针对空气中温度差值难以捕捉的问题,以空气中温度分布的可视化作为研究对象,采用基于最大事后概率的最大似然估计算法,研究空气中温度分布图像化问题。可视化测量系统中,在被测区域设置32个收发分离的超声波换能器,基于一发多收模式实...针对空气中温度差值难以捕捉的问题,以空气中温度分布的可视化作为研究对象,采用基于最大事后概率的最大似然估计算法,研究空气中温度分布图像化问题。可视化测量系统中,在被测区域设置32个收发分离的超声波换能器,基于一发多收模式实现渡越超声信号数据采集,通过实验获取16×16=256个渡越时间参数TOF(Time of Flight)。实验系统采用测量角度插补与渡越时间参数平行插补两种方法进一步补充成像所需渡越时间参数,确保重建图像可读性。对实验数据进行了基于最大似然估计算法的超声波CT图像重建,重建图像结果能成功分辨空气场温度值差异。实验结果表明,基于最大似然估计算法实现空气中温度差异可视化的有效性。展开更多
To estimate the spreading sequence of the direct sequence spread spectrum (DSSS) signal, a fast algorithm based on maximum likelihood function is proposed, and the theoretical derivation of the algorithm is provided. ...To estimate the spreading sequence of the direct sequence spread spectrum (DSSS) signal, a fast algorithm based on maximum likelihood function is proposed, and the theoretical derivation of the algorithm is provided. By simplifying the objective function of maximum likelihood estimation, the algorithm can realize sequence synchronization and sequence estimation via adaptive iteration and sliding window. Since it avoids the correlation matrix computation, the algorithm significantly reduces the storage requirement and the computation complexity. Simulations show that it is a fast convergent algorithm, and can perform well in low signal to noise ratio (SNR).展开更多
文摘针对空气中温度差值难以捕捉的问题,以空气中温度分布的可视化作为研究对象,采用基于最大事后概率的最大似然估计算法,研究空气中温度分布图像化问题。可视化测量系统中,在被测区域设置32个收发分离的超声波换能器,基于一发多收模式实现渡越超声信号数据采集,通过实验获取16×16=256个渡越时间参数TOF(Time of Flight)。实验系统采用测量角度插补与渡越时间参数平行插补两种方法进一步补充成像所需渡越时间参数,确保重建图像可读性。对实验数据进行了基于最大似然估计算法的超声波CT图像重建,重建图像结果能成功分辨空气场温度值差异。实验结果表明,基于最大似然估计算法实现空气中温度差异可视化的有效性。
基金supported by Joint Foundation of and China Academy of Engineering Physical (10676006)
文摘To estimate the spreading sequence of the direct sequence spread spectrum (DSSS) signal, a fast algorithm based on maximum likelihood function is proposed, and the theoretical derivation of the algorithm is provided. By simplifying the objective function of maximum likelihood estimation, the algorithm can realize sequence synchronization and sequence estimation via adaptive iteration and sliding window. Since it avoids the correlation matrix computation, the algorithm significantly reduces the storage requirement and the computation complexity. Simulations show that it is a fast convergent algorithm, and can perform well in low signal to noise ratio (SNR).
