中文摘要：

针对变换域通信系统收发双方电磁环境不一致、频谱失配导致有用信号能量损失和降低系统抗干扰性能的问题,提出接收端多元编码幅度谱成型算法。首先,在发送信号频谱组成未知的条件下,采用变换域技术对接收端干扰和噪声进行处理,由于干扰会在变换域能量聚焦,而噪声没有能量聚焦特性,因此变换域处理能够有效的将干扰和噪声进行分离;然后,对分离后的干扰和噪声幅度谱进行归一化处理;最后,对归一化处理的幅度谱进行多元编码,无干扰的频点幅度谱设置为1,有干扰的频点幅度谱设置为归一化幅度谱最大值与该点的差值。采用该算法生成的基函数能够有效降低有用信号能量的损失并躲避干扰。仿真结果表明,相比接收端直接生成基函数和基于类似等增益合并的算法,在信噪比为-8~8dB的条件下,多元编码幅度谱成型算法能够有效形成接收端幅度谱,基函数误差分别降低了16.2%和13.1%,提高了系统的抗干扰性能。

英文摘要：

A novel shaping algorithm of amplitude spectrums based on multiple code is proposed for the receiver to solve the problem that electromagnetic environments of a transmitter and a receiver are inconsistent and the mismatch of spectrums results in the loss of useful signal energy and the weakening of the anti-interference performance of a system. The transform domain technology is adopted to deal with the interference and noise at the receiver under the condition where the transmit signal spectrum in the receiver is unknown. Based on the fact that the energy of interference will be focused in transform domain while the noise will not, transform domain processing can effectively separate the interference and noise. Then the amplitude spectrums of the separated interference and noise are normalized, and the normalized amplitude spectrums are multiply coded. Specifically, the amplitude spectrum without interference is set to ＂1＂ and the amplitude spectrum with interference is set to the difference between the maximum value of normalized amplitude spectrum and the value of this point. The basic function generated by the proposed method can reduce the loss of useful signal energy and avoid interferences. Simulation results in the range of signal-to-noise ratio from -8 to 8 dB and comparisons with the approach in which the basic function is generated directly by the receiver and the method based on similar to equal-gain combining show that the proposed algorithm forms the receiving amplitude spectrum effectively and basis function errors are reduced by 16.2% and 13.1%, respectively. The anti- interference performance of systems is improved.