Research On Perception And Positioning Based On RSSI Of Wireless Networks

Traditional sensing systems,such as video surveillance technology,need to be monitored without obstructions between the target and the camera.Weather conditions should be good.Heavy rain,snow and fog or obvious changes in the light will cause the perception effect to decline.Radio Frequency Identification technology needs to be tagged and communicated with the reader at close range.These technologies of application are limited.Aiming at the defects of traditional sensing methods,this paper presents a passive device-free sensing method based on wireless sensor networks.Wireless sensor networks are deployed within the monitoring range and data transmission is performed between nodes through radio frequency signals.When someone or obstructs enter the wireless communication link,the Received Signal Strength Indication(RSSI)between the nodes will change.Through analyzing this change and deducing its changing rules,the perception is realized.Compared with traditional infrared sensing and video sensing technologies,wireless sensor networks have the advantages of being insensitive to environmental changes and having fewer blind spots.Passive sensing through a wireless sensor network does not require the target to be tagged or to be cooperated.Only by entering the area can achieve the perception.Therefore,the device-free passive sensing based on wireless sensor networks has not only received wide attention,but also has a wider application range and wider application space.This article first elaborates the propagation characteristics of RF signals then analyzes the path loss in the propagation process of RF signals and points out the reasons that the human body has influence on the received signal strength indicators between sensor nodes.Combined the movement methods of personnel in the monitoring area,a device-free passive sensing algorithm is proposed.In the follow-up study,the flaws of this sensing algorithm has some disadvantages and this sensing algorithm was optimized to improve the perceptual accuracy.Based on the above work,a device-free passive sensing model and experimental method are designed.This includes the description of the model design and the reason of each step in the experiment.According to the experimental results,the feasibility and accuracy of the improved sensing algorithm are further verified.At the same time,perception is extended and positioning research is conducted.Among a large number of positioning methods,an RSSI-based wireless network WiFi fingerprint database location method is introduced.This method is divided into two phases:offline acquisition phase and online positioning phase.The big disadvantage is that the workload is very big.A method which use radial basis functions to perform fingerprint library interpolation is proposed.Relying on the previous work,the interpolation algorithm was verified on the MATLAB experiment platform.The simulation results show that the interpolation method can effectively fill the fingerprint database,which has certain significance for the improvement of positioning efficiency.