Optimization Algorithm Based On RSSI Indoor Positioning Technology

Indoor positioning technology based on Wireless sensor network (WSN) has wide range of needs and huge commercial potential. However, the most important reason that limits the indoor positioning technology development is a larger error of the indoor positioning. At present, the widely used in the method indoor positioning which based on RSSI is strongly dependent on the environment. The factors that interfere with the measurement of RSSI are Long-distance communication, obstruction, burst interference, random disturbance and noise jamming. Furthermore, all of these reduce the accuracy of indoor positioning. The dissertation analyzes the extent of RSSI measurement which is caused by the mentioned factors, and proposes the improved scheme to improve the localization accuracy.We Reduce the impact of long-distance communication and obstacles on RSSI measurements by segmented the overall regional. Environmental parameter fitting will be executed in the divided sub-areas. As the uncertainty of indoor environment, personnel move is inevitable, while will produce burst interference and introduction of deviation value of RSSI observation data. Clustering algorithm is adopted to eliminate the deviation value. Radio signals is vulnerable to interference by random noise, signal reflection, signal diffraction and measurement system itself, such as internal noise of the measurement nodes and A/D quantization noise. These factors make the RSSI observation data generate the random error. By analyzing the characteristics of random disturbance, filtering algorithms is applied to weaken the influence of random disturbance. In addition to the above two kinds of interference, the RSSI measurements are also subject to noise jamming. Due to the noise jamming is conforms to the characteristics of Gaussian white noise, we apply wiener filtering to eliminate the influence caused by noise jamming.This dissertation also studies the strategy of choice reference node in the positioning algorithm, and proposed an improved algorithm. The improved algorithm take the highest reliability of the three reference nodes which measured by weights to complete the localization, thereby reducing the positioning error.Verify the improved scheme through the experiment on Zigbee module, comparison of experimental data before and after improvements, finally verifies the effectiveness of the improved scheme.