Research Article
A Novel Bioinspired Multiobjective Optimization Algorithm for Designing Wireless Sensor Networks in the Internet of Things
Step 1. Initialize the main population (ant colony) ; | for each individual in | initialize its position, search range, organization factor and other variables; | initialize the archive ; | Step 2. Evaluate each individual in the population; | Step 3. Classify the individuals in into dominated individuals and non-dominated individuals ND; | for each individual in | flag = 0; | for each individual in | if is dominated by | ; | else | ; | flag = 1; | if flag == 0 | ; | Step 4. Calculate crowding distance for ND; | Initialize the distance to be zero for all individuals in ND ( denotes the size of ND); | for each objective | sort the individuals in ND based on objective ; | assign infinite distance to boundary values for each individual in , i.e. and ; | for to | ; | /* is the value of the th objective function of the th individual in */ | Step 5. Update the archive ; | ; | classify into dominated individuals and non-dominated individuals ; | ; | Step 6. Generate new population by (3) and re-defined concept of neighbor selection; | Step 7. if terminate is true | Output the population; | else | goto step 2; |
|