Algorithm 1 Human activity recognition method with SDAE.

Input: Raw dataset D

Output: Activity types of testing dataset

1: Data Preprocessing:

2: Segment the dataset according to sampling frequency

3: Apply the random oversampling

4: Standardize the dataset to obtain input vector $x_i$

5: Divide the dataset into training dataset $D_{train}$, validation dataset $D_{validate}$, test dataset $D_{test}$

6: Pretraining:

7: while $l<=$ Hidden layers $N_l$ do

8:  The dataset $x_i$ in $D_{train}$ is corrupted into $x_i^{{}^{\prime }}$ by adding a denoising factor. Then let $x_i^{{}^{\prime }}$ as input to train l-th layer of stacked denoising autoencoder.

9:  The output of l-th layer will be the input of l + 1-th layer

10:  l += 1

11: end while

12: Fine-tuning:

13: Fine-tune the whole network by applying backpropagation. Utilize labeled dataset $D_{train}$ to train softmax layer.

14: Test:

15: Use the $D_{train}$ and $D_{validate}$ to train model and validate performance of model respectively.   Recognize the activity type of test data $D_{test}$.