Decomposition into Orthogonal Subspaces
What You'll Learn
You'll learn how to prove that every vector can be uniquely split into two perpendicular components—one lying in a subspace and one in its orthogonal complement.
The Core Idea
Imagine you're standing in a room and you shine a flashlight at a wall. Your light beam (a vector) can be thought of as having two parts: the bright spot on the wall (the component in the wall's plane) and the part traveling perpendicular to the wall (the component in the orthogonal direction). These two pieces add up to give you the full beam, and this split is unique.
Mathematically, given any subspace W of an inner product space V and any vector v in V, we can write:
v = w + w⊥
where w ∈ W and w⊥ ∈ W⊥ (the orthogonal complement of W).
The Proof Sketch
Existence: If W has an orthonormal basis {u₁, u₂, ..., uₖ}, we can construct:
- w = ⟨v, u₁⟩u₁ + ⟨v, u₂⟩u₂ + ... + ⟨v, uₖ⟩uₖ (the projection onto W)
- w⊥ = v - w (the remainder)
By construction, w⊥ is orthogonal to every basis vector of W, so w⊥ ∈ W⊥.
Uniqueness: Suppose v = w₁ + w₁⊥ = w₂ + w₂⊥. Then (w₁ - w₂) = (w₂⊥ - w₁⊥). The left side is in W, the right side is in W⊥. A vector in both W and W⊥ must be orthogonal to itself, which means it's the zero vector. Therefore w₁ = w₂ and w₁⊥ = w₂⊥.
Key Takeaway: Every vector in an inner product space has a unique decomposition into a component within any subspace and a component in that subspace's orthogonal complement—this is the orthogonal decomposition theorem.