- npORY OF NUMBER Diophantine Equations. The general form of a lincar Diophantive equation in two ariables having integral coefficients is ¥ ax+by =e - QQ) The solution of this equation is trivial unless neither a nor b is zero, so We can SUPpOSe a #0, b +0. Let d = ged(a, b). Then theo- rem 5 shows that there exist integers X9 and y, such that axy+byy=d w (2) Numarical values for Xj) and Yo can be obtained conveniently by applying the Euclidean algorithm, art. 2.7 to the integers | a | and | b|. Now if d|c, we get (x9, yo) as the particular integeral solution of (1). Then all integral solutions of (1) are given by xaxy+ek, y=y¥o-2h d d where k is any integer. But, ifd 4c, then ax+by=c clearly has no solutions in inte- gers. Illustration (i) Consider the Diophantine equation 2x+3y=4. Here ged(2, 3)=1 and 4 is divisible by 1. So the equation 2x+3y = 4 has integral solution. Now 1=2-(-1)+3-1 4=2.(-4)+3-4 Thus one integral solution is xy =-4, ¥9 = 4. Hence all inte- 8ral solutions of the given equation are given by Renda 3h, y=4-2h for any integer k ‘©, y= 443k, y=4-2k for any integer k. iDISCRETE (ai) Consider another Diophantine equation Lax +2ly =5 Here ged(14, 21) = 7 and 5 is not divisible by 7. Henee aen3 equation has no integral solution. Illustrative Examples. y . Ex. 1. If d= ged(x,y), then : and a ave integers prime to each other. Solution. Since d = ged(x,¥)> there exist integers land m such that d=lx+my ==142 +m Since d is the division of x and y, there exist integers U and v that Z=u, Z=0 such that 7 od wlsul+um -, wand v are integers prime to each other. Thus 7 and 7 are integers prime to each other. Ex. 2. If ged(a, 4) =2 and gcd(b, 4) =2, proved that gcd(a+b,4)=4 Solution. Since ged(a, 4) = 2, 80a is a multiple of 2 but not 4 -.a=2p, for some odd integer p Similarly, since ged(b, 4) = 2, we have b = 2q, for some odd integer q| npoRY OF NUMBER 21) .a+b=2%Ap+q) =2-2r where p+q=2r, an even integer, as, p, q are odd =4r _ged(a +b, 4) = gcd(4r, 4) where r is an integer =4 Thus gced(a+b,4)=4. Ex. 3. If ged(a,b)=1 prove that ged(a+b,a-b)=1 or 2 Solution Let gcd(a+b,a—b)=d atb=kd . (@) and a-b=h,d . @ where k,, k, are integers From (1) and (2), we get 2a =(k, +k,)d and 2b=(k, -k,)d -. d divides 2a and 2b ».d < gcd(2a, 2b) = 2.gcd(a, b), by property (C) of at 2.8 =2-1 [- ged(a,b)=1 =2. Hence d =1 or 2 -. ged(a+b,a—6)=lor2 Ex. 4. Find the integers x and y such that 154x + 260y =3 Solution. Here ged(154, 260) = 2. Since 3 is not divisible by 2, it follows that the given equation has no integral solution. Ex. 6. Find the ged of 252 and 596 and express it in the form 252x + 595y Solution. By division algorithm, 595 = 2x 252+91 qa) 252 =291+70 (it)DISCRETE MA?}, 91=1x70+21 70 =3x214+7 21=3x7+0 Since the last non-zero remainder is 7, so ged( 252, 595) = 7 To express the ged in the form 252x + 595y , we have, from (a, 7=70-3x21 = 70-3(91-1x 70), by (iu) =4-70-3-91 = 4(252-2x91)-3-91, by =4-252-11-91 = 4-262-11(595 - 2 x 252), by (i) = 26-252-11-595 <7 = 252x+595y where x=26, y=-11 (u) Ex. 6. Find all solutions in positive integers of 5x +3y =52 Solution. Since gcd(5,3)=1 and 52 is divisible by 1, so the equation has integeral solutions. Now 5=1-34+2 3=1-241 11=8-1-2=3-1(5-1-3) =-1-54+2-3 Hence §2 = -5.52+3-104 Thus one integral solution of the equation is Xp = -62, yy = 104, Hence all integral solutions of the equation are given by ad x= 62+ 3k» ¥=104~6k for all integers k ¢THEORY OF NUMBER (223) As the solution will be positive, we must have ~52+3k>0 and 104-5k>0 _ 52 104 np sks< = - k= 18,19, 20. Thus the given equation has exactly three positive solutions and these are x=-52+3-18=2, y=104-5-18=14 x=-52+3-19=5, y=104-5-19=9 x=-52+3-20=8, y=104-5:20=4. Ex. 7. Using prime factorisation, find gcd and lem of 1300, 3575, Also verify that gcd(a, b)-lcm(a,b) = ab. Solution. We have 1300 = 2?-5*-11°-13" 3575 = 2°-5°-11'-13! -. ged(1300, 3575) = gmin(2.0) a gmim(2 2) y gmim(Ou¥) , pgmin(t yy = 29.57-11°-13! = 325 lem(1300 3575) = 2-9) inmn22)_ppman(O.d) gmat = 2°, 57.11! 13! = 14300 * Bed(1300, 3575) - lem(1300, 3575) = 825 x 14300 = 4647500 = 1300 x 3575 Hence ged(a,b)-lem(a, 6) = ab -ww Ss Ex. 8. Ifa is a prime integer such that a=n?-4 for some integer n, then show that a=5 Solution. We have a =n" — 4 =(n—2)(n + 2) for some integer n. Since a is prime, we must have, either n— 2=1or,n+2=1 But n +21 for some integer 7. Hence n-2=1 ie, n=3- naen?-4=3°-4=5. Ex. 9. Solve the Diophantine equations 158x-47y =9- Solution. Here gcd(158, - 47)=1and 9 is divisible by 1. Hence the given equation has integral solution. Now d 158 = (-3)(-47) +17 -47 =(-3)11+4 17=4-441 1=17-4-4 =17-4{-47- (-3)-17} =(-4)(-47)+(-11)-17 = (—4)(-47) + (-11) -{158 - (-3)(-47)} = (-11)-158+(-37)(-47) 2.9 = (-99)-158 + (-333)(—47) Thus one integral solution of tHe given equation is X= -99, yy = -333. Hence all integral solutions of the equation are given by x =-99-47k, y =-333-158k, for all integers &. >» ¢THEORY OF NUMBER 225) Congruence. Definition. Let a and b be any two integers and m is a posi- tive integer. Then a is said to be congruent to b modulo m, if a—b is divisible by m or, m | (a—), and in this care we write a=b mod (m) Here m is called the modulus of the congruence, and bis called a residue of a mod (m). If a-b is not divisible by m, we say that a is not congruent to b module m and in this case we write a 4b mod (m). Note. If a= (mod m) then a~b=km ie. a=b+km for some integer k. Illustration. (i) Since 57-17 is divisible by 5, so 57 =17 (mod 5) Here 5 is the modulus of the congruent and 17 is the residue of 57(mod 5) (ui) Since 32-3 = 29 is not divisible by 7, so 32 and 3 are in congruent modulo 7. 1.82 #3 (mod 7). Properties of congruence Let a, b, c denote integers. Then : 1. If a =b (mod m), then b = a(mod m) Proof: Since q =} (mod m), a-b is divisible by m 1e., -(b-a) is divisible by m Le., b~a is divisible by m -b=a mod (m) 2 If a=b (mod m), then a(modm) = b(modm) Proof As gap (mod m), 80 a—b = km, for some integer mLet k=k,-k, where k, and k, are integers, a—b=(k,—k,)m a-km=b-kym=r, say a-kme=r,b-kym=r “=a (mod m), r= b (mod m) *.a (mod m) = 6 (mod m) 3. If a= b (mod m), then atc=(btc)m Proof: Since a = b (mod m), a-b is divisible by m Now a-b=(a+c)-(btc) (atc)—(b+¢) is divisible by m =atc=(bic)m. 4. If a= b (mod m), then ac = be (mod m) Proof: Since a = b (mod m), a-b is divisible by m <.(a-b)c = ac— be is also divisible by m ’. ae = be (mod m) Note. The converse of the above property is not always true m 5. If ae = be(mod m) then 4 = b| mod 77 Proof: Since ac = be (mod m), ac- be is divisible by m :.ac—be = kmwhere k is an integer -a-b=h(™) .a=b (mod = ), providesd = is an integer .THEORY OF. NUMBER 27) m. ‘ Since —— isan integer, so c divides m «. ged (c,m)=c b{ mod —™ _ Hence a= (rm gcd (c, =a) Note. If ac=bc (mod m), then a=b (mod mj) only if ged (c,m) = 1 - 6. If a= (mod m) and c=d (mod m), then atc=(b+d)(mod m) Proof : Since a = b(mod m), a—b is divisible by m Similerly c—d is divisible by m. :.(a—b)+(c-d) is also divisible by m. ie., (atc)—(b+d) is divisible by m. -.atc=(b+d) (mod m) Note In general a = (mod m) and c=d (mod m) = ax +cy =(bx +dy) (mod m) where x,y are integers. 7. If a=b (mod m) and ¢=d (mod m), then ac = bd (mod m) Proof : Since q = (mod m), a-b is divisible by ™ “(a~b) ¢ is divisible by m Also, since ¢= q (mod m), c—d is divisible by ™ “(c~d) b is divisible by m Hence (a — b) c+(c—d) b is divisible by m- \e., ac—bd is divisible by ™. “ae = bd (mod m)MATHEMATIC THEORY OF NUMBER pIisCRETE | 1 (mod m) where 1 18 a Positivg Thus a= (mod m) and 8. Ifa = b (mod m), then a" sb’ integer Proof: Obviously, statement 18 (THe for n=1 =.= (mod m) Let the statement is true for 7 =k Hence the congruence: ‘Thus th ot = bt (mod m) wus the congruence r ‘Phen, using property (7, we have Residue class |) = b.b* (mod m) Definition The set of: modulo n is called the con ke = pie! (mod or residue classes of int 1 whenever it is t is truefor or [a] or, 3. Thus Th he statement 1S true for n= k+ e for any positive inte [a]={b_2:b= a(n! n=k Hence by induction the statement is tru ger® : js i For exampleall congrt Theorem. The congruence relation is an equivalence relation, . oO} = 7 Proof. As a-a= 0 is divisible by ™, 8° { | {bez a aza (mod m) for any integer a. ={beZ:b Thus the congruence relation is reflexive efor Again a=0 nod m) fijefbezibel' = pezb-l-# >a-bis divisible by ™ -a) is divisible by ™ 27,-3b5. =-(b 4 3S divisible by ™ = b-a =>bea (mod m) ce the congruence relation 1s symmetric. mod m) 247 n divisible by ™ Hen Lastly a=b ( b=c(mod m)= b-¢ (mod m) i So (a-b)+(b-e) in divisible by ™ Le., a~c is divisible by m a=c (mod m)Thus a= b (mod m) and } = ¢ (mod m) > ame (mod m) Hence the congruence relation is transitive . Thus the congruence relation is an equivalence relation. Residue classes of integer modulo n Definition The set of all integers b which are congruent to a modulo n is called the congruence class of integer modulo n _ orresidue classes of integer modulo n and is denoted by [2], | or {e] or, @. Thus {a]= {be Z:b=a(mod n)}. For example,all congruence classes of integer modulo 4 are [0]= {be Z:b=0(mod 4)}. ={b €Z:b is divisible by 4 or b= 4k for some k} = {---8,-4,0,4,8,--+} [l]={beZ:b=1 (mod 4)} ={beZ:b-1=4k ie b=1+ 4k for some k} = {+6-7,-3,1,5,9,°} Similarly [2]={---6,-2,2,6,10,---} [8]={-5,-1,3,7,11,---} [a]= (8, -4,0,4,8.- }= [0] ‘Thus, there are only 4 distinct congruence classes of integer | modulo 4, namely [0], [1], (2}.[3].Ex] piscrerE MATHEMAN In general [0},[1], (2), [n- 1] are n distinct residue Clases , integer modulo n It is evident that {a]=[a+n]=[a+ 2n]=- [ol=[n]=[2n]= For any positive integer n, Z, denote the set of all congruen, classes of integer modulo n. Thus 2, ={ (0D) 2} [3] In general, 2, ={(00}2)--fe-} Theorem, The number of elements of Z, is finits and th number is n. Proof. left as an exercise Arithmetic of Residue classes. Addition and multiplication for residue classes of integer modu nare defined as given below : [a]+[b] =[a +6] and [a}-[b] = [ad] As an illustration, consider the residue classes of integer mod 3, namely [0},(1], [2]. Thea (i+ (0]=[1), [)+[2]=[3}=[9) f=]. (2}(2]=(4)=0)RY OF NUMBER p31) Following composite tables show the addition and multiplica- tion tables for the residue classes of integer modulo 3: +} PB (ol) (] [>] [>] Q) a @ 2 2} fo] [2] fl Definition. An element [b] ¢Z, is called an inverse of an element [a] €Z, if [a][6]=[1] in Z, . Definition. An element [a] €Z,_ is said to be a unit element in Z,, if [a] has inverse in Z,,. Theorem Let a and n be integers with n > 2 relatively prime. Then (a] has an inverse in Z, if and only ifa and n are relatively Prime Proof. Beyond the scope of this book. ~<—Bs] 2 DISCRETE MATHEMATICS Illustration. (i) The unit elements of Z, are [1].[5]. a8 ged (1, 6) = ged (5,6) =1 Gi) The inverse of [5] in Z, is [3] as [5][3] = [1] Linear Congruence. Let a,b be any two integers. Then a congruence of the form ax = b (mod m) w- (1) Where m is a positive integer and n is an unknown integer is called a linear congruence. An integer x, is called a solution of (1) if a Xo =b (mod m). Any value of x which is a solution of the congruence ax = 1 (mod m) is called an inverse of a modulo m. Illustration (i) Consider the congruence 6x = 3 (mod 9) s+ (2) Since 6-2 =3 (mod 9), so x =9 is asolution of (2). (ii) Since x = 3 is a solution of the congruence 3x =1 (mod 8), so 3 is an inverse of 3 modulo 8. Theorum. If a and m are relatively prime, then the congru- ence ax = b (mod m) has a unique solution. Proof : As a and m are relatively prime, gcd (a,m)=1. -. There exist integers p,q such that pa+qm=1. Then pab+qmb=b Hence pab+qmb-b=0 -. pab + qmb = b (mod m) But qmb=0 (mod m) -. pab = b (mod m)THEORY OF NU ‘MBER = Hence 1 = pb is a solution of ar =4 (mod m) ax =6 mod m) Next let us assume that y be the another solution of ax = (mod m) ay =b (mod m) -- (2) -. From (1) and (2), we have. ax —ya=0 (mod m) ax = ya (mod m) x =y (mod m) [- a and m are relatively prime .] Thus the solution of ax = b (mod m) is unique. Note : If ged (a,m)+d_ the congruence ax = b (mod m) has no solution when d does not divide 6; but if d divides 6, there are exactly d solutions. Illustration Consider the congruence 3x = 2 (mod 8). Since ged (3, 8) =1, the congruence has a unique solution. As gcd (3,8) =1 there exists u and v such that 38u+8v=1 Here u, v can be found out using division algorithm as follows 8=3x242 3=2x1+1 21=3-2-1 =3-1-(3-3-2) =3-3+(-)'8 Hence u=3,v=-1 2.3-34+(-1)-8=1UM RY OF N THEO! gince acd (P,(, DISCRETE MATHEMATICg which implies a’ “ od 8) 3 in i Corollary. If; “3-6 =2 (mo a? Thus x =6 is a solution of the congruence 3x = 2 (mod 8) proof : If p dis Hence the solution of the given congruence is Hence a? =a x =6 (mod 8) If p does not di Fermat's theorem. a? If p be a prime and a is an integer such that p does not divide a, then Hence a? =a‘ a? =] (mod p) Illustration 7 Proof : Consider the integers Fermat's theor @, 2a, 3a,---(p-l)a qa) 258-1 is divi none of these integers is divisible Illustrative E by p. Ex. 1. Show th We show that no two distinct integers of (1) are congruent to Solution. By | each other modulo p. Suppose, if possible ° v ra =sa (mod p) As p is not a divison of a, 3 where 15. Hence we must have m=7—— SO y por OF NUMBER rH EXERCISE Multiple choice Questions. 1. Prime divisors of 12! are (a) 2.5.6, 10,11 (b) 2,3,5,7,11 235.791 @) 1,3,5,7,11 2, By using the Euclidean algorithm find the greatest common aivisor (ged) of (a) 414 and 662 ~—(b) 2689 and 40001 (©) 7469 and 2464 (d) 2947 and 3997 Ans : (a) 2 (b) (11 (©) 77(d) 7 3. Find the ged g of the numbers 1819 and 3587 and then find integers x and y to satisfy 1819x+3587y = g [g=17,x=71y= -36]. 4. Ifa is prime to b and a is prime to c, then a is prime to (a) b? and <2 (b) bec) b+e = @) BW? +c” 5. Integers less than 15 and relatively prime to 15 are (a) 3,7,8,9,11,12,13, (b) 2, 4,7, 8,11, 13,14 (©) 1,3,5,10,15 (d) 2,3,7,9, 11, 13,14 6. Every non-empty subset of natural number has a —— “lement (@) Zero (b) unit —_(c) smallest (d) greatest \ 7. The remainder when the sum 41+ 5!+ 61+ +50! is divided Y 4 is M1 2 @s do 7 —2-40 8. Let ged (a,m)=d. The linear congruence ax = b (mod mn) . d(a,m)=d. has no solutions when (a) d divides a (b) d does not divide 6 (c) d divides (d) d does not divide a b a 9 If ged (a,b)=c, then . and > are (a) prime (b) composite (©) relatively prime (d) none 10. Every integer n is relatively prime to @1 go (©) 2 () none 11. When two integers m and nare relatively prime, 8ed(m,n) = (a)n (b) mn (1 @m 12. Let d= ged (a,b) Then the linear Diophantine equation ax +by=c hasa solu- tion iff (@) d divides a (©) d divides ¢ 13. The number of integral Solutions of the equation 24x4 16y = 2 is (b) d divides 5 (a) none (a) unique (b) nil (a) 3 does not divide 8 (0) 3 divides 51 (©) 3 divides 6 (@) none THEORY OF NUM —_——_— 15. Let a,b,¢ (a) a® = b® (m (©) ac = be (me 16. a=b (mod @) a+b= pm (C) a~b= pm 17. All solutidy @) x=-1 (mo © x=1 (mod 18. The remair @2 @) 2 19. The remain @0 @6 20. The remain @0 1 ( 21. The Congrt “nique module m ) pis prime © pla 22. One solutio: @6 ge2-41 16. Let t a,b,e denote integers, Then qs b (mod m) implies (9) a' = b% (mod m) (b) Va = Yo (mod m) (©) ac = be (mod m) (@) a=b (moa ™) € 16. a=b (mod m) ifand only if for some integer Dp @ a+b= pm ) = pm (©) a~b=pm (@) ab= pm 17. All soluti8ns of 3x =7 (mod 4) are (a) x =~] (mod 4) (b) x =3 (mod 4) (©) x=1 (mod 4) (d) none 18. The remainder when 62 is divided by - 8 is (a)2 (b) -2 0 @6 19. The remainder when 12* is divided by 7 is @0 ©)6 @8 (d)1 20. The remaindes when 310 is divided by 7 is (0 (b) 1 @)7 (d)4 21. The congruence ax = b (mod p) has a solution which is “nique module m if (4) p is prime (b) pfa pla @) pis prime and p}a 13) i 22. One solution of the linear congruence 6x = 1 (mod 13) is %6 de (11 (d) 0 i| 7 pISCRETE MATHEMAT) 23. b is an inverse of a modulo ™ iff — is a solution of ax =| (mod m) (a) b (bo) a (c) 1 A) none 24. For the congruence 5x =1 (mod 3), — 18 an inverse » 5 modulo 3 (a5 @)8 (7 qd) 3 26. inverse of 11 modulo 12 @1 2 @5 (ao is an 26. The number of elements of Zo i (a) 10 =) 10! ©) 10° @9 21. For Ze .{3]+[5)= @] ®2] of] @ [15] 28. For Zs, (2):[7]= . @piy of] off} © ft] 29. The inverse of [1] in Z, is @ [1] ©) [2] © [4] (d) does not exist ANSWERS 1b 2.b "Be 4.b,¢ 5.b 6.¢ 1d 8.b Pc 10.4 lhc 12. ¢ 13. b l4a 15. a,c 16.c 18.a 19d 20.d 22.¢ 23. a 24.b 25. b 26. a 27.b 28.¢ 29. a@) 231,1575 Gi) 439, 95256 Verify also that 8ced (m,n 4, Find ged of 595 and 25: ! 88 it in 25 m+6595n Gti) x= 38 (mod 87) (ii) no Solution ) x= 19 (mod 29 QW) x = _g2 (mod 400) 5. If ged (a, 6)=d then sh ).dem (m,n) = mn 2 and expre iow that b ed (2 a) “ wa(5ug)=1 6. If ged (a, m) = 8cd(b,m) =), then show that &cd (ab, m) =1, 7. Prove that if ged (b,c) =1 and db, then ged (d,c)=1, 8. Show that if ac |bc, then a |e 9. Find all solution in Positive integers : @) 15x4+7y =111 (i) 12%+501y =1 10. Let (a,b)=1 and c>0. Prove that there is an integer x Such that (a+ bx, c)=1 11. Prove that (i) if p is a prime and p|a and p| (a? +6?) then p\b Gi) if a | 3 then alc 12. Prove that if gcd (a,b) = ged (a,c), then 8ed (a, b*) = ged (c’, c’) 13. Prove that ged (a*, b*DISCRETE MATHER, Amy Cy S 14. Prove that if p| ab, p being a prime, then P| 15. Given that ged (a,4)=2 and ged (b, 4) = 2 * Prove 4), &ed (a4 b,4)=2 16. Prove that ged (a, b*) =? if ged (a,b) = ¢ 17. If ged (a,b) =1, prove that ged (2a +b,a4 2b)=1 org Find the inverse of (i) 19 modulo 141 Gi) 144 modulo 233. 18. Prove that n° —1 is divisible by 7 if ged (x,y) = 19. Find the reminder when the sum M2314 5091, divided by 12 20. For any x, prove that acd (a, b) = ged (b,a) = cd (a-b) = ged (a,b+ ax) 21. Find values of x and y to satisfy (a) 256x+116y =2 ) 93x-81y=3 (©) 243x4 198y =9 () 71x-50y =1 22. If x= (mod m), then show that ged (x,m) = ged (y, m) 23. Let f denote a polynomial with integral coefficients, If g=b (mod m), prove that F(a) = f(b) (mod m) 24. Prove that if p is a Prime and q? =? (mod p), then P| (a+b) or P| (a-b) 25. Show that for all n in N, 341 3.5" (mod 11). Obtain a corresponding result about 24*3 and deduce that | (aint + aims) 28. Using Fermat’s theorem, show that (i) 234° 21 (mod 1) (ii) 3° = 9 (mod 11) (ii) 22° =} (mod 41) Construct be addition and multiplication table for Z; 27. Solve (i) 4x = 3 (mod 7) (ti) 6x =7 (mod 9) (ii) 18x = 29 (mod 42)- p0RVOF NUMBER f 8 How many solutions are there to each of the fallowing ces * ee = 95 (mod 35) (i) yx = 24 (mod 35) ii) 15x = 0 (mod 35) go. Find all solutions of the following congruences : ( 20x24 (mod 30) Gili) 59x = 31 (mod 67) Gi) 20r=30 (mod 4) (VY) 25x =15 (mod 29) (iv) 353x = 254 (mod 400) ANSWERS 1. (i) 7,51975 (ii) 216, 190, 512 4, T:m=26,n=-11 9 @) x=6,y=3 (i) no solution 11. (i) 52 Gi) 89 19. 9 U. (a) x=41,y =-9, (b) x=7,y=8 © x=9,y=-11 @) x=31y=44 4%. 23 8.5" (mod 11) 4. i) x= (mod 7) (ti) no solution (ill) = 4 (mod 42)] #5 Gyo Gis 9. (i) no solution ti) no solution (iti) x = 38 (mod 67) Liv) x =~ 82 (mod 400) ) x=19 (mod 29).