Blind Receiver Technologies For Software Radio

Blind receiver is one of the key technologies of software radio. It can improve the spectral efficiency and robustness of communications. It is also important in military area. This paper focuses on the technologies involved in blind receiver,such as signal sampling,blind equalization,symbol synchronization and carrier synchronization. A software radio blind receiver implemented on a high speed DSP platform is also introduced.Chapter 2 firstly introduces several important specifications of A/D converter. The key problem of signal sampling in software radio is then discussed. At the end of the chapter,an application of band-pass sampling used in a radio receiver is described.Chapter 3 is on blind equalization. Different types of equalizer structures and algorithms are introduced and analyzed.' The algorithms based on CMA or Gaussian cluster formation soft decision directed algorithm are discussed in detail. A stop-and-go multi-modulus blind equalizer and an adaptive step-size soft decision directed blind equalizer are proposed. Simulation results show that the proposed algorithms have good performances.In chapter 4,the sampling rate conversion structure and the related filters are discussed. A new scheme of non-integer-factor sampling rate converter using poly-phase structure is given. The structure analysis and design of interpolation filter for symbol synchronization are introduced,based on which a poly-phase structure of interpolation filter with high accuracy and low complexity is presented. Lastly,the ML criterion is used to analyze the NDA timing offset estimation algorithms. Two symbol synchronization methods that use feedback and feed-forward timing offset estimation algorithms are discussed and simulated.Based on the ML criterion,Chapter 5 analyzes some carrier recovery algorithms and proposes a NDA frequency offset estimation algorithm. The ML estimator of carrier phase and the improved algorithms are also discussed. What's more,some improved carrier synchronization algorithms for high order QAM signals are described. Several important algorithms are simulated. In the simulation results,the proposed algorithm shows good performances.Chapter 6 analyzes some blind DFE algorithms and proposes to cascade a super-exponential equalizer with a blind DFE. Combined with the carrier synchronizer proposed in chapter 5,a blind DFE structure is presented. Simulation results show that the presented DFE structure can deal with serious channel distortion and frequency offset.Chapter 7 introduces and analyzes some timing error detection and feed-forward timing recovery algorithms for GMSK signal. The auto-correlation method is improved by using a post-filter. Simulation results show that the improved method has better estimation accuracy. The frequency offset estimation algorithms for GMSK signal are also discussed and a frequency offset estimator with larger estimation range and higher accuracy is proposed. Two kinds of synchronization structures are compared with each other. At last,the synchronizer described above and a differentially coherent demodulator compose a blind GMSK receiver.Chapter 8 describes the structure of the software radio blind receiver implemented on a DSP platform. Some related parameters are also introduced.