Delta Modulation (DM) is a simple analog-to-digital conversion technique that encodes the difference (delta) between consecutive samples using 1-bit quantization.
Key Characteristics:
Example: For input samples [1.2, 1.5, 1.3, 1.6] and Δ = 0.2:
Encoded bits: [+1, +1, -1, +1]
Reconstructed: [1.0, 1.2, 1.4, 1.2, 1.4]
Delta modulation has two main types of noise, i.e slope overload distortion and granularity.
a) Slope overload occurs when the input signal changes too rapidly for the delta modulator to follow, resulting in a distorted approximation.
b) Granularity arises when the input signal changes too slowly, causing the modulator to oscillate between its two quantization levels, producing unwanted noise.
Slope Overload Condition:
Where: Δ = step size, T = sampling interval
Occurs when Δ is too small for small signal variations.
To mitigate noise in Delta modulation, adaptive delta modulation (ADM) was developed. ADM dynamically adjusts the step size based on the input signal's characteristics, allowing it to better track both rapid changes and gradual variations, thus reducing both slope overload and granular noise.
The primary goal is to dynamically adjust step size (Δ) to increase Δ during steep signal slopes (prevent overload) and to decrease Δ for small variations (reduce granular noise)
| Type | Description |
|---|---|
| Continuously Variable Slope Delta Modulation (CVSD) | Δ changes continuously based on signal slope |
| Step-Dela Modulation | Δ changes in discrete steps |
| Feature | DM | PCM | DPCM |
|---|---|---|---|
| Bits/sample | 1 | 8-16 | 2-4 |
| Complexity | Low | High | Medium |
| Quality | Low (voice only) | High | Medium |
Answer: It encodes only the difference (delta) between samples as +1/-1.
Answer: When the input signal changes faster than the staircase can follow (Δ too small).
Answer: By dynamically adjusting step size (Δ) to match signal slope.
Answer: Poor performance for rapidly changing signals (slope overload) and granular noise.