摘要
对低信噪比微弱运动目标的线谱轨迹检测问题,提出了基于动态规划的线谱增强方法。这里将信号线谱增强问题转化为一个多阶段决策的优化问题。首先建立包含运动目标频谱状态变化的系统状态方程,其次依据广义似然比和贝叶斯原理计算递归形式的状态转移得分泛函,利用最小贝叶斯风险准则和虚警指标判断目标数得到对应线谱轨迹。算法利用单一优化过程完成数据互联和微弱运动目标线谱轨迹探测,具有增强处理非高斯噪声的能力,并能有效减少计算量;通过仿真和海试数据验证,表明所建模型合理,理论分析正确。
A line spectrum enhancement method is provided to improve 2D lofargram feature by dynamic programming （DP）. The problem of line spectrum enhancement can be regarded as being the solution of a multi-stage optimization procedure. First, the system state equation including Doppler frequency shift is created. Then, the score functional based on the Log-likelihood ratio （LLR） and Bayes principle is obtained. At last, target frequency shift trace is shown by analyzing the max-value of the score functional. The DP algorithm has the property of reducing computational requirements, and target detection and tracking are combined into a single optimization procedure, and can deal with the non Gaussian noise problem. The simulation and sea trial results indicate that 2D lofargram line spectrum feature in low signal-to-noise ratio （SNR） can be improved by the dynamic programming process.
出处
《应用声学》
CSCD
北大核心
2011年第3期193-201,共9页
Journal of Applied Acoustics
基金
国防装备科研计划项目(4010503010103)资助
关键词
动态规划
线谱增强
Dynamical programming process, Line spectrum enhancement
