On Expansion And Optimization Of Efficient Modulation Systems

Radio frequency spectrum is related to the interests of national political, economic, military and other aspects. As wireless applications continue to grow rapidly, spectrum resources have almost been exhausted. Therefore, the transmission system, with high spectrum and energy efficiency, is needed to solve the spectrum shortage and achieve the robust communication. It also has direct economic benefits and strategic significance.In order to make spectrum efficient system practical, the theory and technology of the spectrum efficient modulation are still to be developed. The spectrum efficient system is expanded and optimized on several aspects, such as improving spectral efficiency and demodulation performance, meeting the practical application requirements, designing new system under strictly limited band. Main research problems and innovations are summarized as follows:1. The existing M-ary phase position shift keying (MPPSK) modulation is improved. Compared with the existing system, the detection performance of the improved MPPSK signal is improved. The effect by different modulation parameters for the improved modulation is analyzed.2. The extended binary phase shift keying (EBPSK) modulation signal is applied to suppress the asynchronous interference of multi-radar system. EBPSK signal is expected to improve the continuous tracking level of secondary surveillance radar, and optimize system detection performance by the matching filtering with double symbols.3. The MPPSK modulation is expanded to transmit by the dual quadrature carriers to double the spectral efficiency. Then, the novel co-channel full-duplex system with single antenna based on the improved MPPSK modulation is proposed, and the optimal band pass filter (BPF) for the coherent receiver is designed. What’s more, the self interference cancellation of the system gains higher isolation than the existing full-duplex system.4. The special impacting filter (SIF) and narrow band pass filter (NBPF), as the shaping filter in the transmitter, are used to strictly limit the band. Under the premise of guaranteed demodulation performance, the order of shaping filter is reduced.5. To further improve the demodulation performance of the strictly band-limited system based on spectral efficient modulation, combining with the advantages of deep learning algorithms, the deep learning algorithm is introduced into the feature extraction of the strictly band-limited signaling. The efficient demodulation structure based on the stacked denoising sparse auto encoder (DSAE) algorithm for the strictly band-limited signaling is presented. The proposed structure, with excellent performance, makes the fast than Nyquist (FTN) communication become possible. The detection performance of different deep learning algorithms is compared. In Gaussian channel and Rayleigh fading channels, the detection performance of stack DSAE network is also analyzed respectively.