Radar walking speed measurements of seniors in their apartments: technology for fall prevention.

Falls are a significant cause of injury and accidental death among persons over the age of 65. Gait velocity is one of the parameters which have been correlated to the risk of falling. We aim to build a system which monitors gait in seniors and reports any changes to caregivers, who can then perform a clinical assessment and perform corrective and preventative actions to reduce the likelihood of falls. In this paper, we deploy a Doppler radar-based gait measurement system into the apartments of thirteen seniors. In scripted walks, we show the system measures gait velocity with a mean error of 14.5% compared to the time recorded by a clinician. With a calibration factor, the mean error is reduced to 10.5%. The radar is a promising sensing technology for gait velocity in a day-to-day senior living environment.

Radar walk detection in the apartments of elderly.

Seniors want to live more independent lifestyles. This comes with some risks including dwindling health and major injuries due to falling. A factor that has been studied and seen to have a correlation to fall risk is change in gait speed. Our goal is to create a passive system that monitors the gait of elderly so that assessments can be given by caregivers if gait changes do occur. This paper will cover a method of using pulse-Doppler radar to detect when walks occur. In unscripted living environments, we are able to detect valid walks. The system does miss walks during the day, but when walks are detected, they are actually valid walks 91.8% of the time using a large data base of radar signals captured in living environments.

A covariance fitting approach for correlated acoustic source mapping.

Microphone arrays are commonly used for noise source localization and power estimation in aeroacoustic measurements. The delay-and-sum (DAS) beamformer, which is the most widely used beamforming algorithm in practice, suffers from low resolution and high sidelobe level problems. Therefore, deconvolution approaches, such as the deconvolution approach for the mapping of acoustic sources (DAMAS), are often used for extracting the actual source powers from the contaminated DAS results. However, most deconvolution approaches assume that the sources are uncorrelated. Although deconvolution algorithms that can deal with correlated sources, such as DAMAS for correlated sources, do exist, these algorithms are computationally impractical even for small scanning grid sizes. This paper presents a covariance fitting approach for the mapping of acoustic correlated sources (MACS), which can work with uncorrelated, partially correlated or even coherent sources with a reasonably low computational complexity. MACS minimizes a quadratic cost function in a cyclic manner by making use of convex optimization and sparsity, and is guaranteed to converge at least locally. Simulations and experimental data acquired at the University of Florida Aeroacoustic Flow Facility with a 63-element logarithmic spiral microphone array in the absence of flow are used to demonstrate the performance of MACS.

Enhanced channel estimation and symbol detection for high speed multi-input multi-output underwater acoustic communications.

The need for achieving higher data rates in underwater acoustic communications leverages the use of multi-input multi-output (MIMO) schemes. In this paper two key issues regarding the design of a MIMO communications system, namely, channel estimation and symbol detection, are addressed. To enhance channel estimation performance, a cyclic approach for designing training sequences and a channel estimation algorithm called the iterative adaptive approach (IAA) are presented. Sparse channel estimates can be obtained by combining IAA with the Bayesian information criterion (BIC). Moreover, the RELAX algorithm can be used to improve the IAA with BIC estimates further. Regarding symbol detection, a minimum mean-squared error based detection scheme, called RELAX-BLAST, which is a combination of vertical Bell Labs layered space-time (V-BLAST) algorithm and the cyclic principle of the RELAX algorithm, is presented and it is shown that RELAX-BLAST outperforms V-BLAST. Both simulated and experimental results are provided to validate the proposed MIMO scheme. RACE'08 experimental results employing a 4 x 24 MIMO system show that the proposed scheme enjoys an average uncoded bit error rate of 0.38% at a payload data rate of 31.25 kbps and an average coded bit error rate of 0% at a payload data rate of 15.63 kbps.

Sparsity constrained deconvolution approaches for acoustic source mapping.

Using microphone arrays for estimating source locations and strengths has become common practice in aeroacoustic applications. The classical delay-and-sum approach suffers from low resolution and high sidelobes and the resulting beamforming maps are difficult to interpret. The deconvolution approach for the mapping of acoustic sources (DAMAS) deconvolution algorithm recovers the actual source levels from the contaminated delay-and-sum results by defining an inverse problem that can be represented as a linear system of equations. In this paper, the deconvolution problem is carried onto the sparse signal representation area and a sparsity constrained deconvolution approach (SC-DAMAS) is presented for solving the DAMAS inverse problem. A sparsity preserving covariance matrix fitting approach (CMF) is also presented to overcome the drawbacks of the DAMAS inverse problem. The proposed algorithms are convex optimization problems. Our simulations show that CMF and SC-DAMAS outperform DAMAS and as the noise in the measurements increases, CMF works better than both DAMAS and SC-DAMAS. It is observed that the proposed algorithms converge faster than DAMAS. A modification to SC-DAMAS is also provided which makes it significantly faster than DAMAS and CMF. For the correlated source case, the CMF-C algorithm is proposed and compared with DAMAS-C. Improvements in performance are obtained similar to the uncorrelated case.