CAJ | 학술논문

Rao-Blackwellized粒子滤波虽然适合系统状态包含线性高斯分量的非线性状态估计，但是由于其计算量较大，不适用于实时性较高的被动目标跟踪情况。针对Rao-Blackwellized粒子滤波的不足，提出了改进的Rao-Blackwellized粒子滤波算法用于WSN被动目标跟踪。新的算法由一个粒子滤波和一个卡尔曼滤波组成，在执行过程中，粒子滤波和卡尔曼滤波相互交换信息，并行运行。计算机仿真结果表明，新的算法能够更好地减少计算量，提高跟踪的实时性。
Rao-Blackwellized입자려파수연괄합계통상태포함선성고사분량적비선성상태고계，단시유우기계산량교대，불괄용우실시성교고적피동목표근종정황。침대Rao-Blackwellized입자려파적불족，제출료개진적Rao-Blackwellized입자려파산법용우WSN피동목표근종。신적산법유일개입자려파화일개잡이만려파조성，재집행과정중，입자려파화잡이만려파상호교환신식，병행운행。계산궤방진결과표명，신적산법능구경호지감소계산량，제고근종적실시성。
Rao-Blackwellized particle filter has advantages in solving nonlinear estimation when the state has linear Gaussian sub-state than particle filter. However,it is limited in passive target tracking which needs high performance of real time for its computational complexity. A new modified Rao-Blackwellized particle filter consist of a particle filer and a Kalman filter is proposed in the paper and be applied in passive target tracking with WSN. At each time,the particle filter and the Kalman filter exchange information with each other for estimation at next time. The simulation results prove that the new algorithm can reduce the computation complexity and improve the performance of real time of passive tracking.