Research On Linearization Technologies For Front-end Parts Of The Transmitter And Receiver

The transmitter and receiver are the indispensable parts of communication systems,meanwhile, their nonlinear characteristics have severe impact on the communicationquality and efficiency. For the transmitter, improving the linearity of the poweramplifier is the focus of research, especially when the non-constant envelopemodulation schemes have been widely used in today’s communication system. In thiscase, the memory and nonlinear characteristics of the power amplifier can be furthersimulated, resulting in in-band distortion and out-band spectral extension for the outputsignal. For the receiver, nonlinearity caused by the mixer, amplifier, filter and the ADCalso cannot be ignored. They may cause distortion of the received signal, reducespurious-free dynamic range of the receiver and impact the ability of the receiver to thedetection of weak signals.With the theory of nonlinear systems and adaptive filter technology, the detailedtheories and specific implement methods of the digital predistortion technology andblind compensation technology are deeply discussed in this dissertation. Thecontribution and innovation points of this dissertation are summarized as follows:1. Systematically summarizes the basic theories and research methods of the fieldof transmitter and receiver linearization, as well as the existing research results.2. Since the traditional RLS identification algorithm for digital predistorter hassome shortcomings, such as huge complexity and resource consumption, a novelidentification method based on ERLS-DCD algorithm is presented. Meanwhile, thisdissertation proposes the NiosII+multi-look-up table structure, making it possible torealize the entire predistortion system in a single FPGA. The scheme improves thesystem integration, and reduces the implementation costs.3. In the receiver linearization, introduces the advanced research on digital receiverfront-end adaptive identification and compensation techniques, including the nonlinearmodel and extraction algorithm criterion for the blind identification. On this basis, somedevelopments and further explorations are presented in the dissertation. The relatedsimulation eventually verifies the validity and reliability of the system. 4. In this dissertation, researches have been made to solve the distortion caused bythe analog-to-digital converter (ADC) in the receiver system.The meaning and testmethods of various performance indicators for ADC are discussed in detail. Postcalibration based on blind compensation method is proposed to eliminate thenonlinearity and conversion error. Since this method can greatly enhance the ADC’seffective number of bits (ENOB) and the maximum spurious-free dynamic range(SFDR) without changing its circuit or increasing the manufacturing costs, it has a verywide prospect of appliance.