Mathwords logoMathwords

Graph Power — Definition, Formula & Examples

Graph power is an operation that creates a new graph from an existing one by connecting every pair of vertices whose distance in the original graph is at most a specified value k.

Given a graph G=(V,E)G = (V, E) and a positive integer kk, the kk-th power of GG, denoted GkG^k, is the graph on the same vertex set VV where two distinct vertices uu and vv are adjacent in GkG^k if and only if the shortest path distance dG(u,v)kd_G(u, v) \leq k.

Key Formula

E(Gk)={{u,v}:u,vV,  uv,  dG(u,v)k}E(G^k) = \{\{u, v\} : u, v \in V,\; u \neq v,\; d_G(u, v) \leq k\}
Where:
  • GkG^k = The k-th power of graph G
  • VV = Vertex set of the original graph G
  • dG(u,v)d_G(u, v) = Shortest path distance between u and v in G
  • kk = Positive integer specifying the maximum distance for adjacency

How It Works

To construct GkG^k, start with the same set of vertices as GG. For each pair of vertices, compute their shortest path distance in GG. If that distance is kk or less, draw an edge between them in GkG^k. The original edges of GG (distance 1) are always preserved, and new edges are added for vertices reachable in 2, 3, ..., up to kk steps. Note that G1=GG^1 = G, and as kk increases, GkG^k becomes denser, eventually becoming a complete graph when kk equals or exceeds the diameter of GG.

Worked Example

Problem: Let G be the path graph on 4 vertices: v1v2v3v4v_1 - v_2 - v_3 - v_4. Find G2G^2.
Step 1: List all pairwise distances in G.
d(v1,v2)=1,  d(v1,v3)=2,  d(v1,v4)=3,  d(v2,v3)=1,  d(v2,v4)=2,  d(v3,v4)=1d(v_1,v_2)=1,\; d(v_1,v_3)=2,\; d(v_1,v_4)=3,\; d(v_2,v_3)=1,\; d(v_2,v_4)=2,\; d(v_3,v_4)=1
Step 2: Include an edge in G2G^2 for every pair with distance at most 2.
E(G2)={{v1,v2},{v1,v3},{v2,v3},{v2,v4},{v3,v4}}E(G^2) = \{\{v_1,v_2\},\{v_1,v_3\},\{v_2,v_3\},\{v_2,v_4\},\{v_3,v_4\}\}
Step 3: The pair {v1,v4}\{v_1, v_4\} has distance 3, which exceeds 2, so it is not an edge in G2G^2.
Answer: G2G^2 has 5 edges. Compared to the original 3 edges, two new edges ({v1,v3}\{v_1,v_3\} and {v2,v4}\{v_2,v_4\}) were added.

Why It Matters

Graph powers appear in frequency assignment problems for wireless networks, where transmitters within kk hops of each other must use different channels. They also play a role in graph coloring theory — a classical result states that the cube of every connected graph is Hamiltonian.

Common Mistakes

Mistake: Confusing graph power GkG^k with the Cartesian or tensor product of GG with itself kk times.
Correction: Graph power keeps the same vertex set and adds edges based on distance. Graph products create a new, larger vertex set from ordered tuples. These are fundamentally different operations.