Tree traversal is a process of visiting each node in a tree exactly once. Unlike linear data structures such as an array or linked list which is canonically traversed in linear order, a tree may be traversed in depth-first or breadth-first order

Breadth First Traversal

Breadth-first traversal visits all nodes on the same level before going to another lower level

For the above tree, assuming that the left subtree are chosen before the right subtree, breadth-first traversal orders will be 7-2-9-1-3

Implementation example on a binary tree

We can follow these steps to implement a bread first traversal algorithm on a binary tree

• Create a queue Q to track the traversal path

• Add the root node into Q

• Do the following while Q still contains at least one node

  • Remove a node N from Q

  • Print the value of N

  • Add left child of N into Q if it is not null

  • Add right child of N into Q if it is not null

import java.util.ArrayDeque;  
import java.util.Queue;

public class TreeBreadthFirstTraversal {  
    static void breadthFirstTraversal(BSTByLinkedList.Node root) {
        Queue<BSTByLinkedList.Node> queue = new ArrayDeque<>();
        queue.offer(root);

        while (!queue.isEmpty()) {
            BSTByLinkedList.Node currentNode = queue.poll();
            System.out.println(currentNode);

            if (currentNode.left != null)
                queue.offer(currentNode.left);

            if (currentNode.right != null)
                queue.offer(currentNode.right);
        }
    }

    public static void main(String[] args) {
        BSTByLinkedList tree = new BSTByLinkedList();
        tree.insert(7);
        tree.insert(2);
        tree.insert(3);
        tree.insert(1);
        tree.insert(9);

        breadthFirstTraversal(tree.root);
    }
}

Output

data: 7, left: 2, right: 9  
data: 2, left: 1, right: 3  
data: 9, left: null, right: null  
data: 1, left: null, right: null  
data: 3, left: null, right: null

BSTByLinkedList is defined in Binary Search Tree Data Structure