Showing papers by "Gangolf Hirtz published in 2015"

Assessment and care system based on people detection for elderly suffering from dementia

Abstract: This paper presents a novel AAL (Ambient Assisted Living) concept related to the care of people suffering from dementia In our project, we are developing a stereo vision-based system designed to identify the deficits of demented people who are still capable of living at home In this way, the assessment of their need of care can be improved Furthermore, the system supports demented people during their daily life at home By this means, they can stay longer in their familiar environments, which contributes to the preservation of their health

Hom-e-call - An enhanced fall detection system based on accelerometer and optical sensors applicable in domestic environment

Abstract: A reliable detection of falls is an important challenge for applications of ambient assisted living. Especially for elderly people living autonomously at their home, such a system would help to treat possible injuries faster. Most of the currently used systems are only based on the analysis of the movement data recorded by a wearable device. In this study an automatic fall detection system is developed that can be installed in domestic environment. Based on an accelerometer fixed to the body and optical sensors mounted on the wall, the fusion unit analyses the actual posture of the patient and detects fall events. By means of simulated fall scenarios performed by young athletes, the overall system performance is tested. To identify the false alarm rate of the fall detection system elderly people have performed activities of daily living in a furnished test flat. The combined system approach can be used to improve conventional fall detection algorithms.

OPDEMIVA: An integrated assistance and information system for elderly with dementia

Abstract: Due to demographic developments (rising life expectancy and fall in birth rates), western countries have to face major challenges. One promising remedy for the crisis in elderly care are novel consumer healthcare systems such as support systems for people with special demands. In this paper we present a novel assistance and information system for elderly people suffering from dementia. The system aims to support both the patient as well as the medical professionals, care personnel and informal caregivers. The system is designed to detect daily activities and to use the collected data for assistance (e.g. to remind for medication) and information (e.g. to track progress of disease). For this purpose, we developed a stereo sensor providing RGB-D data for 3-D people detection, object recognition and pose estimation in order to form a system of understanding human behavior. Multiple interfaces provide valuable information for all involved parties.

A modified symbol timing and frequency synchronization method based on cyclic prefix for OFDM systems

Abstract: In this paper, a new blind symbol timing and frequency offset estimation algorithm is proposed for Orthogonal Frequency Division Multiplexing (OFDM) systems in multipath fading channels. The method utilizes the redundant information contained in the cyclic prefix and is suitable within the acquisition as well as tracking phase of an OFDM receiver. The method is based on a new timing function which allows to estimate jointly the symbol timing offset, the frequency offset as well as the channel dispersion. In order to demonstrate the performance of the proposed method, the most recent blind synchronization algorithms in the literature are compared with the new proce- dure under different realistic frequency-selective fading channel conditions.

Low complexity Intercarrier Interference reduction for high mobility wireless systems

Abstract: With the growing number of infotainment services, the demand on high speed internet in vehicles is increasing. Yet, ensuring a high speed internet access at high mobility is challenging. Indeed, high mobility leads to Doppler shifts that result in Inter Carrier Interference (ICI) in OFDM-based wireless systems. Many ICI equalization techniques have been proposed, yet these incur an additional computational complexity and hence higher energy consumption. In this work, we propose a very low complexity interference reduction method. Based on a linear modeling of the channel variation, we derive relations between the interference terms at the different subcarriers and show that interference can be significantly reduced through simple linear operations.

A blind timing and frequency synchronization method based on the correlation characteristics of an OFDM signal

Abstract: Orthogonal Frequency Division Multiplexing (OFDM) systems are very sensitive to symbol timing and frequency offsets Therefore, it is very important to achieve a high precision time and frequency synchronization With the absence of pilot symbols, blind synchronization algorithms using different metric functions or statistics of the received signal can be used In this paper, a new blind symbol timing and frequency offset estimation algorithm is proposed for OFDM systems in multipath fading channels The method utilizes the redundant information contained in the cyclic prefix and does not necessitate any channel profile knowledge In contrary to other works that propose a timing function, the proposed method extracts the timing offset and the channel length directly from the correlation characteristics between the received samples Consequently, an interference-free region is specified and the frequency offset is jointly estimated The performance of the proposed method is demonstrated using Monte Carlo simulations over Rayleigh multipath fading channels of different delay spreads The results show that the proposed algorithm outperforms considerably other algorithms in related work and is robust to varying channel conditions

Mobility Assessment of Demented People Using Pose Estimation and Movement Detection

Abstract: The European population will steadily be growing older in the following decades. At the same time, the risk of getting dementia increases with higher age. Both these factors are apt to cause serious problems for the society, especially with regard to the caring sector, which also suffers from the lack of qualified personnel. As technical support systems can be of assistance to medical staff and patients, a mobility assessment system for demented people is presented in this paper. The grade of mobility is measured by means of the person’s pose and movements in a monitored area. For this purpose, pose estimation and movement detection algorithms have been developed. These process 3-D data, which are provided by an optical stereo sensor installed in a living environment. In order to train and test a discriminative classifier a variety of labelled training and test data was recorded. Moreover, we designed a discriminative and universal feature vector for pose estimation. The experiments demonstrated that the algorithms work robustly. In connection with a human machine interface, the system facilitates a mobilisation as well as a more valid assessment of the patient’s medical condition than it is presently the case.

Pose Estimation and Movement Detection for Mobility Assessment of Elderly People in an Ambient Assisted Living Application

Abstract: In European countries, the increasing number of elderly with dementia causes serious problems for the society, especially with regard to the caring sector. As technical support systems can be of assistance to caregivers and patients, a mobility assessment system for demented people is presented. The grade of mobility is measured by means of the person’s pose and movements in a monitored area. For this purpose, pose estimation and movement detection algorithms have been developed. These algorithms process 3-D data, which are provided by an optical stereo sensor installed in a living environment. The experiments demonstrated that the algorithms work robustly. In connection with a human machine interface, the system facilitates a mobilisation as well as a more valid assessment of the patient’s medical condition than it is presently the case. Moreover, recent advances with regard to action recognition as well as an outlook about necessary developments are presented.

An omnidirectional stereo sensor for human behavior analysis in complex indoor environments

Abstract: Optical sensor-based health care systems have to be both cost-efficient and easily integrable into a domestic environment. Consequently, the reduction of the overall number of necessary sensors is a crucial issue. We present an optical system employing a single omnidirectional stereo sensor for human behavior analysis in complex living environments.

Robust Head-shoulder Detection using Deformable Part-based Models

Abstract: Conventional person detection algorithms lack of robustness, especially when the person is partially occluded. We propose thereby a robust head-shoulder detector in 2-D images using deformable part-based models. This detector can be used in a variety of applications such as people counting and person dwell time measurements. In experiments, we compare the head-shoulder detector with the full body detector quantitatively and analyze the robustness of the detector in realistic scenarios. In the results, we show that the model learned with our method outperforms other methods proposed in related work on an ambient assisted living application.

Low complexity intercarrier interference reduction techniques for OFDM systems

Abstract: In OFDM systems, fast channel variation leads to the loss of orthogonality between the subcarriers. This results in intercarrier interference which cannot be mitigated with conventional single-tap equalization, and hence more complex equalization techniques are required. We seek in this paper low complexity interference equalization (linear in FFT size). Whereas successive interference cancellation targets removing the interference with previously detected symbols, we aim at reducing the interference with the other symbols prior to detection. Based on a proposed channel modeling and approximation procedure, we develop two low-complexity interference reduction techniques that reduce the remaining intercarrier interference significantly. The first targets maximizing the average SINR (Signal to Interference + Noise Ratio), whereas the second is based on minimizing the mean squared error. We compare these 2 methods using Monte Carlo simulations, and demonstrate their effectiveness compared to other methods of similar order of complexity.