Dissertation > Excellent graduate degree dissertation topics show

Design and Verification of Fuzzy Logic Algorithm Based on Signal Strength

Author: WangJiWei
Tutor: MengWeiXiao
School: Harbin Institute of Technology
Course: Information and Communication Engineering
Keywords: positioning algorithm subtractive clustering fuzzy logic fingerprint
CLC: TN929.5
Type: Master's thesis
Year: 2009
Downloads: 24
Quote: 0
Read: Download Dissertation

Abstract


With an increasing popularity of a variety of wireless devices, the perceived demand of location-based information is gradually rising up in recent years, which made the wireless location technology be widely used. At present, most of the WLAN positioning technologies are based on Time of Arrival, Time Difference of Arrival and Direction of Arrival etc, which requires synchronization between the transmitter and receiver, and is difficult to achieve in many applications. Another choice of WLAN positioning is based on Receiving Signal Strength (RSS) which is a better choice of indoor positioning for its broad coverage.The non-line-of-sight transmission of wireless network signal and the difference between physical properties and software of client lead to the large difference of RSS characters. The Fingerprint method the paper adopts is a good solution for the problems above, which selects RSS as the location character. During the off-line phase, the system establishes Radio Map through the analysis of indoor signal propagation and pre-processing part, then enters the off-line fuzzy logic positioning algorithm training one, reselects the initial reference points on the condition of over fitting or over matching. During the on-line phase, this method makes the real-time space signal sampling at the checking points on the basis of Radio map, inputs pre-processing sampling data in fuzzy logic positioning algorithm for searching and matching period, finally achieves the predicted point coordinates.This paper adopts the clustering method for the structure identification. After defining and amending the density of each point, the clustering method selects the data with the highest density as the cluster center, then take the iterative process constantly until identify all the effective cluster centers,so as to divide the fuzzy set into sub sets. This paper adopts the BP algorithm of the neural network model to optimize the parameters for the objective of minimum systematic error. The mean of RSS has good location dependence, in two experimental environments, this paper selects the mean of AP as the features of location and the algorithm input. The RSS straight distribution shows bimodal distribution characteristics which greatly influences the location dependence and is the main reason for the decline of generalization ability. The further experiment shows the connection status of APs and network card has great influence on bimodal distribution, which is proved in the reference experiment of the network card. In the second environment, the result shows an improvement with a large upgrade. For a more detailed analysis of system performance, the algorithm makes an ideal based on simulation free space propagation model. Finally, this paper states the simulation of the algorithm and a comparation of NN algorithm, which shows the fuzzy logic algorithm a good system performance based on signal strength.

Related Dissertations

  1. The Relation between Fuzzy Logic Guidance Law and Homing Quality,TJ765
  2. Study on the Pharmacokinetics of Xiaoaiping and Chlorogenic Acid in Rats and the Quality Standard of Xiaoaiping Preparations,R285
  3. Studies on the Commodity Survey and Quality Evaluation of Radix Paeonia Rubra,R282.71
  4. Short-Range Wireless Localization Algorithms for Internet of Things,TN929.5
  5. Quality Standard Revise and Finger Print Study of Pipazhike Granule,R286.0
  6. Drug Susceptibility Testing of Staphylococcus Aureus from Children Impetigo and Typing of Staphylococcus Aureus by Randomly Amplified Polymorphic DNA in Chengdu,R440
  7. Road Traffic Safety Comprehensive Evaluation Method,U492.8
  8. With the mobile node localization algorithm for wireless sensor networks,TP212.9
  9. Research and Design on Asymmetric Encryption and Decryption Algorithm Based on FFTT,TP309.7
  10. Research on Pressure Sensed Fingerprint Identification System and Algorithem,TP391.41
  11. Research on Methods of Fingerprint Image Segmentation,TP391.41
  12. Design and Implementation of family planning management system in Yongkang,TP311.52
  13. The Study of Positioning and Tracking of Moving Objects Based on ZigBee,TN929.5
  14. An Improvement Protocol of WLAN MAC Based on Fuzzy Logic Control,TN925.93
  15. The Technology of Browser Security Access and Fingerprint Identification,TP393.092
  16. Research on Multi-modalbiometric Fusion Personalidentification Technology Based on Fingerprint and Name Speech,TP391.41
  17. A Research on ISPs’ Liabilities in China and Foreign Countries,D923.41
  18. Inline deduplication technology research and implementation,TP333
  19. Based on pattern recognition of fingerprint information system design and implementation,TP391.41
  20. Cloud backup pair of fingerprint verification technology with multithreading transfer,TP309.3
  21. Collaborative learning and digital fingerprint based on copyright protection and tracking technology,TP309

CLC: > Industrial Technology > Radio electronics, telecommunications technology > Wireless communications > Mobile Communications
© 2012 www.DissertationTopic.Net  Mobile