Solution:

In PCM, each sample is encoded with " n " bits whereas, in DPCM, the difference between the actual sample and its predicted sample value is encoded with " n " bits.

$ \mathrm{e}\left(\mathrm{nT}_{\mathrm{s}}\right)$ range is small when it is compared to the range of the actual sampled signal. Therefore we need a lesser number of quantization levels to perform quantization.

The bandwidth required for the DPCM signal is less when it is compared to the PCM.

Is used in (cellular, and telephony) personal communication services to digitize voice message signals.

DPCM predictor (delay line filter) predicts future sample values.

The predicted sample value may not be exactly equal to the actual sample value. But it is very close to the actual sample value.

In DPCM quantization error is reduced with a number of quantization levels i.e. less number of bits.

$\mathrm{n} \rightarrow$ less

$\mathrm{L} \rightarrow 2^{\mathrm{n}} \rightarrow$ less

$\mathrm{q}_{\mathrm{i}} \rightarrow$ less

Bitrates decreases $\mathrm{R}_{\mathrm{b}} \rightarrow$ less

$ \mathrm{BW}_{\mathrm{T}} \rightarrow$ less