Differential Privacy Location Protection Method Based on the Markov Model

Algorithm 2

Location protection based on Differential privacy (LPADP).

Input: ; // location from Algorithm 2

Output: The two locations with the largest prediction probability are protected by adding noise

1. Constructing LPT;

2. void fun(int , int , int ) // select the two nodes with the largest prediction probability on LPT(X₁,X₂)

3. {

4.  int i;

5.  ;

6.  for(i=1; ; i++)

7.  if() ; // select the nodes with the highest prediction probability X₁

8.  ;

9.  for(i=1; ; i++)

10.  {

11.   if( && )

12.   ; // select the next largest value node X₂

13.  }

14.  ;//output X₁, X₂

15. }

16. ; //ε₁<ε₂

17. // Laplacian noise is added to the two location nodes with the largest prediction value