A.2.3.5 Pass Transistor Logic

The general idea of pass-transistor logic as shown in Fig. A.13(i) is to multiplex (i.e., to select from) existing signals x_i by using product-term switching functions of a set of selection signals $y_i, \, \overline{y_i}$ and summing up. The product terms are simply series connections of NMOS transistors to pass zeros, and PMOS transistors are used to pass ones. The logic function can be expressed as

$$\sum x_i P_i({\bf y}) ,$$ (A.22)

where P_i are the product terms. Fig. A.15 shows a four-to-one multiplexer n pass-transistor logic.

The advantages of pass-transistor logic are the simple design, the reuse of already available signals, and the low contribution to static power. However, as against other types of logic, the x signals are not regenerated, instead, they are passed through an RC structure. This requires combination with other logic styles and makes it somewhat difficult to verify the timing of pass-transistor logic and to optimize for performance.

Figure A.15: Four-to-one multiplexer in pass-transistor logic