Meaning
Digital distortion represents the rounding error introduced when an analog signal is converted into a discrete digital value by an analog to digital converter. In signal processing chains and communication receivers, quantization noise limits the achievable dynamic range and signal to noise ratio of the system. This error behaves as a random white noise source under the assumption of a busy, complex input signal.
It is determined primarily by the resolution of the converter, measured in bits.
Resolution Impact
Each additional bit of converter resolution reduces the amplitude of the rounding error by a factor of two. For a sinusoidal input signal, this relationship yields a six decibel increase in the signal to noise ratio for every additional bit. Designing for high resolution signals requires twelve or sixteen bit converters to keep this noise floor below the level of the analog thermal noise.
However, higher resolution converters increase both the cost and power consumption of the module, forcing the designer to select the minimum resolution that meets the system requirements.
Spectral Distribution
The power of this rounding error is distributed evenly across the Nyquist bandwidth of the converter. By sampling the signal at a higher rate, the noise is spread over a wider band. This allows a digital filter to remove the out of band noise.
System Verification
Verifying that the system is not limited by this digital noise requires measuring the total harmonic distortion plus noise of the receiver. If the measured noise floor is higher than the calculated theoretical limit, the issue may lie in clock jitter or board layout problems. Testing is conducted with a high purity signal generator to ensure that the input signal does not introduce its own noise.
This verification is essential for high performance instrumentation.
