devices that were applied to provide the ground truth for breathing and heart rates.

The recently reported Wirtinger flow (WF) algorithm has been demonstrated as a promising method for solving the problem of phase retrieval by applying a gradient descent scheme. An empirical choice of stepsize is suggested in practice. However, this heuristic stepsize selection rule is not optimal. In order to accelerate the convergence rate, we propose an improved WF with optimal stepsize. It is revealed that this optimal stepsize is the solution of a univariate cubic equation with real-valued coefficients. Finding its roots is computationally simple because a closed-form expression exists. Furthermore, compared with obtaining the coefficients of the cubic equation, calculating the gradient is still the leading cost. Therefore, the proposed approach has the same dominant cost as WF in each iteration. Simulation results are provided to validate its efficiency compared to the existing technique.

This paper investigates the second-order cyclostationarity of block transmitted-single carrier linearly digitally modulated (BT-SCLD) signals, and its applications to signal classification and blind (non-data aided) parameter estimation. Analytical closed-form expressions are derived for the cyclic autocorrelation function (CAF), cyclic spectrum (CS), complementary CAF (CCAF), complementary CS (CCS), and corresponding cycle frequencies (CFs). Furthermore, the conditions for avoiding aliasing in the cycle and spectral frequency domains are obtained. Based on these findings, we propose algorithms for classifying BTSCLD, orthogonal frequency division multiplexing (OFDM), and SCLD signals, and for the blind estimation of the BT-SCLD block transmission parameters. Simulation and laboratory experiments demonstrate the effectiveness of the proposed algorithms under low signal-to-noise ratios (SNRs), short sensing times, and various channel conditions. Furthermore, these algorithms have the advantage of not requiring the recovery of carrier, waveform, and symbol timing information, or the estimation of signal and noise powers.

Cognitive radio (CR) represents a possible solution to the paradoxical problem of simultaneous scarcity and underutilization of the electromagnetic spectrum. Spectrum awareness, a key task of such a radio, encompasses the recognition of the received signal type and parameters. This paper investigates the cyclostationarity approach for the recognition of cyclically prefixed single carrier linearly digitally modulated (CP-SCLD) signals versus SCLD and orthogonal frequency division multiplexing (OFDM) signals under practical conditions, including time-dispersive channels, additive Gaussian noise, and phase, frequency and timing offsets. Analytical closed-form expressions are derived for the cyclic autocorrelation function (CAF) and the set of cycle frequencies (CFs) of CP-SCLD signals. These results are the basis of the proposed signal recognition algorithm. This algorithm has the advantage of avoiding requirements for the recovery of carrier, waveform, and symbol timing information, and the estimation of signal and noise powers.

Previous studies on the cyclostationarity aspect of orthogonal frequency division multiplexing (OFDM) and single carrier linearly digitally modulated (SCLD) signals assumed simplified signal and channel models or considered only second-order cyclostationarity. This paper presents new results concerning the cyclostationarity of these signals under more general conditions, including time dispersive channels, additive Gaussian noise, and carrier phase, frequency, and timing offsets. Analytical closed-form expressions are derived for time- and frequency-domain parameters of the cyclostationarity of OFDM and SCLD signals. In addition, a condition to eliminate aliasing in the cycle and spectral frequency domains is derived. Based on these results, an algorithm is developed for recognizing OFDM versus SCLD signals. This algorithm obviates the need for commonly required signal preprocessing tasks, such as signal and noise power estimation and the recovery of symbol timing and carrier information.

The second-order cyclostationarity of cyclically prefixed single carrier linear digital (CP-tSCLD) modulated signals is investigated with emphasis on its applicability to signal recognition. Analytical closed-form expressions for the second-order (one-conjugate) cyclic cumulants (CCs) and the set of cycle frequencies (CFs) for CP-SCLD modulated signals are derived. Based on these results, an algorithm is proposed for the recognition of CP-tSCLD against SCLD and orthogonal frequency division multiplexing (OFDM) signals. This algorithm obviates the need for signal pre-processing tasks, such as symbol timing, carrier and waveform recovery and estimation of signal and noise powers.

The sensing of the radio frequency (RF) environment is a fundamental concept of cognitive radio (CR). It follows that the detection and classification of very low signal-to-noise ratio signals with relaxed a priori information on signal parameters is a problem of considerable relevance to CR. This paper proposes an algorithm based on first-order cyclostationarity for joint detection and classification of frequency-shift-keying (FSK) signals. This algorithm does not require timing and frequency recovery, and estimation of signal and noise powers. The theoretical analysis of the algorithm performance is validated by simulation results.

This paper studies the nth-order cyclostationarity of orthogonal frequency division multiplexing (OFDM) and single carrier linearly digitally modulated (SCLD) signals affected by a time dispersive channel, additive Gaussian noise, carrier phase, and frequency and timing offsets. The analytical closed-form expressions for the nth-order cyclic cumulants (CCs) and cycle frequencies (CFs) of OFDM and SCLD signals are derived. Furthermore, a second-order CC-based algorithm is developed to recognize OFDM against SCLD signals under the aforementioned conditions. This algorithm obviates the need for signal preprocessing tasks, such as symbol timing estimation, carrier and waveform recovery, and signal and noise power estimation. Simulation experiments confirm the theoretical analysis.

The paper studies the cyclostationarity of an orthogonal frequency division multiplexing (OFDM) with a view to recognizing OFDM against single carrier linear digital (SCLD) modulations. The analytical expressions for the nth-order cyclic cumulants (CCs) and cycle frequencies of an OFDM signal embedded in additive white Gaussian noise and subject to phase, frequency and timing offsets are derived An algorithm based on a second-order CC is proposed to recognize OFDM against SCLD modulations. The recognition algorithm of the authors obviates the need for preprocessing tasks, such as symbol timing estimation, carrier and waveform recovery, and signal and noise power estimation. The results of simulation experiments confirm the theoretical analysis.

In this paper, first-order cyclostationarity of M-ary frequency shift keying (M-FSK) signals affected by additive Gaussian noise, phase, frequency offset and timing errors is investigated, and applied for modulation order recognition. A novel recognition algorithm is proposed, which employs the number of first-order cycle frequencies (CFs) of the received signal as a discriminating feature. The algorithm has the advantage that it requires neither symbol timing and carrier recovery, nor estimation of signal and noise powers as preprocessing tasks. Simulations are carried out to confirm theoretical developments.

This paper proposes an improved classification strategy using misclassified training samples. It is shown that a subset of the misclassified training set forms isolated pockets. In the proposed approach, apart from providing the parameters derived out of the training samples to a classifier, the location of these misclassified pockets is also provided. The proposed strategy overcomes any weakness a given classifier may have by changing the classification decision for a given test sample based on the location of the test sample with respect to the misclassified pockets. Three diversely different classifiers and a simple composite classifier are used to test the strategy. The proposed strategy is implemented on both simulated and real data and it is shown that a reduced error rate can be obtained when this strategy is used.