Storage Of The Data From Wireless Signal Detector And Neural Network Spectrum Estimation

With the development of wireless communication and satellite communicationtechnology, it has become more and more important in the real-time monitoring of thewireless signal. In modern communication systems, signal spectrum estimation isimportant. It can help people to extract the useful information of the signal to beanalyzed.This paper is based on project named “The Development of the Wireless SignalDetector”, the signal is collected by the radio frequency (RF), and then converted to70MHz intermediate frequency (IF) signal. The intermediate-frequency signal issampled by the A/D converter and digital down conversion (DDC) by FPGA and thenturn to baseband signal, after the filtering process, the signal can be analyzed onDM6446platform and the computer. With the combination between modern spectralestimation theory and the neural network, the related algorithm and extract signalinformation can be finished to meet the project needed.This paper uses the Wireless Signal Detector as hardware platform. Firstly, thehardware environment is introduced and a new method for large volume data storageis developed, which can present the efficient transmission of signal. Then, based onthe idea of software radio, data can be analyzed got from Wireless Signal Detector,some different frequently-used methods and the related parameter estimationalgorithms are compared. At last, according to the characteristic of BP neural network,a new method is proposed based on signal spectrum estimation of neural net. Bysimulating the practical data on MATLAB, the factor of neural network can be trainedto extract the signal spectrum. The proposed algorithm is presented by C language andimplemented on the Wireless Signal Detector.