In multi-sensor data fusion target tracking system, using information filtering can implement distributed location with uncorrelated measurement noises, but the measurement noises of different sensors are often correlated. If measurement noises are correlated, the covariance matrix of measurement noises is not a diagonal matrix. We can not use information filtering to implement distributed location with correlated measurement noises. By using the matrix theory, the covariance matrix of measurement noises can be transformed to a diagonal matrix. The observation models are transformed to new observation models, and the multi-sensor measurements with correlated measurement noises are transformed to equivalent pseudo ones with uncorrelated measurement noises. There are many methods in the matrix theory, we use Cholesky fatorization in this thesis. Cholesky fatorization is from Gaussian elimination, and there are many advantages in the computation process.However, the observation models need to be transformed to new observation models, and the measurement datas for the approach need to be separated and recombined. For measurement datas being separated and recombined, every sensor must communicate with each other. In practice, one sensor does not directly communicate with other sensors except its direct neighbors. By formulating the Cholesky factorization process, we present architectures which are applied in wireless distributed location. Distributed architectures with clustered nodes are proposed to achieve measurement exchange and information sharing for wireless location and target tracking. With limited times of data exchanges between clustered nodes, the correlated noise components in the measurements are transformed into uncorrelated ones through the Cholesky process, and the resultant information can be directly shared and processed by the derived extended information filters at the nodes in the distributed system. Hybrid TDOA/AOA wireless location systems with the NLOS error effects are used as examples in investigating the distributed information architecture. Simulation results show that the proposed distributed information processing and data fusion architecture effectively achieve improved location and tracking accuracy.

A white Gaussian noise measurement model is widely used in target tracking problem formulation. In practice, the measurement noise may not be white. This phenomenon is due to the scintillation of the target. In many radar systems, the measurement frequency is high enough so that the correlation cannot be ignored without degrading tracking performance. In this thesis, target tracking problem with correlated measurement noise is considered. The correlated measurement noise is modeled by a first-order Markov model. The effect of correlation is thought as interference, and Optimum Decoding Based Smoothing Algorithm is applied. For linear models, the estimation performances of Optimum Decoding Based Smoothing Algorithm are compared with the performances of Alpha-Beta Filter Algorithm. For nonlinear models, the estimation performances of Optimum Decoding Based Smoothing Algorithm are compared with the performances of Extended Kalman Filter by performing various simulations.