Proof of the Structure Theorem via Smith Normal Form
We prove that every matrix over a PID can be reduced to Smith normal form by row and column operations, from which the structure theorem follows.
Smith Normal Form
Let be a PID and an matrix over . There exist invertible matrices and such that is in Smith normal form:
where , each , and .
Proof
The proof uses three types of elementary operations (each achievable by multiplication by an invertible matrix):
- Swap two rows (or columns).
- Multiply a row (or column) by a unit.
- Add a multiple of one row (or column) to another.
Step 1: If , we are done. Otherwise, by swapping rows/columns, we may assume .
Step 2: Make divide all entries in the first row and column. If , perform the Euclidean algorithm: write with and (where is the Euclidean function; in a general PID, use that ideals are principal to argue). Replace column by column column , getting in position . Swap columns to put in position . The "size" of strictly decreases, so this terminates. Similarly for column entries.
Step 3: After Step 2, divides all entries in row 1 and column 1. Subtract multiples to zero them out: column column column 1 for , and similarly for rows.
Step 4: If does not divide some with : add row to row 1, creating a nonzero entry in row 1 not divisible by . Return to Step 2. Since the ideal strictly increases each time this happens, and is Noetherian, this terminates.
Step 5: Now divides all entries of . The matrix has the form . Apply the procedure recursively to .
Step 6: The resulting diagonal entries satisfy by construction (Step 4 ensures each diagonal entry divides all later ones).
Deriving the Structure Theorem
Given a finitely generated -module with generators, for some matrix (the relation matrix). Reducing to Smith normal form gives:
The divisibility chain gives the invariant factors.
over . After row/column operations: Smith form . So .