Research And Implementation Of Wi-Fi Location Algorithm Based On Map-Reduce

With the popularity of Wi-Fi (Wireless Fidelity) technology and Wi-Fi hotspot coverage in recent years, continuous improvement has made Wi-Fi targeting possible and feasible. Whether residential buildings,schools, offices, or supermarkets, museums, buses and other public places have been Wi-Fi signal coverage. At the same time, the daily life of people increasingly need location-based services, this demand for the development of positioning technology has brought opportunities, more and more researchers will focus on positioning algorithm research.However, in complex environments, the existing algorithms such as ToA(Time of Arrival), TDoA (Time Difference of Arrival), AoA (Angle of Arrival) and other localization algorithms will cause errors because of the interference of radio waves. So, how to improve the accuracy of positioning to provide better location services is a popular research topic in recent years.The emergence of cloud computing promotes the further development of wireless location technology. Hadoop is a Google Cloud Computing platform open source implementation. It uses the programming model of MapReduce to realize parallel computing of large-scale data sets, which makes the system provided with the characteristics of distributed storage of massive data, parallel processing tasks, high performance and reliability. In the experiment, we take in-depth analysis on the realization and improvement of the model, task scheduling, load balancing and other aspects based on the MapReduce calculation model. How to improve the Wi-Fi location algorithm from the traditional ones on the Hadoop cloud computing platform and how to use the MapReduce idea to realize the analysis and processing of the massive location information are the innovation and difficulty points of this paper.Based on the Hadoop cloud platform, this paper analyzes and studies the application of MapReduce parallel processing technology in cloud computing to the implementation and improvement of Wi-Fi location algorithm. In this paper, the research status of Wi-Fi positioning system is introduced firstly and a lot of research is emphasized on the algorithm of ranging location based on Wi-Fi. Then, with the previous research, we introduce the concept of local linear regression and position weighting and propose an improved Wi-Fi location algorithm, which is based on MapReduce and applied to the designed experimental system. At last, the experiment environment of Hadoop cluster is set up, on which the function and perforamance of the algorithm are tested. The analysis of the test results shows that the improved Wi-Fi location algorithm has reliable localization result and good execution efficiency, and can complete the positioning process in a short time when a large amount of data is flooded into the system.