The Research On Test Methods Of RFID Signal Modulation

RFID (Radio Frequency Identification) is a newly invented automatic identification technology. Compared with other identification technologies, RFID has many advantages. As the rapid development of electronic and manufacturing technology, RFID will become more and more useful in many fields, such as purchasing and distribution of the goods, commerce and trade, manufacturing, identification, security and phisical application. As a key part of RFID system, the testing methods of RFID have received great attention. And it has become an important and urgent task to investigate on the testing methods. In fact, the testing methods of RFID have great theoretical and practical value for the RFID system. This thesis mainly discusses about the testing method for RFID signal modulation.Software Radio develops at an amzing speed in the recent years. First, the thesis introduces the basic theories and the three architectures of software radio. According to the present designing condition, the wideband and band-pass sample architecture is proved to be the most practical one. Based on the architecture, the general model of multi-mode modulation is proposed. Then the thesis discusses about the three traditional method of digital modulation. When the channel condition is good, the parameter histogram distribution method is introduced to make a statistics of the three basic character parameters. The thesis theoretically proves that almost all the amplitude modulation signals could be classified with the parameter R. If the channel condition is not so good, the decision theory method which realizes easily and fits to analyze online is adept to recognize the modulation signals. Based on the discussion, this thesis proposes a characteristic parameter Ra . Then a new testing method for RFID signal modulation which is based on software radio is proposed. By using 3 characteristic parameters and a classifier based on decision tree and artificial neural networks, the thesis could make correct identification among 5 modulation types, including 2ASK,4ASK,2FSK,4FSK and BPSK. The thesis discusses about the design procedure and presents the simulation results. The results of the experiments proves that this method recognizes the signals accurately in real time with a low SNR.