Detection of human vital signs through obstructive barriers using UWB GPR

This thesis involves detection of human vital signs detected through obstructive barriers using UWB GPR systems. After a short introduction and background, radar and ground penetrating radar(GPR) principles, clutter reduction techniques and signal processing algorithms for analysis of human vital signs in frequency spectrum are discussed. Radar emits radio waves through media, and back-scattered waves from objects are collected using various antenna types. Radar systems basically consist of a radar device to generate radio signals, antennas for transmission and reception, and computers for recording and processing data. History of radar goes back to World War II. Today's radar systems are widely used both in civil and defense industry. Detection of hazardous fields in enemy territory and monitoring of ships and planes are some of the uses. In civil life, radar systems can be used in traffic system and weather forecast. Ground Penetrating Radar or shortly GPR is technique among radar systems that uses radio waves through ground to detect buried pipes, mines or buried objects. GPR mostly uses pulse technique to generate radio signals. Ground types such as soil, moist soil, and rocky fields represent different attenuation characteristics. Thus, type of ground, thickness of material, and depth are important parameters for object imaging and detection. Ultra-wideband(UWB) systems are modern high gain&resolution techniques applied to radar systems. High resolution is provided by wide band-width. Plus, fast data speed, target recognition capability and reduced radar dead zones make those systems attractive for detection objects through walls and/or under rubble. Detection of human vital signs, such as, respiratory rate and heart beat through obstacles are a hot topic for research related to rescue operations after natural disasters, such as earthquakes. Such techniques may also make possible remote (non contact) monitoring of vital signs of patients in a hospital environment. Continuous observation of vital signs are important because most signals in human body are by nature continuous. A portable radar device which detects human vital signs under collapsed buildings and rubble can become life saving. Target recognition capability, high gain and wide bandwidth makes UWB systems excellent candidate to detect human vital signs through wall or under debris. Currently, UWB GPR systems are used for such purposes. Human respiratory motion carries characteristic information and its frequency is between 0,2 Hz and 0,5 Hz. During breathing lungs expand and contract. These expansions and contractions lead to small displacements in chest cavity. These tiny displacements can be detected using UWB GPR systems. Signal processing algorithms are crucial for accurate detection. Background subtraction and signal localization play a big role in data analysis. Clutter type and thickness of clutter makes the environment more complex. Suppression of clutter is essential. Frame Differencing, Principal Component Analysis and Singular Value Decomposition Techniques are well known algorithms for background elimination. Analysis of human vital signs in frequency domain can be performed with conventional transform techniques, like Fast Fourier Transform, but such analyses may also be carried out with modern techniques. Wavelet Transform and Hilbert Huang Transform are the latest techniques for data interpretation. Hilbert Huang Transform is composed of Hilbert Transform and Empirical Mode Decomposition algorithm. This technique is effective on both non-linear and non-stationary data analysis. Also, the method is adaptive and local. In this study, a UWB GPR system is designed for detection of human vital signs through obstacles. After successful data acquisition, signal processing algorithms are explained in two sections. First one is background subtraction techniques to eliminate obstacle effect. The last one is signal localization algorithms and covers Hilbert Huang Transform and Fast Fourier Transform. Finally, results will be presented and conclusions drawn.