Modulation Recognition, Signal Processing Architecture

Modulation Classification (MC) is extremely important both in commercial (such as software defined radio) and military (such as and electronic warfare) communication systems. Based on the communication signal modulation classification algorithm and software defined radio technology, this paper represents the signal processing framework which is applied in the implementation of the whole MC platform.There are 8 chapters in this paper. Chapter 1 is an introduction to MC and digital receiver technology. Chapter 2 introduces a 90-degree phase shifter which plays an important role in the whole system. Our design is based on the IIR half-band elliptic filter instead of the traditional FIR filter model. The hardware consumption is highly reduced. Chapter 3 represents a full digital automatic gain control loop based on the logarithmic model. After comprehensive analysis, we get the exact expressions of the time-varying gain and output corresponding to input signals. Chapter 4 introduces the extraction algorithms of tree key features, such as instantaneous phase, instantaneous frequency and instantaneous amplitude. We analyze the precision of implementing this algorithm in the fixed-point system and derive the theoretical system error function. Chapter 5 represents the statistical parameters and classification flow for MC. Chapter 6 introduces the demodulation methods and the implementation framework. Chapter 7 design a modulation classification platform based on the fundamental modules and the whole system is tested. Chap 8 gives a summary of the thesis and points out several prospective directions for further study.