發(fā)布時間：2019-04-04 10:32

【摘要】：無線電頻譜關系到國家政治、軍事、經(jīng)濟等各方面的利益,已經(jīng)成為國家的重要戰(zhàn)略性資源。在通信業(yè)務快速增長的現(xiàn)在,人們普遍意識到在這種靜態(tài)固定的頻譜分配模式下可用的無線電頻譜已經(jīng)快要耗盡了。因此,同時具有高頻譜利用率和高能量利用率的高效傳輸體制,能最大限度地壓縮無線傳輸頻譜,是解決頻譜短缺和實現(xiàn)高效通信的有效途徑,具有直接的經(jīng)濟效益和重要的戰(zhàn)略意義。為了使高效調(diào)制通信系統(tǒng)走向?qū)嵱?高效調(diào)制理論和技術仍需發(fā)展和完善。本論文重點在提高頻譜利用率和解調(diào)性能、解決實際應用需求、構(gòu)思新型體制和施加嚴格限帶幾方面對高效調(diào)制系統(tǒng)進行了拓展與優(yōu)化。論文主要工作和創(chuàng)新性成果如下：1.優(yōu)化了現(xiàn)有的多進制相移鍵控(MPPSK)調(diào)制方式,提高了MPPSK調(diào)制的解調(diào)性能,分析了不同調(diào)制參數(shù)對調(diào)制信號功率譜的影響；2.將擴展的二元相移鍵控(EBPSK)調(diào)制用于多基地雷達的異步干擾抑制,實現(xiàn)了多基站的測速、測距,可提高二次監(jiān)視雷達連續(xù)跟蹤時的自動化水平,并通過雙碼元匹配濾波提高了系統(tǒng)的檢測性能；3.將MPPSK調(diào)制拓展到正交載波雙路傳輸,使得頻譜效率倍增；進而提出了一種新穎的基于改進MPPSK調(diào)制的共信道全雙工系統(tǒng),優(yōu)化設計了其相干接收機最佳帶通濾波器,得到了超出現(xiàn)有同時同頻全雙工系統(tǒng)的數(shù)字域收發(fā)隔離度；4.提出了利用沖擊濾波器加窄帶帶通濾波器作為發(fā)送端成型濾波器限制帶寬,在保證一定解調(diào)性能的前提下,降低了成型濾波器的階數(shù)；5.為提高嚴格限帶高效調(diào)制系統(tǒng)的解調(diào)性能,結(jié)合深度學習網(wǎng)絡算法,將深度學習算法和網(wǎng)絡引入嚴格限帶高效調(diào)制信號的特征提取中,提出了基于堆棧降噪稀疏自編碼(DSAE)學習算法的嚴格限帶信號的高效解調(diào)結(jié)構(gòu),解調(diào)性能表現(xiàn)優(yōu)異,為實現(xiàn)超奈奎斯特速率通信提供了一種可能；對比了多種深度學習算法的檢測性能,并在高斯信道、瑞利衰落信道中分析了堆棧DSAE學習網(wǎng)絡的檢測性能。
[Abstract]:Radio spectrum is related to the national political, military, economic and other interests, and has become an important strategic resource of the country. With the rapid growth of communication services, it is widely recognized that the radio spectrum available in this static fixed spectrum allocation mode is running out. Therefore, an efficient transmission system with both high spectrum efficiency and high energy efficiency can compress the spectrum of wireless transmission to the maximum extent, and it is an effective way to solve the spectrum shortage and realize efficient communication. It has direct economic benefits and important strategic significance. In order to make the high-efficiency modulation communication system practical, the theory and technology of high-efficiency modulation still need to be developed and perfected. In this paper, the high efficiency modulation system is expanded and optimized in the aspects of improving the spectrum efficiency and demodulation performance, solving the practical application demand, conceptualizing a new system and imposing strict band limit. The main work and innovative results of this paper are as follows: 1. The existing multi-ary phase shift keying (MPPSK) modulation is optimized, the demodulation performance of MPPSK modulation is improved, and the influence of different modulation parameters on the modulation signal power spectrum is analyzed. The extended binary phase shift keying (EBPSK) modulation is used to suppress asynchronous jamming of multi-static radar, and the speed measurement and ranging of multi-base stations are realized, which can improve the automation level of continuous tracking of secondary surveillance radar. At the same time, the detection performance of the system is improved by dual-symbol matched filtering. 3. The MPPSK modulation is extended to the orthogonal carrier dual-channel transmission, which makes the spectrum efficiency multiply. Furthermore, a novel co-channel full-duplex system based on improved MPPSK modulation is proposed, and the optimal band-pass filter of its coherent receiver is optimally designed, and the digital domain transceiver isolation beyond the existing simultaneous co-frequency full-duplex system is obtained. 4. In this paper, the impulse filter and narrow band pass filter are used to limit the bandwidth of the shaping filter at the transmitter end, and the order of the shaping filter is reduced under the premise of certain demodulation performance. 5. In order to improve the demodulation performance of the band-limited high-efficiency modulation system, combined with the deep learning network algorithm, the depth-learning algorithm and the network are introduced into the feature extraction of the strictly band-limited high-efficiency modulation signal. An efficient demodulation structure of strictly band-limited signal based on stack de-noising sparse self-coding (DSAE) learning algorithm is proposed. The performance of demodulation is excellent, which provides a possibility for super-Nyquist rate communication. The detection performance of multiple deep learning algorithms is compared, and the detection performance of stack DSAE learning network is analyzed in Gao Si channel and Rayleigh fading channel.
【學位授予單位】：東南大學
【學位級別】：博士
【學位授予年份】：2016
【分類號】：TN914

【參考文獻】

相關期刊論文 前3條

1 王紅星;王洪利;毛忠陽;趙志勇;;基于循環(huán)譜特性的擴展二元相移鍵控解調(diào)算法[J];電波科學學報;2010年05期

2 馮熳;吳樂南;;統(tǒng)一的BPSK信號功率譜分析[J];電氣電子教學學報;2008年06期

3 宋軍,曹蕾,錢建生,朱乃鵬;多進制數(shù)字基帶信號的頻譜分析[J];通信技術;2003年06期

相關博士學位論文 前1條

1 宗華;雙基地高頻雷達數(shù)據(jù)處理NFE技術研究[D];哈爾濱工業(yè)大學;2010年

相關碩士學位論文 前2條

1 李旭森;向下兼容的數(shù)字調(diào)幅廣播系統(tǒng)設計與實現(xiàn)[D];東南大學;2015年

2 朱超越;向下兼容的數(shù)字調(diào)幅廣播算法與設計[D];東南大學;2015年

，

本文編號：2453728