Research On Key Technologies Of Energy Self-sufficient And Low Power Wireless Sensor Network

Wireless Sensor Network (WSN), the basis of "The Internet of things" technology, enhances people's ability to acquire information. It connects the information between objects as well as that between objects and people. WSN can overcome the spacial obstacles, providing accurate, prompt and effective information. At present, Wireless Sensor Network has been applied in many areas about industry and people's livelihood and it enjoys potential applications. The development of WSN will promote "The Internet of things" and comprehensively enhance the information perception, information interoperability, intelligent decision-making abilities in industry and daily life. WSN's development will also help our contry to compete for commanding heights in international science and technology, accomplish the strategic aim of establishing the innovative country, and develop our country to achieve a substantial development. In order to ahieve an advantage in technological competition in this area both home and abroad, we should research the key technology with intellectual property in accordance with the practical applications of Wireless Sensor Network.In so doing, we can enhance our core competence in the key technology.This paper is based on the research results of two programmes. One is SOPC-based low energy consumption WSN nodes design and applied research of networking (08ZCKFGX00500), the key project of Tianjin science and technology support programme. The other is "Research on the multi-Vidor parameter optical fiber diffraction grating wireless sensor network" (2006AA01Z217), the project of national high technology research and development programme. Based on the profound analysis about the key technological development in this area, the paper makes a deep research on issues such as node low-power core design, security mechanism, power supply mode and the applications of compressed sensing theory. The paper also puts forward the energy self-sufficient low-power Wireless Sensor Network, which is based on data collection of compressed sensing theory. This provides a basis for the large-scale application of WSN. This paper designs the WSN node dedicated processor based on transmission trigger mechanis and the 16-bit RISC construction system. The instruction bus is separated from data bus. Based on the processor core design of arithmetic logic unit, instruction and data memory, general-purpose registers, program pointer and stack pointer, the paper extends many general interface modules like SPI, UART and 12C to meet the needs of different peripheral interconnection. What's more, through extending dynamic power management module and using multi-processor node of dynamic switching, the paper designs the low-power node processor. In accordance with the network security, the paper analyzes the features of WNS. The specific constraints of its security mechanism are used to optimize nodes'encryption and decryption algorithm. The paper takes SOPC as the platform to make design the co-processor based on RC5 encryption and decryption algorithm.Starting from the energy consumption of sensor nodes, the paper designs solar energy-based energy self-sufficient WSN node energy management system. The system uses MPPT algorithm for low power and takes super capacitor as storage of system energy. It adopts energy monitoring mechanism, adapting to the working state of management system which manages the collectors according to light conditions, so as to increase solar energy utilization. The self-energy homeostasis equation of solar energy is analyzed to establish the original model of sensor nodes. The experiments suggest that the use of solar energy collector can supplement effective energy to sensor nodes, thus greatly prolonging the lifetime of network.According to the data collection process of large-scale WSN, the paper introduces compressed sensing theory in a creative way. In accordance with various types of data in network, a compressed sensing data cycle collection model is established, which can initiate and passively motivate abnormal inspecting mode, precisely restore network data, greatly optimize the performance of network and enhance the practical application value of WSN. The current data collection and data fusion technologies of WSN are analyzed in this paper. On the basis of a deep research on relevant key theories and technologies of compressed sensing, the paper adopts the specific features of WSN and constructs many WSN data collection modes which are based on compressed sensing, using random measurement matrixes to realize dmensional reduction in collecting large amounts of data in network, achieving optimal balancing of node load so as to greatly enhance the stability of network. In this paper, Bayesian compressive sensing theory is used to test sparse signals in WSN. Fast relevance vector machine algorithms are used and improved. Maximum marginal likelihood estimation is used to estimate sparse coefficients so as to finish the original construction of measuring signals quickly and accurately. The paper uses MATLAB to verify the accuracy and the effectiveness of reconstruction algorithm. The paper also use network simulation platform to verify the performance of new network data collection.