Research On The Radio Tomography Image Based On Wireless Sensor Network

In recent years, indoor localization techniques have become one hop topic of wireless sensor network and been widely used in many applications ranging from smart home, logistics, healthcare, emergency relief and so on. Many of the localization systems rely on the target attaching an electronic device and cooperate with the localization systems reporting its location information constantly. However, these kinds of systems are not capable of locating survivors during fires or earthquakes, helping police to locate terrorists, who are unwilling to be located, through wall. Radio tomography image is one critical part of Device-Free Localization, which does not require any electronic device to be attached on the target. However, the accuracy of radio tomography image deteriorates seriously in environment with Wi Fi and multipath interference. State-of-the-art methods do not efficiently solve these problems.In this dissertation, we introduce an algorithm of fusing dual-band images in order to obtain an enhanced image inferring more precise location and propose a multi-channel adaptive weighing model algorithm that regulating the weight of the measurement channel according to the interference intensity of Wi Fi and multipath. In order to mitigate the influence of Wi Fi to the accuracy of radio tomography image, we propose a dual-band radio tomography image algorithm. The basic idea is to combine the advantages of both 2.4 GHz and 433 MHz band to establish a RTI system that is robust to Wi Fi interference and allows for accurate indoor localization even with sever interference. To make sure that the two radio tomography images, one is from 2.4 GHz and the other is from 433 MHz, reflect the targets at exactly the same location, a dual-band data synchronization protocol is proposed. Field experiments show that dual-band radio tomography image algorithm can improve the accuracy of radio tomography image further in the environment with Wi Fi interference. To settle the problem that Wi Fi and multipath interference leading to serious deterioration of radio tomography image accuracy at the same time, a novel radio tomography image algorithm based on multi-channel adaptive weight model is proposed to improve the accuracy of radio tomography image further in the environment with Wi Fi and multipath. Real-world experiments show that the radio tomography image based on multi-channel adaptive weight mode outperforms the dual-band radio tomography image algorithm.