Research On Key Techniques Of Receiver For CPM Signals In Wireless Fading Channels

Continuous phase modulation(CPM) is an excellent digital modulation scheme with constant modulus and continuity phase. It is used widely in modern mobile, satellite communications and military tactical communication thanks to its attractive spectral and power efficiency. In recent years, much attention is paid to it in communication reconnaissance system. While CPM is a non-liner modulation scheme, many conventional signal processing methods used in liner modulation are not suitable for it. So the receiving and processing of CPM signals are always difficult in signal processing fields. Modern wireless communication, which is characterized by high transmission rate and low power, and the complexity of wireless channels bring great challenge to receiving of CPM signals. Especially, the received signal is distorted by channel fading, which causes more difficulties for signal analyzing in non-cooperative reception conditions such as electromagnetic spectrum monitoring, signal reconnaissance and electronic countermeasure. So the study of key technologies of blind demodulation of CPM signals in wireless fading channels has important theoretical significance and practical value.This dissertation is devoted to a study of receiving CPM signals high efficiently in wireless fading channels at low signal-to-noise ratio, based on the background of modern wireless communication. Through the analysis of equalization, synchronization and iterative decoding, the corresponding receiver programs have been established for different task requirements and solving the existing problems. The key technologies including Turbo frequency domain equalization for CPM signals, blind Turbo equalization in parametric stabilization channel for CPM signals, blind equalization for CPM signals over time-varying frequency-selective channel, timing synchronization at low signal-noise ratio for CPM signals, time-varying carrier phase synchronization for CPM signals. The main works and contributions obtained in this thesis can be summarized in the following six aspects.1. In the research on the equalization of CPM signals, when the channel response is known, Turbo frequency domain equalization method based on Rimoldi decomposition for CPM signals is addressed in detail. To solve the issues of the high complexity and poor performance in the equalization of high order and partial response CPM signals, we analyze the framework of tilted phase CPM transmitted signals from the perspective of Rimoldi decomposition. A novel frequency domain equalization algorithm for high order CPM signals is proposed based on the combination of single-carrier frequency domain equalization(FDE) and Turbo equalization. This algorithm avoids the large matrix inversions in computation of the equalizer coefficients in time domain by transforming the signal equalization to frequency domain. Simultaneously, it improves the system performance by soft information iterative process. The analysis and simulation results show that the proposed algorithm provides a relatively lower computational complexity and a good performance gain in multipath fading channels with serious inter-symbol interference。2. Facing to the problem of blind equalization for CPM signals on condition that the channel is static and the channel response is unknown, we deeply study blind Turbo equalization algorithm based on hidden Markov model for CPM signals. To solve the issues of the poor performance and convergence difficulties in the blind equalization of high order CPM signals, a new EM-SOVA channel blind equalization method is developed from the perspective of hidden Markov model(HMM). A novel blind Turbo equalization algorithm for high order CPM signals is proposed based on the combination of this method and Turbo equalization. The algorithm improves the equalization performance by applying the soft-output Viterbi algorithm(SOVA) within an EM iteration. Simultaneously, a soft information iterative process is used for further improvement of the system performance at a low signal-to-noise ratio. The analysis and simulation results show that the proposed algorithm provides a good blind equalization performance and convergence.3. To solve the issues of the high complexity, poor performance and slow convergence speed in the blind equalization of high order CPM signals in time-varying frequency-selective fading channels, a new blind equalization method based on forward adaptive backward adaptive soft-input soft-output(FABA-SISO) algorithm used in linear modulation signals is developed from the perspective of bidirectional adaptive channel equalization. A novel adaptive blind equalization algorithm for high order CPM signals is proposed based on the combination of PSP and Kalman filtering. The algorithm improves the equalization performance by applying the FABA-SISO which used the past, the present and the future observation to implement Kalman filtering channel estimation. Simultaneously, a per-survivor processing is used for further improvement of the system complexity, so that the algorithm is better suitable for engineering application. The analysis and simulation results show that the proposed algorithm provides a good blind equalization performance and convergence.4. Targeting at the problem of the poor performance and false lock in the blind timing synchronization for M-ary partial-response CPM signals, a new PAM-based timing synchronization scheme is developed from the perspective of pulse amplitude modulation(PAM) decomposition. A novel blind timing synchronization algorithm for M-ary partial-response CPM signals is proposed on the basis of the combination of this scheme and timing false lock detection based on Markov chain model. The algorithm reduces the number of matched filters and trellis states by optimizing the PAM pulse. Simultaneously, an auxiliary timing error detector and a false lock detector are used to monitor the appearance of false lock in real time. The analysis and simulation results show that the proposed algorithm provides accurate and rapid synchronization under the low SNR and large initial timing error condition.5. In order to improve the performance and reduce the complexity in the time-varying carrier phase synchronization for M-ary partial-response CPM signals, a new PAM-based phase synchronization scheme is developed from the perspective of pulse amplitude modulation(PAM) decomposition. A novel phase synchronization algorithm for M-ary partial-response CPM signals is proposed on the basis of the combination of the factor graph model and EM-SPA. The algorithm reduces the number of matched filters and trellis states by adopting the principal Laurent pulses. Simultaneously, EM-SPA based on the factor graph model was used to reduce the complexity further. The analysis and simulation results show that the proposed algorithm provides accurate and rapid synchronization under low SNR and in presence of strong phase noise.6. The present algorithms of iterative detection for serially concatenated continuous phase modulation had a high computational complexity and poor convergence at low signal-to-noise ratio(SNR). The low complexity joint iterative demodulation and decoding algorithm is proposed. A modified Gaussian wavelet of fast convergence and delayed feedback control method are used to control positive feedback and improve the convergence of system effectively. Combined with hard decision-aided iterative stopping criterions, the algorithm reduces the average number of iterations. A large number of matched filters are required to compute a priori information of codeword in iterative decoding. Because of large condition number, effective matched filters have a strong correlation. So truncated small eigenvalues method is used to reduce the number of matched filters. The analysis and simulation results show that the proposed algorithm exhibits a lower complexity and better convergence.