Skip to content

Latest commit

 

History

History
 
 

Radix_Sort

RADIX SORT

Radix Sort is an efficient non-comparison based sorting algorithm which can sort a dataset in linear O(N) time complexity & hence, can be better than other competitive algorithm like Quick Sort.

It uses another algorithm namely Counting Sort as a subroutine.

Radix sort was developed to sort large integers. It considers integer as a string of digits so we can use Radix Sort to sort strings as well.

Algorithm

  1. Do following for each digit i where i varies from least significant digit to most significant digit.
          Sort input array using Counting Sort (or any stable sort) according to the i'th digit.

Example 1

Assume the input array is: [326, 453, 608, 835, 751, 435, 704, 690]

Based on the algorithm , we will sort the input array according to one's digit(least significant digit).

The original array is sorted based on [6, 3, 8, 5, 1, 5, 4, 0] using Counting Sort.

So, the array becomes [690, 751, 453, 704, 835, 435, 326, 608].

Now, we'll sort according to ten's place :

The above partially sorted array is sorted based on [9, 5, 5, 0, 3, 3, 2, 0] using counting sort .

Now array becomes : [704, 608, 326, 835, 435, 751, 453, 690]

Finally, we sort according to hundred's place(most significant digit) :

the above partially sorted array is sorted based on [7, 6, 3, 8, 4, 7, 4, 6] using Counting Sort

The array becomes : [326, 435, 453, 608, 690, 704, 751, 835] which is now sorted .

Example 2

Radix Sort Process example 2

Example 3

Radix sort Process example 3

Complexity Analysis

Best case time complexity : Ω(nk)

Average case time complexity : Θ(nk)

Worst case time complexity : O(nk)

Space complexity : O(n + k)

where n = number of input data , k = maximum element in input data.

Let there be d digits in input integers. Radix sort takes O(d(n+b)) time where b is the base for representing numbers, for example for decimal system b is 10. What is the value of d ? If K is maximum possible value , then d would be O(logb(k)). So, overall time complexity is O((n + b) * logb(k)) . which looks more than the time complexity of comparison based sorting algorithm for large K.

Let use first limit K. Let k <= nc where c is constant . In that case, complexity becomes O(nlogb(n)). But it still doesn't beat comparison based sorting algorithms. What if we make value of b larger ? . what should be the value of b to make time complexity linear ? If we set b as n, we get the time complexity as O(n). In other words , we can sort an array of integers with range from 1 to nc if the numbers are represented in base n (or every digit takes log2(n) bits ).

Advantages

  1. Fast when the keys are short i.e. when the range of the array elements is less.
  2. Used in suffix array construction algorithms like Manber's algorithm & DC3 algorithm.
  3. Radix sort is stable sort as relative order of elements with equal values is maintained .

Disadvantages

  1. Since Radix Sort depends on digits or letters, Radix Sort is much less flexible than other sorts . Hence, for every different type of data it needs to be rewritten.
  2. The constant for Radix sort is greater compared to other sorting algorithms .
  3. It takes more space compared to Quick Sort which is inplace Sorting .
  4. It can be slower than other sorting algorithms like merge or quick sort , if the operations are not efficient enough. These operations include insert & delete functions of the sub-list & the process of isolating the digits we want.