Contents

1 Introduction 1
1.1 Technical Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.1 Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.2 Quantization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.3 Encoding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.4 Signal-to-Quantization-Noise Ratio (SQNR) . . . . . . . . . . . . . . 2
1.1.5 Types of Quantization . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.6 Companding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2 Analysis & Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Review of Current Requirements . . . . . . . . . . . . . . . . . . . . . . . . 3
1.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.5 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2 Introduction 1
2.1 Technical Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2.1.1 Line Coding Techniques . . . . . . . . . . . . . . . . . . . . . . . . . 1
2.1.2 Power Spectral Density (PSD) . . . . . . . . . . . . . . . . . . . . . . 3
2.1.3 Mathematical Analysis of Average Power . . . . . . . . . . . . . . . . 3
2.2 Analysis & Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.3 Review of Current Requirements . . . . . . . . . . . . . . . . . . . . . . . . 4
2.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.5 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

3 Introduction 1
3.0.1 Differential Pulse Code Modulation (DPCM) . . . . . . . . . . . . . . 1
3.0.2 Delta Modulation (DM) . . . . . . . . . . . . . . . . . . . . . . . . . 1
3.0.3 Relevance and Applications . . . . . . . . . . . . . . . . . . . . . . . 1
3.1 Technical Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
3.1.1 Differential Pulse Code Modulation (DPCM) . . . . . . . . . . . . . . 2
3.1.2 Delta Modulation (DM) . . . . . . . . . . . . . . . . . . . . . . . . . 3
3.1.3 Adaptive Delta Modulation (ADM) . . . . . . . . . . . . . . . . . . . 3
3.2 Analysis & Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
3.3 Review of Current Requirements . . . . . . . . . . . . . . . . . . . . . . . . 4
3.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
3.5 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

1
Pulse Code Modulation (PCM) Report

Course Code: EC-503

Student Name: Rajdeep Dey
Roll Number: 12000323069
Course: Digital Communication Systems
Submission Date: September 16, 2025
1 Introduction
Pulse Code Modulation (PCM) is a widely employed digital technique for representing analog
signals in digital form. It plays a foundational role in many digital communication systems
such as telephony, digital audio broadcasting, and multimedia streaming. The fundamental
goal of PCM is to enable the accurate and efficient transmission, storage, and reproduction
of analog signals by converting them into a sequence of binary digits. The process of PCM
consists of three essential steps:

1. Sampling: The continuous-time analog signal is sampled at discrete time intervals.

2. Quantization: Each sampled value is mapped to a finite set of discrete amplitude

levels.

3. Encoding: The discrete quantized values are converted into binary code for digital
processing.

To ensure faithful representation of the original analog signal, the sampling rate must satisfy
the Nyquist criterion:
fs ≥ 2fmax
This prevents aliasing and guarantees that the analog signal can be exactly reconstructed
from its samples. Quantization introduces quantization error, the difference between the
actual sampled amplitude and its nearest quantization level. This manifests as quantization
noise, characterized by the Signal-to-Quantization-Noise Ratio (SQNR). Higher bit depths
improve accuracy but increase bandwidth requirements. To address this, companding tech-
niques are used:

• Compressing the dynamic range of the signal before quantization.

• Expanding the signal at the receiver.

Two companding laws dominate:

• A-Law: Used in Europe and international systems.

• µ-Law: Used in North America and Japan.

These are standardized in ITU-T G.711, enabling efficient digital voice transmission.

1.1 Technical Description

1.1.1 Sampling
According to the Nyquist-Shannon theorem:

fs ≥ 2fm

1
where fm is the highest frequency in m(t). Samples are:

m(n) = m(nTs ), n = 0, 1, 2, . . .
1
with Ts = .
fs

1.1.2 Quantization
Quantization step size:
Vmax − Vmin
∆=
L
Quantization error:
∆
e(n) = m(n) − mq (n), |e(n)| ≤
2
Quantization noise power:
∆2
Nq =
12

1.1.3 Encoding
Each quantized level is encoded into N bits:

L = 2N , Rb = N fs

1.1.4 Signal-to-Quantization-Noise Ratio (SQNR)

Signal power for sinusoid of amplitude A:

A2
Ps =
2
SQNR:
SQNRdB = 6.02N + 1.76 dB

1.1.5 Types of Quantization

• Uniform Quantization — constant ∆.

• Non-uniform Quantization — variable ∆, often via companding.

2
1.1.6 Companding
A-Law:
A|x| 1


 , 0 ≤ |x| < ,
F (x) = 1 + ln A A
1 + ln(A|x|) 1
, ≤ |x| ≤ 1,


1 + ln A A
where A ≈ 87.6.
µ-Law:
ln(1 + µ|x|)
F (x) = , 0 ≤ |x| ≤ 1,
ln(1 + µ)
where µ ≈ 255.

1.2 Analysis & Discussion

PCM is robust and simple but suffers from quantization noise. Increasing N reduces noise
but requires more bandwidth. Companding enhances low-level signals without increasing N .
The ITU-T G.711 standard uses:

fs = 8 kHz, N = 8, Rb = 64 kbps.

Limitations:

• Non-linear distortion from companding.

• Incompatibility between A-Law and µ-Law systems.

• High bit rate requirement for high-fidelity audio.

Recent advances include adaptive quantization, packetized PCM for VoIP, and extensions
like G.711.1 (wideband audio).

1.3 Review of Current Requirements

• Adaptive quantization techniques.

• Wideband PCM (beyond 8 kHz).

• Integration with IP-based systems.

• Compression extensions like G.711.0.

• Low-power PCM for IoT devices.

3
1.4 Conclusion
Pulse Code Modulation (PCM) is a fundamental technique in digital communication for
converting analog signals into digital format. By sampling the signal according to the Nyquist
criterion and quantizing the samples, PCM ensures accurate representation of the original
signal. However, quantization introduces noise, which can be mitigated by increasing bit
depth or applying companding techniques like A-Law and µ-Law.
These companding laws effectively improve the dynamic range and signal quality without
demanding higher bit rates, making them essential in telephony. PCM’s robustness, ease of
error correction, and compatibility with digital systems have solidified its role in audio, voice,
and multimedia communication.
Despite advances in audio compression, PCM and companding remain widely used stan-
dards due to their simplicity and proven performance. Understanding PCM and its compo-
nents is crucial for designing efficient, high-quality digital communication systems.

1.5 References
1. T. Yoneyama and S. Nishida, ”Nonradiative dielectric waveguide for millimeter wave
integrated circuits”, IEEE Trans. Microwave Theory Tec., vol. 29, pp.1188-1192, Nov.
1981.

2. K. Wu, L. Han, ”Hybrid integration technology of planar circuits and NRD-guide for
cost effective microwave and millimeter-wave applications”, IEEE Trans. Microwave
Theory and Tech., vol. 45, pp.946-954, June 1997.

3. K. Wu, Y. Cassivi, ”Recent advances of hybrid planar/NRDguide technology for millimeter-

wave integrated circuit design,” TELSIKS’01, Nis, Yugoslavia, Sep. 19-21.

4
Line Coding Techniques Report

Course Code: EC-503

Student Name: Rajdeep Dey
Roll Number: 12000323069
Course: Digital Communication Systems
Submission Date: September 16, 2025
2 Introduction
Line coding is a fundamental technique in digital communications that involves transforming
binary data bits into digital signals suitable for transmission over physical media. The im-
portance of line coding stems from the need to ensure that digital data can be accurately and
efficiently transmitted, synchronized, and recovered with minimal error under real physical
channel constraints.
Key objectives and properties of line coding include:
• Data Representation: Converting digital bits into electrical pulses or waveforms
suitable for the communication channel.

• Synchronization: Embedding timing information within the signal waveform to en-

able the receiver to recover the transmitter’s clock, essential for proper bit delimitation
and error-free decoding.

• DC Component Elimination or Minimization: Prevents baseline wander and

allows the use of transformers or capacitive coupling, which cannot pass DC signals.

• Bandwidth Utilization: Designing signals with spectral properties that fit the chan-
nel bandwidth efficiently and reduce interference.

• Error Detection: Some line codes have inherent capabilities to detect certain types
of errors based on waveform violations.

• Noise Immunity: Maximizing signal robustness to minimize bit errors even in noisy
environments.

• Complexity and Practicality: Balancing intricate design requirements with prac-

tical implementation feasibility.
In essence, line coding facilitates reliable digital communication by crafting signal waveforms
that maintain signal integrity, enable clock recovery, and optimize spectrum usage according
to system requirements.

2.1 Technical Description

2.1.1 Line Coding Techniques
Line coding techniques dictate how bits are mapped into signal levels or transitions on the
transmission medium. Major classes include:

Unipolar Line Coding

• Uses a single polarity voltage to represent binary ‘1’ and zero voltage to represent ‘0’.

• Common Types:

1
 – Unipolar NRZ (Non-Return-to-Zero): Maintains the signal level for the
entire bit interval. (
A, b = 1
s(t) = , 0≤t<T
0, b = 0
– Limitations: Strong DC component, poor synchronization due to no guaranteed
transitions.

Polar Line Coding

• Uses two polarities to represent bits: +A for ‘1’ and −A for ‘0’.
• Key Types:
– Polar NRZ-L: Direct mapping of bits to positive or negative voltage.
s(t) = A(2b − 1)
– Polar NRZ-I: Signal transitions occur on ‘1’; no change on ‘0’.
• Advantages: No DC component, reduces baseline wander.
• Drawbacks: Long runs of identical bits lead to synchronization issues.

Bipolar Line Coding

• Employs three voltage levels: positive, zero, and negative.
• Example: Alternate Mark Inversion (AMI): Logical ‘0’ is zero volts, logical ‘1’s
alternate between +A and −A.
si (t) = A × (−1)i

• Advantages: Zero average voltage (no DC), simplifies error detection as bipolar vio-
lations indicate errors.

Manchester Coding
• Combines data and clock into a single self-synchronizing signal.
• Each bit period T contains a transition in the middle:
– ‘1’ → High to Low transition.
– ‘0’ → Low to High transition.
• Bandwidth ≈ 2Rb .
• Advantages: Excellent synchronization, no DC component.
• Drawbacks: Doubles required bandwidth compared to NRZ.

2
Differential Manchester Coding

• Transition at bit start for clocking; mid-bit transition depends on data.

• Allows clock recovery with fewer transitions compared to Manchester encoding.

2.1.2 Power Spectral Density (PSD)

PSD describes how signal power is distributed with frequency, critical for channel design and
interference management.
" ∞
#
2
|P (f )| X
S(f ) = 1+2 Rk cos(2πkf T )
T k=1

where:

• P (f ) = Fourier transform of the pulse shape of duration T .

• Rk = E[an an+k ] = autocorrelation of the bit sequence.

PSD of Common Line Codes:

• Unipolar NRZ:
A2 T
SU N RZ (f ) = sinc2 (f T ) + δ(f )
4
Strong DC component.

• Polar NRZ:
SP N RZ (f ) = A2 T sinc2 (f T )
No DC component.

• AMI: PSD shows null at DC due to zero average voltage.

• Manchester: Null at DC and at bit rate; bandwidth ≈ 2Rb .

2.1.3 Mathematical Analysis of Average Power

For line codes, average power is:
Z T
1
Pavg = |s(t)|2 dt
T 0

A2
• For Unipolar NRZ: Pavg = .
2
A2
• For Polar NRZ: Pavg = .
4
3
2.2 Analysis & Discussion
• Synchronization Capabilities: Codes with guaranteed transitions improve clock
recovery — important in Manchester encoding.

• Presence of DC Component: DC components limit coupling options and cause

baseline wander; thus, codes like AMI are preferred to eliminate DC.

• Error Detection: Bipolar codes’ violation patterns can indicate transmission errors.

• Bandwidth Efficiency: Manchester coding doubles bandwidth compared to NRZ

but maximizes synchronization.

• Complexity: Polar and bipolar coding are simpler to implement but may suffer syn-
chronization issues; Manchester addresses this at bandwidth cost.

2.3 Review of Current Requirements

• Communications with increased bit rates necessitate multi-level codes for spectral ef-
ficiency.

• Fiber optic and high-speed communication adopt line codes minimizing DC and con-
centrating energy within channel passband.

• Standard protocols (Ethernet, DSL) specify particular line coding schemes suited for
media and data rates.

• Research continues on advanced error-tolerant and bandwidth-efficient line codes.

2.4 Conclusion
Line coding is integral to reliable digital data transmission, balancing synchronization, band-
width, power efficiency, and error mitigation.
Different line coding schemes offer distinct advantages:

• Unipolar codes are simple but inefficient.

• Polar and Bipolar codes improve spectral properties and DC balance.

• Manchester codes provide robust clock recovery at the cost of bandwidth.

PSD analysis aids in understanding and optimizing bandwidth and signal integrity. Modern
communication systems need line codes that harmonize efficiency, resilience, and simplicity.

4
2.5 References
1. B. P. Lathi and Z. Ding, Modern Digital and Analog Communication Systems, 4th ed.,
Oxford Univ. Press, 2010.

2. J. G. Proakis, M. Salehi, and G. Bauch, Modern Communication Systems Using MAT-

LAB, 3rd ed., CENGAGE, 2013.

3. “Line Coding Techniques,” Tutorialspoint, 2023. Available: [Link]

com/digital_communication/digital_communication_line_codes.htm. [Accessed:
Sept. 16, 2025].

4. “Line Coding,” GeeksforGeeks, Oct. 2022. Available: [Link]

org/computer-networks/line-coding/. [Accessed: Sept. 16, 2025].

5. “Power Spectral Density,” Tutorialspoint, Jan. 2022. Available: [Link]

com/what-is-power-spectral-density. [Accessed: Sept. 16, 2025].

5
 Differential Pulse Code Modulation
(DPCM) and Delta Modulation (DM)

Course Code: EC-503

Student Name: Rajdeep Dey
Roll Number: 12000323069
Course: Digital Communication Systems
Submission Date: September 16, 2025
3 Introduction
In digital communication, analog signals are converted into digital signals for efficient pro-
cessing, transmission, and storage. While Pulse Code Modulation (PCM) is the fundamental
method to digitize analog signals, it often requires high bit rates and bandwidth due to trans-
mitting full sample values.
To improve efficiency, two important modulation techniques—Differential Pulse Code
Modulation (DPCM) and Delta Modulation (DM)—are commonly used. Both techniques
focus on encoding changes or differences between signal samples rather than absolute values,
thus reducing data redundancy and achieving compression.

3.0.1 Differential Pulse Code Modulation (DPCM)

• Predicts the current sample based on past reconstructed samples.
• Encodes the difference between the actual sample and this predicted value.
• Enables better compression for signals with high temporal correlation (e.g., speech,
audio).
• Balances compression with signal quality by handling prediction and quantization
noise.

3.0.2 Delta Modulation (DM)

• Simplifies encoding by sending only one bit per sample indicating whether the input
signal is rising or falling compared to the previous approximation.
• Uses a fixed step size to update the reconstructed signal, resulting in a staircase-like
approximation.
• Advantages include simplicity and low hardware complexity.
• Limitations include slope overload distortion and granular noise, mitigated by Adaptive
Delta Modulation (ADM) where step size varies dynamically.

3.0.3 Relevance and Applications

• Both DPCM and DM are foundational in digital speech coding, multimedia compres-
sion, and telemetry systems.
• They help reduce bandwidth requirements and power consumption in contemporary
communication devices.

This report presents an in-depth exploration of the working principles, technical specifica-
tions, and practical implications of DPCM and DM, along with their ongoing relevance in
digital communications.

1
3.1 Technical Description
3.1.1 Differential Pulse Code Modulation (DPCM)
DPCM is a predictive coding technique designed to reduce redundancy in the digitized
samples of an analog signal by encoding the difference between the actual sample and its
predicted value.

Signal Sampling The continuous analog signal x(t) is sampled at discrete intervals t =
nTs , producing samples x(n), where Ts is the sampling period.

Prediction Instead of encoding x(n) directly, DPCM uses a predictor to estimate the
current sample x̂(n) based on previously reconstructed samples:
M
X
x̂(n) = ak x̃(n − k)
k=1

where x̃(n − k) are the previously quantized and reconstructed samples, and ak are predictor
coefficients chosen to minimize the mean square error. The order M defines how many
previous samples are used.

Difference or Prediction Error The prediction error is computed as:

e(n) = x(n) − x̂(n)

Since x(n) and x̂(n) are highly correlated, e(n) typically has lower variance than x(n).

Quantization The difference signal e(n) is quantized into:

v(n) = Q[e(n)] = e(n) + q(n)

where Q[·] denotes quantization and q(n) is quantization noise.

Encoding and Transmission The quantized difference v(n) is encoded into a binary
bitstream and sent over the channel.

Reconstruction At the receiver:

x̃(n) = x̂(n) + v(n)

This reconstructed value is fed back into the predictor for future predictions, creating a
closed loop.

2
3.1.2 Delta Modulation (DM)
Delta Modulation encodes only the direction of change of the signal instead of the absolute
difference.

Working Principle At each sampling instant nTs , the encoder compares x(n) with x̃(n −
1) and outputs: (
+1, x(n) > x̃(n − 1)
d(n) =
−1, x(n) ≤ x̃(n − 1)
The receiver reconstructs:
x̃(n) = x̃(n − 1) + d(n)∆
where ∆ is the step size.

Key Characteristics

• Step Size ∆: Fixed change per sample interval.

• Sampling Rate fs : Must be much higher than Nyquist rate to track the signal
accurately.

Distortions

• Slope Overload Distortion: Occurs if ∆ is too small for rapid changes.

• Granular Noise: Occurs if ∆ is too large for slowly varying signals.

3.1.3 Adaptive Delta Modulation (ADM)

ADM dynamically adjusts ∆ depending on the input signal slope:

∆ = f (slope of x(n))

Mathematical Summary:

• d(n) = sign[x(n) − x̃(n − 1)]

• x̃(n) = x̃(n − 1) + d(n)∆

Bandwidth and Sampling The bit rate is:

R = fs × 1 bit/s

3
3.2 Analysis & Discussion
• Compression Efficiency: DPCM achieves better compression by using predictors,
while DM is less efficient.

• Signal Quality: DPCM provides higher SNR and reduces distortions; DM is simpler
but noisier.

• Complexity: DM is very simple (comparator + 1-bit quantizer), while DPCM requires

predictors.

• Bandwidth and Bit Rate: DM needs higher sampling rates; DPCM uses fewer bits
compared to PCM.

• Distortions: DM faces slope overload and granular noise; ADM mitigates these.
DPCM handles them better with finer quantization.

• Applications: DPCM is used in speech, image, and video coding. DM/ADM are
used in low-power systems.

3.3 Review of Current Requirements

• Adaptive Coding: Modern applications use ADPCM and ADM for variable signals.

• Low Latency: Needed for real-time communication.

• Power Efficiency: DM and ADM suit embedded devices with low power.

• Robustness: Error correction and prediction improve reliability in packet networks.

• Quality at Low Bit Rates: Research continues in predictive and machine learning
techniques.

3.4 Conclusion
DPCM and DM provide effective frameworks for bandwidth-efficient digital signal represen-
tation. DPCM balances quality and compression through prediction and quantization. DM
offers hardware simplicity but requires adaptive steps to reduce distortion. Both methods
are evolving with adaptive and machine learning techniques for modern telecommunication
challenges.

3.5 References
• S. Haykin, Digital Communications, 4th ed., Wiley India, 2011.

• A. B. Carlson and P. B. Crilly, Communication Systems, 5th ed., McGraw-Hill, 2010.

4
• B. P. Lathi and Z. Ding, Modern Digital and Analog Communication Systems, 4th ed.,
Oxford Univ. Press, 2010.

• J. G. Proakis, M. Salehi, and G. Bauch, Modern Communication Systems Using MAT-

LAB, 3rd ed., CENGAGE, 2013.

• “Differential Pulse Code Modulation,” Wikipedia, 2004. [Online]. Available: https://

[Link]/wiki/Differential_pulse-code_modulation. [Accessed: Sept.
16, 2025].

• “Delta Modulation,” Tutorialspoint, 2023. [Online]. Available: [Link]

com/digital_communication/delta_modulation.htm. [Accessed: Sept. 16, 2025].