Knitting Sort: A Novel Stable and Inplace Sorting Technique (An Extension of Exchange Sorting with Bitonic Sorting Network)

One of the most frequent operations performed on database is searching. To perform this operation we have different kinds of searching algorithms, some of which are Binary Search, Linear Search, Quadratic search[1], but these and all other searching algorithms work only on data, which are previously sorted. An efficient algorithm is required in order to make the searching algorithm fast and efficient. Sorting is an algorithm that arranges all elements of an array, orderly. Sorting Technique is frequently used in a large variety of important applications to arrange the data in ascending or descending order. Several Sorting Algorithms of different time and space complexity are exist and used[9], but they are not suitable because some are stable but not inplace such as merge sort, and other are inplace but not stable such as Quick sort. Two approaches of 2-way merging (odd-even merging and bitonic merging ) have been proposed by Batcher[16]. kway bitonic sort based on 2-way merging(i.e. two monotonic sequences of n numbers merging is completed by k m-way bitonc sorters[12] and m k-key bitonic sorters, where 2n=km and “k-key means k numbers”) has been proposed by Nakatani et al.[17] and Batcher[18].Since the early work of Batcher, much work has been done on parallel processing system,multi-access memory systems, and so on. This research paper presents a new sorting algorithm named as Knitting Sort which is stable as well as Inplace also. In this algorithm the technique of exchanging elements is used. It is shown in this paper that by various manners of comparisons and swapping the array is sorted. The technique of bitonic sorting network is used to implement this algorithm. The Knitting Sort is implemented on an array to sort it in ascending or descending order. The average case time complexity for this algorithm is ɵ(n).