# Table 1 Pseudo-code for the particle filter-based ATR algorithm

• Initialization: Draw ${\mathbf{X}}_{0}^{j}~N\left({X}_{0},1\right)$, and ${\alpha }_{0}^{j}={\alpha }_{0}$ j {1,..., N p }. Here X0 and α0 are the initial kinematic state and identity values, respectively.

• For t = 1,..., T (number of frames)

1. For j = 1,..., N p (number of particles)

1.1 Draw samples ${\mathbf{X}}_{t}^{j}~p\left({\mathbf{X}}_{t}^{j}\mid {\mathbf{X}}_{t-1}^{j}\right)$ and ${\alpha }_{t}^{j}~p\left({\alpha }_{t}^{j}\mid {\alpha }_{t-1}^{j}\right)$ as in (10) and (11).

1.2 Compute weights ${w}_{t}^{j}=p\left({\mathbf{z}}_{t}\mid {\alpha }_{t}^{j},\phantom{\rule{2.77695pt}{0ex}}{\mathbf{X}}_{t}^{j}\right)$ using (12).

End

2. Normalize the weights such that ${\sum }_{j=1}^{{N}_{p}}{w}_{t}^{j}=1$.

3. Compute the mean estimates of the kinematics and identity ${\stackrel{^}{\mathbf{X}}}_{t}={\sum }_{j=1}^{{N}_{p}}{w}_{t}^{j}{\mathbf{X}}_{t}^{j}$ and ${\stackrel{^}{\alpha }}_{t}={\sum }_{j=1}^{{N}_{p}}{w}_{t}^{j}{\alpha }_{t}^{j}$.

4. Set $\left[{\alpha }_{t}^{j},\phantom{\rule{2.77695pt}{0ex}}{\mathbf{X}}_{t}^{j}\right]=\text{resample}\left({\alpha }_{t}^{j},\phantom{\rule{2.77695pt}{0ex}}{\mathbf{X}}_{k}^{j},\phantom{\rule{2.77695pt}{0ex}}{w}_{k}^{j}\right)$ to increase the effective number of particles .

• End 