The Design And Implementation Of Spectrum Analyzer Based On Embedded System

With the rapid development of electronics and communication, the technology of digital signal processing is widely used in various fields. In the process of digital signal analysis and processing, an important measuring instrument will be usually used-a spectrum analyzer. Spectrum analyzer can be used to show users the information of each frequency point of signal directly. By analyzing the sampled information we can determine whether the signal is subject to noise, so that to make improvements in time. Traditional spectrum analyzers are mainly based on analog circuits, and more filters are used, making the system complex, bulky, and inaccuracy. The high price also makes its applications limited. Thus, the traditional spectrum analyzer cannot meet the needs of the current users.The design of spectrum analyzer is suggested in the thesis, which is based on embedded microprocessor platform and Fast Fourier Transform, aiming at the low-frequency voice signal (0-20 kHz) which the human’s ears can identify. Compared to traditional spectrum analyzer, embedded microprocessor is used as core controller of spectrum analyzer, so it can reduce the production cost greatly. Its smaller bulk makes spectrum analyzer more portable, and its low power consumption makes itself can be powered by battery. The rich peripheral interfaces of ARM platform allow spectrum analyzer with more extensions in the future. The Algorithm of Fast Fourier Transform has been used in this design to improve the speed of spectrum analysis greatly.The main idea of this thesis is to introduce the design and implementation of a spectrum analysis system by using embedded ARM as micro-controller core and μ-C/OS-II operating system as the platform of software. The part of hardware mainly discusses the design of core circuit of micro-controller of STM32 which based on the core of Cortex-M3, signal generating circuit of DDS, data acquisition circuit of 8 bits A/D, protection circuit based on diode, the amplitude adjusting circuit based on programmable amplifier, the displayed circuit etc. Its platform of software has been based on the hardware above, μ C/OS-II operating system controls each task appropriately, and it can improve the stability of the spectral analysis system and reduce the difficulty of development, while providing convenience for upgrades of future. Then the programming design of the functional modules’drivers are introduced in the thesis, including A/D sampling, FFT algorithm and results displaying.Finally, by doing some experiments, debugging some problems about it, then we make the conclusion that all functions of this spectrum analysis system have been realized, and it runs in good condition. The results of spectral analysis are accurate, they can be saved by screen shots. The measuring range of low-frequency signal is from 20Hz to 25 kHz. And it only takes 1.768ms to complete a FFT of 1024-point.