Codiode

Fixed Point Rounding Logic

HardLogic CircuitBuild

Digital signal processing pipelines require word length reduction after multiplication operations. Simple truncation introduces a systematic negative bias into the signal, requiring round-to-nearest logic to maintain mathematical accuracy.

The circuit takes an 8-bit input din and produces a 4-bit output dout. The upper four bits din[7:4] represent the integer portion to be kept. The lower four bits din[3:0] represent the fractional portion to be discarded. Implement round-to-half-up logic: conditionally add one to the least significant kept bit based on the value of the most significant discarded bit.

din dout
01010000 0101
01010111 0101
01011000 0110
11111000 0000

Constraints

  • Combinational logic only.
  • Ignore overflow carry bits beyond the 4-bit output width.

Topics

fixed-pointdsprounding

Solve this problem

Place the gates, wire them up and watch the signals settle. Every submission runs on the same simulation engine that grades it.

This problem is part of Codiode Pro. The statement above is free to read.

The circuit builder and code editor need a desktop screen. On a phone, read the problem here and open it on a laptop to solve.

Part of Arithmetic and ALU Design

Problem 25 of 32 in the Arithmetic and ALU Design track.

Related problems

Browse all problems · Digital logic design practice problems · Learning tracks

Solve this on a laptopSend yourself the link and continue there.