Table 4 Overview of variable specification and corresponding share of observations Variables Percentage of observations S.D. Definition Household structure Single person 1 12.98 0.44 Household consists of 1 adult, single.
○ denotes state and □ denotes observation variable.
The observation variable is distributed according to p y t |x t ), defined as the observation likelihood probability.
The output of BLE fingerprint positioning and the estimated speed and heading are selected as the observation variable to construct the equation of observation below.
A hidden Markov model (HMM) is a discrete-time stochastic process, with state variable in some discrete space, and an observation variable, such that (3).
It is convenient to assume a three-dimensional configuration with observation variable x = (x,y,z), source variable ( {varvec{xi}} = (xi,eta,zeta ) ), and time ( t ge 0 ).
To decrease the nonlinearity of the measurement equation and make the filter more robust, the heading estimation difference is deployed as the observation variable.
In order to reduce the nonlinearity of the measurement equation, we deploy the difference between user heading derived from the landmark and estimation result of RMPCA as the observation variable.
To decrease the nonlinearity of the measurement equation and make the filter more robust, we deploy the difference between user heading derived from the landmark and estimation result of PCA-based approach as the observation variable.
Each observation variable is also ternary valued.
The biggest advantage of the model is that it allows the testing of the relationship between observation variable and latent, and the latent variables under the situation in which there are measurement errors.
