时频双选水声信道下高谱效OQAM-OFDM通信的关键技术研究

Underwater acoustic (UWA) channel is quite band-limited with time-frequency doubly selective fading. It is thus very difficult to achieve reliable UWA communication in high data rate. Compared to traditional OFDM (Orthogonal Frequency Division Multiplexing), OQAM-OFDM ( Offset Quadrature Amplitude Modulation - OFDM ) is more spectrum-efficient, and less sensitive to both time-selectivity and frequency selectivity of the UWA channels. Thus OQAM-OFDM may have a potential use in UWA communications in the future. In the literature about OQAM-OFDM, it is often assumed that the channel is time-invariant. However, for UWA channels, the changing of the channel within the symbol time can’t be neglected. Aiming at this problem of UWA channels, this project is devoted to advance key technologies in synchronization, channel estimation and channel equalization for OQAM-OFDM system. There are three scientific problems to be solved as follows: (1) timing metric function which would exploit the multipath and be immune to the carrier frequency offset; (2) reducing the dimensions of the channel parameters and building estimation model for the channel parameters. (3)building objective function for the frequency-domain equalizer based on the minimum symbol error rate criterion. The results of this research would provide theoretic guidance for the use of OQAM-OFDM in UWA communications.

水声信道带宽很窄，且呈现时间-频率双重选择性衰落，实现高速可靠通信的难度大。相比于传统OFDM（正交频分复用）技术，OQAM-OFDM（基于偏移正交幅度调制的正交频分复用）技术具有频谱效率更高、抗时频双选衰落能力更强的特点，对于水声通信具有潜在的研究价值。现有OQAM-OFDM研究一般针对时不变信道，在水声信道下还须考虑信道在符号时间内变化的问题。针对该问题，本项目研究水声OQAM-OFDM通信的关键支撑技术，包括同步、信道估计及信道均衡。本项目需要解决三个科学问题：（1）抗尺度压扩、抗频偏并可利用多径的符号同步测度函数；（2）信道参数降维及信道参数估计模型的建模；（3）频域均衡的最小误码率目标函数及其最优解。本项目的研究成果有望为OQAM-OFDM技术在水声通信中的应用提供有力的理论支撑。

水声双选信道下实现OQAM-OFDM高速通信的相关支持技术亟待研究。本项目针对双选水声信道各多径分量路径延迟、尺度压扩系数、衰减系数各不相同的特点，深入分析OFDM水声通信系统在该类信道的子载波间干扰特征，在理论上证明了子载波间干扰系数在统计意义上随着子载波频率的升高而单调递增，并得到了系统遍历容量的闭合表达式。以系统遍历容量最大化为目标，提出了一种OFDM系统在时频双选水声信道下的功率分配算法，该算法可为实现OQAM-OFDM水声通信提供功率分配算法上的理论指导。项目组还就高鲁棒性信号帧同步、信号峰均比抑制、误码性能提升等技术展开了深入的研究。通过项目的实施，项目组成员学术水平得到显著提升，发表学术论文十余篇，其中SCI索引8篇，2名成员获得了博士学位。相关研究成果可为OQAM-OFDM在水声信道下的应用提供关键技术支持。