Analog and Digital Signals

To travel from one point to another, data must be transformed into electromagnetic signals. These signals can take either analog or digital forms, depending on how they represent information across a transmission medium.

1. Analog vs. Digital Paradigms

At the physical layer of networking, understanding the distinction between data types and signal representations is essential.

Analog Data and Signals: Analog data is continuous and takes on continuous values. Similarly, an analog signal has an infinite number of values within a defined range. As time progresses, the transition from one value to another is completely seamless.
Digital Data and Signals: Digital data has discrete states and takes on discrete values (such as binary $0$ and $1$ ). A digital signal can only have a limited, discrete number of defined values.

text

Analog Signal (Continuous)            Digital Signal (Discrete States)
     +V   .-.     .-.                      +V  ┌───┐       ┌───┐
         /   \   /   \                         │   │       │   │
  0V ---/-----\-/-----\---              0V ────┘   └───┬───┘   └───
       '       '       '                           │   │
     -V                                        -V  └───┘

In typical data communications, we employ periodic analog signals to represent carrier waves and nonperiodic digital signals to represent binary streams over short-haul links.

2. Mathematical Modeling of Periodic Analog Signals

A simple periodic analog signal, such as a sine wave, cannot be decomposed into simpler signals. It is modeled mathematically as:

$s(t) = A \sin(2\pi f t + \phi)$

Where:

$A$ is the Peak Amplitude (representing the maximum strength of the signal, usually in Volts).
$f$ is the Frequency (the number of cycles completed in one second, measured in Hertz, $\text{Hz}$ ).
$\phi$ is the Phase (describing the position of the wave relative to time $t = 0$ , in degrees or radians).

Period and Frequency

The period ( $T$ ) is the amount of time, in seconds, required to complete one full cycle. Frequency ( $f$ ) and period are mathematical inverses:

$f = \frac{1}{T} \quad \text{and} \quad T = \frac{1}{f}$

| Unit of Period ( $T$ ) | Equivalent | Unit of Frequency ( $f$ ) | Equivalent | | :--- | :--- | :--- | :--- | | Seconds ( $\text{s}$ ) | $1\text{ s}$ | Hertz ( $\text{Hz}$ ) | $1\text{ Hz}$ | | Milliseconds ( $\text{ms}$ ) | | Kilohertz () | | | Microseconds () | | Megahertz () | | | Nanoseconds () | | Gigahertz () | |

Phase Shift and Wavelength

Phase describes the fractional offset of the wave starting point from the origin. For example, a $90^\circ$ ( $\frac{\pi}{2}$ radians) phase shift moves the waveform origin to its positive peak.
Wavelength ( $\lambda$ ) measures the physical distance a simple signal propagates in a medium during a single period. It relates to the propagation speed ( $v$ ) of the medium:

$\lambda = v \times T = \frac{v}{f}$

3. Time-Domain vs. Frequency-Domain Representation

A signal can be analyzed in two distinct domains:

text

Time-Domain (Amplitude vs. Time)       Frequency-Domain (Amplitude vs. Freq)
    Amp                                    Amp
     ^      .-.     .-.                     ^
     |     /   \   /   \                    |      |
     |----/-----\-/-----\----> t            |      |
     |   '       '       '                  +------|------------> Freq
     0                                      0      f1

Time-Domain Plot: Shows changes in signal amplitude over time. It is useful for observing the actual waveform shape and measuring phase.
Frequency-Domain Plot: Shows the relationship between frequency and peak amplitude. A complete, single-frequency sine wave is represented by a single vertical spike.
Composite Signals: In data communications, a single-frequency sine wave is useless for sending information. We must send composite signals, which are composed of multiple sine waves of varying frequencies, amplitudes, and phases (derived using Fourier analysis).
Bandwidth: The bandwidth ( $B$ ) of a composite signal is the difference between its highest ( $f_h$ ) and lowest ( $f_l$ ) frequencies:

$B = f_h - f_l$

4. Digital Signals and Transmission Modes

Digital signals represent discrete states. If a digital signal has $L$ levels, the number of bits sent per signal level ( $n$ ) is:

$n = \log_2(L)$

Baseband vs. Bandpass Transmission

Baseband Transmission: Transmitting a digital signal directly over a channel without modification. This requires a low-pass channel (a channel with a bandwidth starting at $0\text{ Hz}$ ). Preserving the precise shape of a digital signal requires an extremely wide or infinite bandwidth.
Bandpass Transmission: Converting the digital signal into an analog signal using modulation (e.g., modifying amplitude, frequency, or phase of a high-frequency carrier wave). This allows transmission over a bandpass channel (where the bandwidth does not start at $0\text{ Hz}$ ), such as wireless mediums or telephone copper networks.

5. Transmission Impairments

Signals travel through imperfect transmission media, which causes impairments. The received signal never matches the transmitted signal perfectly.

text

       +-------------+     Impairments:
Tx --->| Trans Medium|---> Rx   1. Attenuation (Loss of signal strength)
       +-------------+          2. Distortion (Waveform changes shape)
                                3. Noise (Addition of unwanted signals)

Attenuation and Decibels ( $dB$ )

Attenuation is the loss of energy as a signal propagates. To measure signal strength differences, engineers use decibels:

$dB = 10 \log_{10}\left(\frac{P_2}{P_1}\right)$

Where $P_1$ is the input power and $P_2$ is the output power.

A negative decibel value ( $dB < 0$ ) indicates attenuation (loss of power).
A positive decibel value ( $dB > 0$ ) indicates amplification (gain of power).
Decibel values can be added or subtracted when analyzing cascaded systems.
To measure absolute power levels relative to $1\text{ mW}$ , engineers use $dBm$ :

$dBm = 10 \log_{10}\left(\frac{P_{\text{measured}}}{1\text{ mW}}\right)$

Distortion and Noise

Distortion: Changes the shape of a composite signal because different frequency components propagate at different speeds through a medium, arriving at the destination with different phase shifts.
Noise: Unwanted electrical energy inserted into the communication link (e.g., thermal noise, induction, crosstalk, or impulse spikes).
Signal-to-Noise Ratio (SNR): Measures the relative strength of the desired signal to the background noise:

$SNR = \frac{\text{Average Signal Power}}{\text{Average Noise Power}}$

$SNR_{dB} = 10 \log_{10}(SNR)$

6. Mathematical Limits on Data Rate

The maximum data rate of a communication channel is bounded by physical laws.

Noiseless Channels: Nyquist Bit Rate

For a noiseless channel, the theoretical maximum bit rate ( $C_{\text{Nyquist}}$ ) is restricted only by the channel bandwidth ( $B$ ) and the number of discrete signal levels ( $L$ ):

$C_{\text{Nyquist}} = 2 \times B \times \log_2(L)$

Noisy Channels: Shannon Capacity

In reality, all channels are noisy. The Shannon Capacity formula defines the absolute theoretical upper limit of the data rate ( $C_{\text{Shannon}}$ ) for a noisy channel:

$C_{\text{Shannon}} = B \times \log_2(1 + SNR)$

System Engineering Application

Engineers use both formulas to design physical-layer hardware. The Shannon capacity provides the absolute ceiling for throughput, while the Nyquist formula determines the specific number of discrete signal levels ( $L$ ) required to meet that target speed.

Here is a Python utility demonstrating how to compute both limits for a communication link:

python

import math
 
def calculate_channel_limits(bandwidth, snr_db, signal_levels):
    # Convert SNR from dB to linear scale
    snr_linear = 10 ** (snr_db / 10.0)
    
    # Calculate Shannon Capacity
    shannon_capacity = bandwidth * math.log2(1 + snr_linear)
    
    # Calculate Nyquist Bit Rate
    nyquist_rate = 2 * bandwidth * math.log2(signal_levels)
    
    return shannon_capacity, nyquist_rate
 
# Example: 3 kHz telephone line, 30 dB SNR, 4 signal levels
b = 3000          # Bandwidth in Hz
snr = 30          # SNR in dB
levels = 4        # Signal levels
 
shannon, nyquist = calculate_channel_limits(b, snr, levels)

7. Network Performance Metrics

To evaluate how well a network transfers signals, several metrics are used:

Bandwidth: Can refer to analog bandwidth (measured in $\text{Hz}$ , the range of frequencies) or digital bandwidth (measured in bits per second, the raw speed of transmission).
Throughput: The actual measure of how fast data can be sent through a network. Throughput ( $T$ ) is always less than the theoretical digital bandwidth ( $B$ ).
Latency (Delay): The total time it takes for a message to completely traverse the network. It is calculated as:

$\text{Latency} = \text{Propagation Time} + \text{Transmission Time} + \text{Queuing Time} + \text{Processing Delay}$

$\text{Propagation Time} = \frac{\text{Distance}}{\text{Propagation Speed}}$

$\text{Transmission Time} = \frac{\text{Message Size}}{\text{Bandwidth}}$

Bandwidth-Delay Product (BDP): Defines the volume of bits that can fill a physical link. It behaves like a pipe, where the cross-section is the bandwidth ( $B$ ) and the length is the round-trip propagation delay ( $D$ ):

$\text{BDP} = B \times D$

Summary

Signals can be analog (continuous) or digital (discrete). Signals must be periodic or nonperiodic to represent data.
A simple sine wave is characterized by its amplitude, frequency, and phase, and is analyzed in either the time domain or the frequency domain.
Fourier analysis shows that any composite signal is a mixture of multiple simple sine waves across a frequency spectrum, creating the signal's bandwidth.
Transmission impairments include attenuation (measured in $dB$ ), distortion (phase shifts over frequencies), and noise (extraneous background signals).
Nyquist formula limits the data rate of a noiseless channel: $C = 2 B \log_2(L)$ .
limits the data rate of a noisy channel: .

Assessed progress

Analog and Digital Signals

Analog and Digital Signals

1. Analog vs. Digital Paradigms

2. Mathematical Modeling of Periodic Analog Signals

Period and Frequency

Phase Shift and Wavelength

3. Time-Domain vs. Frequency-Domain Representation

4. Digital Signals and Transmission Modes

Baseband vs. Bandpass Transmission

5. Transmission Impairments

Attenuation and Decibels ( $dB$ )

Distortion and Noise

6. Mathematical Limits on Data Rate

Noiseless Channels: Nyquist Bit Rate

Noisy Channels: Shannon Capacity

System Engineering Application

7. Network Performance Metrics

Summary

Key Concepts

Analog vs. Digital Signals

Time and Frequency Domains

Transmission Impairments

Nyquist Bit Rate

Shannon Capacity

Quick Check

Computer Networks

Computer Networks