A Polynomial Diophantine Generator Function for Integer Residuals
Two Diophantine equation generator function for integer residuals produced by integer division over closed intervals are presented. One each for the closed intervals [1,Floor(n^0.5)] and [Ceiling(n^0.5),n], respectively.
Authors: ** (논문에 저자 정보가 명시되지 않아 제공할 수 없습니다.) **
Page 1 of 7 A Pol yn o mi al Diophantine Gener ator Function for Int eger Residuals Charles Sauerbier January 2010 Abstract: Two Diophantine equation generator function for integer residuals produced by integer division over closed intervals are presented. One each for the closed intervals 1, √ and √ , , respectively. 1. Preliminaries In this paper we address the problem of determining residual values in integer division by mathematical computation more am enable to analytic methods than iterative division, as well as the values of the corresponding integer quotient. Diophantine generator functions derived from empirical observation of the pattern present in integer val ues as a result of sequential execution of the division operation over each of the respective closed integer intervals is presented that admits the direct computation of corresponding residual and quotient by means of polynomial functions. Unlike the process of si mple iterative division the resulting generator functions admit application of mathematical tools to analysis of applicable problems not otherwise admitted by the sequential division process. We have not found the results presented addressed by others in available literature. The focus of the literature of general relation to the problem and problems it impacts having taken avenues premised on more conventional methods, such as algebraic structures and numeric sieves. The approach taken in derivation of the generator functions is by way of difference expression [1] on the 3 ‐ tuple < x, y, r > in context of the commonly known relation . 2. Generator Functions The observed behavior of the integer values in the two intervals leads us to produce distinct generator functions for each interval. This dic hotomy works to benefit the objective of locating the zeros of the function , as the interval √ , contains precisely one value for ‘x’ such that ‘x’ divides ‘n’. 2.1. Conditions We consider the following conditions: Given a value ‘n’, 3 ‐ tuples where = √ , √ , √ √ . The values of x 0 , y 0 ha ve the relation x 0 ≤ y 0 . For our purpose here we define on the values i = x 0 and j = y 0 the following delta: 1 0 . [1] (Kelley & Peterson, 1990) (Elaydi, 2005) (Goldberg, 1986 ) Page 2 of 7 2.2. X Decrement , The generator functions here are premised on decrementing ‘x’ over th e interval √ 1 , where ‘x’ is an integer index value. The generator function is derived from n = (x * y) – r. We obtain the set of difference expressions of Lemma 2.2.1 for the respective values of x, y, r from that relation. Lemma 2.2.1 Given the conditions as stated, above in 2.1, for all values k in the closed interval 1, over integers: – 1 ,and; . The difference expressions follow by basic algebra from the relations , as any amount by which x is reduced needs be account for in the other term s of the equation to maintain the equality relation between the left and right sides. The expressions for computing the resultant values of ‘y’ and ‘r’ distribute the subtracted value of ‘x’ in the context of the equation across those variables in accordance with the algebraic manipulation: 1 1 1 1 Theorem 2.2.1 Given the conditions as stated , abov e in 2.1, under the difference expressions of Lemma 2.2.1, where –1 , the residual value where ; whichreducesto . We observe that under the given conditions the value √ will be either an integer or real number with a non ‐ zero decimal component. The result of the assignment of values to , is such that either or 1 for any given value of ‘n’. With representing | | we have then the follow sequence: Page 3 of 7 The sequence of relations above only hold where . However, in foregoing the application of the mod and div operations we arrive at the following: , 1 , 1 Which then leads to the following: 1 1 1 1 1 2 1 12 Observing that is equal to 0 or 1 and present in each iteration we have for any the product . This leaves the summation of for each giving us ∑ ∑ . From which we can thus derive: ; which by algebraic operations on the summations reduces to . Testing this hypothesis for base values 0 and 1 we obtain the following results: 0 0 1 1 1 . By induction we then assume that the expression holds for all k, for k+1 we then have 1 1 2 1 2 1 2 1 Observing that absent application of the quotient of that the difference between the corresponding values of , increase with k by 2, including the constant 1 resulting from the decrementing action of 1 the above then reduces to , proving the theorem. Page 4 of 7 Theorem 2.2.2 Given the conditions as stated , abov e in 2.1, under the difference expressions of Lemma 2.2.1, as applied in Theorem 2.2.1: . . Let us assume and ; then implies divides ‘n’, in contradiction of the assumption. Similarly, assume and ; then implies divides ‘n’, in contradiction of the assumption. Corollary 2.2.1 Given the conditions as stated , abov e in 2.1, under the difference expressions of Lemma 2.2.1, as applied in Theorem 2.2.1: . This corollary follows from Theorem 3 by substitution of ; where the equality follows from the definition of as the k th subtraction of the value 1. Lemma 2.2.2 Given the conditions as stated , abov e in 2.1, under the difference expressions of Lemma 2.2.1, as applied in Theorem 2.2.1: Lemma 2.2.3 Given the conditions as stated , abov e in 2.1, under the difference expressions of Lemma 2.2.1, as applied in Theorem 2.2.1 and using the result of Lemm2.2.2: Noting that the value computed as is absent the addition of integer quotient of (i.e. “r div x”) and that accounts for that integer quotient in the absence of the mod operation being applied, from Lemma 2.2.1 it follows that , producing a generator function for the quotient of integer division of ‘n’ by ‘x’. 2.3. Y ‐ Increment , The generator functions here are premised on increm enting ‘y’ over the interval √ , where ‘y’ is an integer index value. The gen erator function is derived from n = (x * y) – r. We obtain the set of difference expressions of Lemma 2.1.1 for the respective values of x, y, r from that relation. Page 5 of 7 The generator functions presented in this section operate over the interval √ . Computation of ‘x’ without modular normalization of values will cause ‘x’ to become negative, which in computer applications may become problematic. Lemma 2.3.1 Given the conditions as stated, above in 2.1, for all values k in the closed interval , over integers: 1 ,and; . The difference expressions follow by basic algebra from the relations , as any amount by which y is increases needs be account for in the other terms of the equation to maintain the equality relation between the left and right sides. The expressions for computing the resultant values of ‘x’ and ‘r’ absorb the added value of ‘y’ in the context of the equation across those variables in accordance with the algebraic manipulation: 1 1 1 1 Theorem 2.3.1 Given the conditions as stated , abov e in 2.1, under the difference expressions of Lemma 2.3.1, where 1 , for all values k in the closed interval , over integers the residual value where ; whichreducesto . We observe that under the given conditions the value √ will be either an integer or real number with a non ‐ zero decimal component. The result of the assignment of values to , is such that either or 1 for any given value of ‘n’. With representing | | we have then the follow sequence: Page 6 of 7 The sequence of relations above only hold where . However, in foregoing the application of the mod and div operations we arrive at the following: , 1 , 1 Which then leads to the following: 1 1 1 1 1 2 1 12 Observing that is equal to 0 or 1 and present in each iteration we have for any the product . This leaves the summation of for each giving us ∑ ∑ . From which we can thus derive: ; which by algebraic operations on the summations reduces to . The proof then follows in consequence of Theorem 2.2.1. Theorem 2.3.2 Given the conditions as stated , abov e in 2.1, under the difference expressions of Lemma 2.3.1, as applied in Theorem 2.3.1, for all values k in the closed interval , over integers: . . Let us assume and ; then implies divides ‘n’, in contradiction of the assumption. Similarly, assume and ; then implies divides ‘n’, in contradiction of the assumption. Corollary 2.3.1 Given the for egoing conditions under the difference expressions of Lemma 2.3.1, as applied in Theore m 2.3.1, for all values k in the closed int erval , over integers: . Page 7 of 7 This corollary follows from Theorem 3 by substitution of ; where the equality follows from the definition of as the k th addition of the value 1. Lemma 2.3.2 Given the conditions as stated , abov e in 2.1, under the difference expressions of Lemma 2.3.1, as applied in Theorem 2.3.1, for all values k in the closed interval , over integers: Follows directly from definition of difference expressions, with the except ion that where , the relation does not hold. Lemma 2.3.3 Given the conditions as stated , abov e in 2.1, under the difference expressions of Lemma 2.3.1, as applied in Theorem 2.3.1 and using th e result of Lemm2.3.2, for all values k in the closed interval , over integers: Noting that the value computed as is absent the addition of integer quotient of (i.e. “r div y”) and that accounts for that integer quotient in the absence of the mod operation being applied, from Lemma 2.3.1 it follows that , producing a generator function for the quotient of the integer division of ‘n’ by ‘y’. 3. Conclusion As a result of observation of empirical data sets a Diophantine generator function of integer quotient and residual in integer di vision was derived. The generator functions admits the computation of both quotient and residual values using continuous polynomial functions on basis of an index in either of the closed integer intervals 1, √ or √ , . The generator functions provide means to de termin e the zeros within either closed interval where ‘n’ is the dividend. The results presented have potential application in determining the factors of an integer value, as well as implications for other applications. 4. References Elaydi, S. (2005). An Introduction to Difference Equations. Springer. Goldberg, S. (1986). Introduction to Difference Equations. Dover Publications. Kelley, W. G., & Peterson, A. C. (1990) . Difference Equations : An introduction with Applications. Academic Press.
Original Paper
Loading high-quality paper...
Comments & Academic Discussion
Loading comments...
Leave a Comment