Design Of Network Base Station In Smart Water Meter Reading System

Remote meter reading system(RMRS), which makes the acquisition, transmission, management and sharing of meter data completely automated, is a comprehensive application of many modern technologies. RMRS avoids disadvantages of traditional manual meter reading such as low efficiency, high cost and household inconvenience. Under the tiered pricing situation, RMRS can be a good solution to obtain dosage and time information accurately. Also, RMRS lays the data foundation for deeper information mining in this big data age. Remote meter reading is the developing trend, while in support of state, it spreads rapidly.As the bridge between wireless sensor network(WSN) and server, concentrator plays a very important role. However, under the background that communication data and functional requirements grow explosively, and in the trend which includes equipment operation becomes more intelligent, data handling becomes localization, extension becomes more convenient and maintenance becomes more humanized, existing schemes have been difficult to meet the requirements. Therefore, this paper re-examines the concentrator design concept, decomposes concentrator into WSN root device and network base station, researches and puts forward a solution of network base station.For the design of network base station, a lot of work was done in water prediction, platform building, software design and system testing. The main researches are as follows:Firstly, the typical structure and principle of RMRS are introduced, the effect when bringing into network base station is analyzed, and a detail requirements analysis is made. According to the analysis, the overall framework is proposed from three aspects, which are modeling, platform building and software design.Secondly, this paper introduces the principle and algorithm of back propagation(BP) neural network and Markov chain. Based on practical application, the water prediction model is established by using BP neural network predict first and then correcting the BP prediction’s error with Markov chain. Also, the algorithm implementation is given.Thirdly, the detail design scheme of platform building and software is expounded. In the hardware aspect of platform, with minimum system board which based on Cortex-A8 processor as the core, this paper designs the peripheral circuits which includes Ethernet front-end port, serial port, SD card interface, etc. In the system software aspect of platform, this paper chooses Linux operating system, installs embedded Java Virtual Machine(JVM) and designs a monitoring script. In the application software aspect, communication management module, data processing module, database operation module, basic functional module, system startup module and program update subsystem are designed. The protocol compatibility and protocol extension are research priorities, also, the guarantee of data processing performance is scientific designed.Finally, this paper describes the installation and deployment information of network base station, expounds the test from functional testing, performance testing and reliability testing, then demonstrates and analyzes the results. Meanwhile, the achievements of this paper are summarized and the subsequent research is prospected.