(1−i)x1−ix2=02x1+(1−i)x2=0 Solution: We have {(1−i)x1−ix2=02x1+(1−i)x2=0⇒{2x1−(1+i)ix2=02x1+(1−i)x2=0⇒{2x1+(1−i)x2=02x1+(1−i)x2=0 so let x2=c, we have x1=(i−1)c/2, so all solutions to the system is x1=2(i−1)c,x2=c,∀c∈C
2. If
A=⎣⎡321−11−3210⎦⎤
find all solutions of AX=0 by row-reducing A.
Solution: Since ⎣⎡321−11−3210⎦⎤→⎣⎡00187−3210⎦⎤→⎣⎡00117−3110⎦⎤→⎣⎡0011001−63⎦⎤→⎣⎡001100113⎦⎤→⎣⎡001100010⎦⎤ so AX=0 has only trivial solutions.
3. If
A=⎣⎡64−1−4−20003⎦⎤
find all solutions of AX=2X and AX=3X.
Solution: It’s equivalent to solving (A−2I)X=0 and (A−3I)X=0, use row-reducing we have A−2I=⎣⎡44−1−4−40001⎦⎤→⎣⎡100−101001⎦⎤ so let x2=c, we have x1=c and x3=−c, so all solutions to the system (A−2I)X=0 is (c,c,−c),c∈C. A−3I=⎣⎡34−1−4−50000⎦⎤→⎣⎡001010000⎦⎤ so all solutions to the system (A−3I)X=0 is (0,0,c),c∈C.
4. Find a row-reduced matrix which is row-equivalent to
5. Prove that the following two matrices are not row-equivalent:
⎣⎡2ab0−1c003⎦⎤,⎣⎡1−211032−15⎦⎤ Solution: The first matrix can be row reduced to ⎣⎡2ab0−1c003⎦⎤→⎣⎡1000−1c003⎦⎤→⎣⎡100010001⎦⎤ the second matrix can be row-reduced to ⎣⎡1−211032−15⎦⎤→⎣⎡100122233⎦⎤→⎣⎡10011021/20⎦⎤→⎣⎡10001021/20⎦⎤ thus they are not row-equivalent.
6.Let A=[acbd] be a 2×2 matrix with complex entries. Suppose that A is row-reduced and also that a+b+c+d=0. Prove that there are exactly three such matrices.
Solution: The zero matrix [0000] is row-reduced and satisfy a+b+c+d=0, suppose there’re nonzero entries, then at least two entries shall be nonzero, otherwise contradicting a+b+c+d=0, consider the case when there’re exactly 2 nonzero entries, then they can’t belong to different rows, since then the matrix would be [1001] or [0110], under the restriction of row-reduced, but then contradicting a+b+c+d=0, in the case when they belong to the same row, we get the matrix [10−10] or [010−1] satisfy the condition. Now consider the case when there’re exactly 3 nonzero entries, then the matrix is of the form [10bd] or [01bd], but notice then either b=0,d=1 or b=1,d=0, under the restriction of row-reduced, this contradicting a+b+c+d=0. Finally the case when all 4 entries is nonzero contradicts the restriction of row-reduced. Thus there’re exactly three such matrices: [0000], [10−10] and [010−1].
7.Prove that the interchange of two rows of a matrix can be accomplished by a finite sequence of elementary row operations of the other two types.
Solution: It’s enough to show the process for a two-row matrix: [a1b1⋯⋯anbn]add row1 to row2[a1b1+a1⋯⋯anbn+an]add (-1)*row2 to row1[−b1b1+a1⋯⋯−bnbn+an]add row1 to row2[−b1a1⋯⋯−bnan]multiply row1 by -1[b1a1⋯⋯bnan]
8. Consider the system of equations AX=0 where A=[acbd] is a 2×2 matrix over the field F. Prove the following:
( a ) If every entry of A is 0, then every pair (x1,x2) is a solution of AX=0.
( b ) If ad−bc=0, the system AX=0 has only the trivial solution x1=x2=0.
( c ) If ad−bc=0, and some entry of A is different from 0, then there is a solution (x10,x20) such that (x1,x2) is a solution if and only if there is a scalar y such that x1=yx10,x2=yx20.
Solution: ( a ) We have 0X=0 for every X. ( b ) Since a and c can’t both be 0, we suppose a=0, use row reduction, we have A=[acbd]→[acabda]→[a0bad−bc]→[1001] and the conclusion follows. ( c ) If both a and c is 0, then A is row-equivalent to [0010] or [0001]. If not, then A is row-equivalent to [a0bad−bc] or [0cbc−add], in all cases A is row-equivalent to a matrix that has nonzero entries in only one row, thus the system AX=0 has the same solution with αx1+βx2=0,(α=0)∨(β=0) let (x10,x20) be a nonzero solution, then αx10=−βx20, thus either α=0, which leads to x10=kx20,k=−β/α, or β=0, which leads to x20=lx10,l=−α/β. If there’s any (x1,x2) which satisfies αx1+βx2=0, then we must also have x1=kx2 or x2=lx1, and the conclusion follows. In fact, (x10,x20) is a basis for the solution space in this case, so any (x1,x2) being a solution, we must have (x1,x2)=y(x10,x20).