A Reinforcement Learning-Based Dynamic Clustering Algorithm for Compressive Data Gathering in Wireless Sensor Networks

Algorithm 1

1: initialize the algorithm’s parameters, iteration round , maximum iteration number , action set , reward table of each action , and ASV table

2: whiledo

3: wait for START message

4: for all nodes

5:   if

6:    randomly select an action from A

7:   else

8:    select the action a with the max value in

9:   end if

10:   send the node’s sensing data to the selected CH

11:   calculate the corresponding reward using (11) and

    update the value of using (8)

12: end for

13: for all CH_i

14:  CH_i receives intra-cluster data packets,

15:  calculates the sparsity K_i of intracluster data,

16:  generates a random Gaussian

   measurement matrix ,

17:  compresses the intra-cluster data using (1) and gets

   compressed measurement vector ,

18:  sends to the sink

19: end for

20: if data transmission is over

21:  the sink broadcasts START message

22:  

23: end if

24: end while