A diode matrix is a two-dimensional grid of wires: each "intersection" wherein one-row crosses over another has either a diode connecting them, or the wires are isolated from each other.
It is one of the popular techniques for implementing a read-only memory. A diode matrix is used as the control store or microprogram in many early computers. A logically equivalent transistor matrix is still used as the control store or microprogram or 'decode ROM' in many modern microprocessors.
A single row of the diode matrix (or transistor matrix) is activated at any one instant. Charge flows through each diode connected to that row. That activates the column corresponding to each row. The only activated control signals during that instant were those whose corresponding column wire was connected with a diode to that row.