π Lecture 12 β Problems 4.3
π Overview: This lecture explores the fundamental characteristics of sine waves, including period, frequency, and phase. It explains how signals can be controlled by these attributes and introduces the concepts of time and frequency domains, culminating in the idea of composite signals composed of multiple sine waves.
ποΈ Topics Covered
The lecture begins by solving problems on period and frequency calculation. It then discusses another way to look at frequency, including the two extremes of frequency (zero and infinite). The concept of phase is introduced and explained through examples. Signal control via amplitude, frequency, and phase is demonstrated. The distinction between time domain and frequency domain plots is made, followed by an introduction to composite signals and their decomposition using Fourier analysis.
π Lecture Summary
Frequency and Period
A sine waveβs period (T) is the time it takes to complete one cycle. Its frequency (f) is the number of cycles per second. They are inversely related.
π Definition β Period (T): The time needed to complete one cycle. π Formula: T = 1/f β The period is the inverse of the frequency. π Example: For a sine wave with a frequency of 6 Hz, the period is T = 1/6 = 0.17 seconds. For a sine wave that completes one cycle in 4 seconds, its frequency is f = 1/4 = 0.25 Hz.
Another Way to look at Frequency
Frequency can be viewed as the measurement of the rate of change of a sine wave. A higher frequency means the signal changes from its lowest to highest point more quickly. A 40 Hz signal has half the frequency of an 80 Hz signal, so each cycle takes twice as long to complete.
π Definition β Rate of Change: The speed at which a sine wave moves from its lowest to its highest point. π Example: A 40 Hz signal has a longer period than an 80 Hz signal.
Two Extremes Frequency
If a signal maintains a constant voltage level and does not change, it completes zero cycles, so its frequency is zero. If a signal has instantaneous changes, it implies an infinite frequency.
π Definition β Zero Frequency: A signal that never changes, completing no cycles. π Definition β Infinite Frequency: A signal that changes instantaneously. π‘ Why this matters: These extremes define the boundaries of signal behavior. A constant DC signal has zero frequency, while a perfect square wave would require infinite frequency components.
Phase
Phase describes the position of a waveform relative to time zero. It indicates the amount of shift (backward or forward) along the time axis for the first cycle. Phase is measured in degrees or radians, where 360 degrees = 2Ο radians.
π Definition β Phase: The amount of shift of a waveform relative to time zero. π Relationship: A phase shift of 360 degrees corresponds to a shift of a complete period. A phase shift of 180 degrees is half a period, and 90 degrees is a quarter period. π Example: A sine wave offset by 1/6 of a cycle has a phase of (1/6) * 360 = 60 degrees.
Control of Signals
A signal can be controlled by three attributes: Amplitude, Frequency, and Phase. Varying these attributes allows for the creation of different signals for communication.
π Definition β Control of Signals: The ability to manipulate a signal by changing its amplitude, frequency, and phase. π‘ Why this matters: This is the fundamental principle behind modulation, where information is encoded by changing these signal characteristics.
Time and Frequency Domain
Time domain plots show signal amplitude changing with respect to time. Phase and frequency are not explicitly measured on a time domain plot. A frequency domain plot shows the relationship between amplitude and frequency.
π Definition β Time Domain: A plot showing changes in signal amplitude with respect to time. π Definition β Frequency Domain: A plot showing the maximum amplitude of a signal with respect to its frequency components. π Example: A low-frequency signal in the frequency domain corresponds to a signal with a longer period in the time domain. A signal changing rapidly in the time domain corresponds to a high frequency in the frequency domain.
Composite Signals
A composite signal is the second type of analog signal, composed of multiple sine waves. Any periodic signal, no matter how complex, can be decomposed into a collection of sine waves. This process is called Fourier analysis.
π Definition β Composite Signal: A periodic signal made up of multiple sine wave components. π Definition β Fourier Analysis: The mathematical method used to decompose a composite signal into its constituent sine waves, each with a measurable amplitude, frequency, and phase. π Example: A periodic signal can be decomposed into two sine waves. One has a frequency of '6' and the other has a frequency of '0' (a DC component). Adding these two sine waves point by point results in the original composite signal. In the time domain, a single point is visible, while in the frequency domain, two component frequencies are present.
β Key Takeaways
A student must understand the inverse relationship between frequency and period and be able to calculate one from the other. The concepts of zero and infinite frequency define the extremes of signal behavior. Phase is a measure of a waveform's shift relative to time zero, measured in degrees or radians. The three attributes for controlling a signal are amplitude, frequency, and phase. Finally, any complex periodic signal can be broken down into a set of simple sine waves using Fourier analysis, resulting in a different representation in the time and frequency domains.
π§ Quick Revision Questions
- What is the period of a sine wave with a frequency of 10 Hz?
- What does a phase shift of 180 degrees represent in terms of a cycle?
- What is the frequency of a signal that never changes its voltage level?
- What is the key difference between a time domain plot and a frequency domain plot?
- What mathematical technique is used to decompose a complex periodic signal into its constituent sine waves?
π Lecture 13 β Frequency Spectrum / Bandwidth
π Overview: This lecture defines Frequency Spectrum and Bandwidth for periodic signals, explains various definitions of bandwidth (including 3 dB, Equivalent Noise, Null-to-Null, and Power Bandwidth), and introduces digital signalsβtheir representation via Bit Interval and Bit Rate, and their frequency spectrum characteristics. Understanding these concepts is essential for selecting transmission media and ensuring accurate signal reconstruction.
ποΈ Topics Covered
The lecture begins with the fundamental definitions of Frequency Spectrum and Bandwidth, including a worked example. It then explores alternative bandwidth definitions: 3 dB (Half Power) Bandwidth, Equivalent Noise Bandwidth, Null-to-Null (Zero Crossing) Bandwidth, and Power Bandwidth. Finally, it introduces digital signals, explaining Bit Interval, Bit Rate, and the infinite frequency spectrum of a digital signal whose significant components can be transmitted for reconstruction.
π Lecture Summary
Frequency Spectrum / Bandwidth
The frequency spectrum of a signal is the collection of all the component frequencies it contains. It is shown using a frequency domain graph. The bandwidth of a signal is the width of the frequency spectrum. In other words, Bandwidth refers to the range of the component frequencies, and Frequency Spectrum refers to the elements within that range.
π Definition β Frequency Spectrum: the collection of all the component frequencies a signal contains. π Definition β Bandwidth: the width of the frequency spectrum (range of component frequencies).
How to calculate Bandwidth?
To calculate Bandwidth, subtract the lowest frequency from the lowest frequency.
π Formula: ( \text{Bandwidth} = f_h - f_l ) β (highest frequency minus lowest frequency in the spectrum)
Example 4.8 If a periodic signal is decomposed into five sine waves with frequencies 100, 300, 500, 700, and 900 Hz, what is the Bandwidth?
Solution: ( f_h = 900 , \text{Hz} ) ( f_l = 100 , \text{Hz} ) ( \text{Bandwidth} = 900 - 100 = 800 , \text{Hz} ) π Example: For a signal composed of 100, 300, 500, 700, and 900 Hz components, the bandwidth is 900 β 100 = 800 Hz.
Other Definitions of Bandwidth
3 dB Bandwidth or Half Power Bandwidth For the magnitude spectrum ( |H(f)| ), the range of the spectrum that does not fall lower than ( \frac{1}{\sqrt{2}} ) times the maximum. This is called the 3 dB bandwidth because a ratio of ( \frac{1}{\sqrt{2}} ) corresponds to -3 dB on a logarithmic scale.
Equivalent Noise Bandwidth The width of a fictitious rectangular spectrum created to have the same power in the rectangular band as the power of the signal in positive frequencies.
Null-to-Null Bandwidth or Zero Crossing Bandwidth If the maximum frequency in a spectrum is ( f_o ), the first null above and below ( f_o ) will be ( f_1 ) and ( f_2 ), where ( f_1 - f_2 ) is the Null-to-Null Bandwidth.
Power Bandwidth Frequency band in which 99% of the total power resides.
Digital Signals
In addition to being represented by Analog Signals, data can also be represented by a digital signal.
Bit Interval and Bit Rate
Most digital signals are aperiodic and thus Period and Frequency are not the appropriate terms to describe them.
π Definition β Bit Interval (seconds): Time required to send one single bit. π Formula: Bit Interval = ( \frac{1}{\text{Bit Rate}} ) (seconds per bit) β time per single bit.
π Definition β Bit Rate (bps): Number of bits sent per second. π Formula: Bit Rate = ( \frac{1}{\text{Bit Interval}} ) (bits per second) β bits transmitted per second.
Frequency Spectrum of a Digital Signal
- Frequency spectrum of a digital signal contains an infinite number of frequencies with different amplitudes.
- Ideally we want to send all the components, but if we send only those components whose amplitudes are significant, we can still recreate the digital signal with reasonable accuracy at the receiver. π‘ Why this matters: This principle explains why we can transmit digital data over limited-bandwidth channels (e.g., telephone lines) by filtering out very high-frequency components, as long as the significant ones (containing most of the signal energy) are preserved.
β Key Takeaways
A student must remember that Frequency Spectrum is the collection of all component frequencies, and Bandwidth is the width of that range, calculated as highest minus lowest frequency. The lecture defines several bandwidth types: 3 dB (Half Power) where magnitude does not fall below 1/β2 of max; Equivalent Noise Bandwidth from a rectangular power-equivalent spectrum; Null-to-Null Bandwidth between first spectral nulls; and Power Bandwidth containing 99% of power. For digital signals, the key parameters are Bit Interval (time per bit) and Bit Rate (bits per second), and the signal's frequency spectrum is infiniteβbut only significant-amplitude components are needed for reasonable reconstruction.
π§ Quick Revision Questions
- What is the difference between Frequency Spectrum and Bandwidth?
- If a periodic signal contains frequencies 200, 400, 600, 800, and 1000 Hz, calculate its bandwidth.
- Define 3 dB Bandwidth and explain why it is also called Half Power Bandwidth.
- What are the definitions of Bit Interval and Bit Rate, and what is the mathematical relationship between them?
- Why can a digital signal be recreated with reasonable accuracy even if its high-frequency components are omitted?
π Lecture 14 β Conversions
π Overview: This lecture explores how information must be converted into signals for transmission across communication media. It introduces the four fundamental types of conversion (Digital-to-Digital, Analog-to-Digital, Digital-to-Analog, and Analog-to-Analog), with a detailed focus on Digital-to-Digital Conversion and its primary encoding techniques: Unipolar and Polar (specifically NRZ-L and NRZ-I).
ποΈ Topics Covered
The lecture begins with an introduction to why information must be transformed into signals, followed by a high-level overview of the four types of conversions. It then dives deep into Digital-to-Digital Conversion, explaining its purpose and the three main encoding techniques: Unipolar, Polar, and Bipolar. The focus is primarily on Unipolar encoding (with its pros and cons, including DC Component and Synchronization problems) and Polar encoding, specifically NRZ-L and NRZ-I.
π Lecture Summary
Introduction
Information must be transformed into signals before it can be transported across communication media. How this transformation occurs depends on both the original format of the information and the format used by the communication device. For example, sending a letter via smoke signals requires knowing which smoke patterns correspond to which words. The words are the Information, and the puffs of smoke are the representation of that information.
Introduction to the type of Conversions
- Digital-to-Digital Conversion: Data stored in a computer (0βs and 1βs) is converted to digital signals for transmission.
- Analog-to-Digital Conversion: Analog signals (e.g., telephone conversations) are converted to digital signals to decrease the effect of noise; this is called Digitizing an Analog Signal.
- Digital-to-Analog Conversion: Digital signals from a computer are converted to analog signals to be sent over media designed for analog signals (e.g., telephone lines); this is called Modulating a digital Signal.
- Analog-to-Analog Conversion: Analog signals (e.g., voice or music) are modulated onto a higher-frequency signal for long-distance transmission through media like air; this is called Modulating an analog Signal.
Types of Conversions
Digital-to-Digital Conversion
Digital-to-Digital conversion/encoding is the representation of digital information by a digital signal. For example, when you transmit data from a computer to a printer, both the original and transmitted data are digital. In this type of encoding, 1βs and 0βs generated by the computer are translated into voltage pulses that can be propagated over the wire.
π‘ Why this matters: This is the most basic form of encoding used in local connections like USB or printer cables, where the medium inherently supports digital signals.
Types of Digital-to-Digital Encoding
The three main types are Unipolar, Polar, and Bipolar.
UNIPOLAR
- Encoding is simple, with only one technique in use.
- It is simple and primitive, but almost obsolete today.
- Studying it provides an introduction to concepts and problems involved with more complex encoding systems.
- A digital transmission system works by sending voltage pulses on the transmission medium.
- One voltage level stands for binary 0, while the other stands for binary 1.
- It is called Unipolar because it uses only one polarity. This polarity is assigned to one of the two binary states (usually a β1β), while the other state (usually a β0β) is represented by zero voltage.
Pros and Cons of Unipolar Encoding
- PROS: Straightforward and simple, inexpensive to implement.
- CONS: DC Component and Synchronization problems.
DC Component
- The average amplitude of a unipolar encoded signal is non-zero, called a DC Component (a component with zero frequency).
- When a signal contains a DC Component, it cannot travel through a transmission medium that cannot handle DC components.
π Definition β DC Component: A non-zero average amplitude of a signal, which prevents it from passing through media that cannot handle zero-frequency components.
Synchronization
- When the signal is unvarying, the receiver cannot determine the beginning and ending of each bit.
- The Synchronization Problem can occur when data consists of long streams of 1βs or 0βs.
- Therefore, the receiver has to rely on a TIMER.
- π Example: Consider a bit rate of 1000 bps (1000 bits in 1 second, so 1 bit in 0.001 second). If a +ve voltage lasts 0.005 sec, it reads five 1βs. Sometimes it stretches to 0.006 seconds, and an extra one bit is read by the receiver.
- Solution: A Separate Parallel Line carries a clock pulse and allows the receiver to resynchronize its timer to that of the signal. However, doubling the number of transmission lines increases cost and proves uneconomical.
POLAR
Polar encoding has 3 subcategories: Non Return to Zero (NRZ), Return to Zero (RZ), and Bi phase (two of which have multiple variations).
- Polar encoding uses two voltage levels β one positive and one negative.
- By using two voltage levels, the average voltage level on the line is reduced, and the DC Component problem of unipolar encoding is alleviated.
Non Return to Zero (NRZ) In NRZ Encoding, the level of the signal is either positive or negative.
NRZ-L (Level)
- The level of the signal depends on the type of bit it represents.
- A +ve voltage usually means the bit is a 1, and a βve voltage means the bit is a 0 (or vice versa).
- β οΈ Problem with NRZ-L: When long streams of 0βs or 1βs are in the data, the receiver receives a continuous voltage and must determine how many bits are sent by relying on its clock, which may or may not be synchronized with the senderβs clock.
NRZ-I (Invert)
- The inversion of the level represents a 1 bit.
- A bit 0 is represented by no change.
- NRZ-I is superior to NRZ-L due to the synchronization provided by a signal change each time a 1 bit is encountered.
- The string of 0βs can still cause a problem, but since 0βs are not as likely, they are less of a problem.
π Definition β NRZ-I: A polar encoding technique where a transition (inversion) at the beginning of a bit period represents a binary 1, and no change represents a binary 0.
π Formula/NRZ Logic: NRZ-L: Voltage level = bit value (e.g., +V for 1, -V for 0). β Plain-English: The signal's voltage directly tells you the bit. NRZ-I: Transition = bit 1, no transition = bit 0. β Plain-English: You look for a change in voltage to know if the bit is a 1.
β Key Takeaways
The lecture establishes that all data must be converted into appropriate signals for transmission, and the conversion type depends on both the data's original format and the medium. Digital-to-Digital conversion is the representation of binary data as voltage pulses, with Unipolar being the simplest but flawed due to DC Component and Synchronization issues. Polar encoding solves the DC component problem by using two voltage levels. Within Polar, NRZ-I is superior to NRZ-L because the signal change on a '1' bit provides inherent synchronization, making long streams of 1's less problematic than long streams of 0's.
π§ Quick Revision Questions
- What are the four main types of conversions discussed in this lecture?
- Explain the two main disadvantages of Unipolar encoding.
- What is the key difference between NRZ-L and NRZ-I encoding schemes?
- How does Polar encoding solve the DC Component problem present in Unipolar encoding?
- Why is NRZ-I considered superior to NRZ-L for synchronization?
π Lecture 15 β Conversions
π Overview: This lecture explores various methods of digital-to-digital encoding used to synchronize sender and receiver clocks, prevent loss of timing during long sequences of identical bits, and manage bandwidth. It covers Return to Zero (RZ), Biphase (Manchester and Differential Manchester), and Bipolar (AMI, B8ZS, HDB3) encoding schemes, highlighting their mechanisms, advantages, and trade-offs.
ποΈ Topics Covered
This lecture covers three main encoding categories: Polar Encoding with Return to Zero (RZ) encoding, Biphase Encoding including Manchester and Differential Manchester, and Bipolar Encoding including Alternate Mark Inversion (AMI) and its two variations B8ZS and HDB3. Each scheme is explained in terms of synchronization, voltage levels, bandwidth requirements, and problem-solving for long strings of zeros.
π Lecture Summary
Return to Zero (RZ)
Any time data contains long strings of 1βs or 0βs, the receiver can lose its timing. In unipolar encoding, a separate timing signal was a solution, but this is both expensive and full of error. A better solution is to somehow include synchronization in the encoded signal, somewhat similar to NRZ-I, but it should work for both strings of 0 and 1. One solution is RZ encoding, which uses 3 values: Positive, Negative, and Zero. Signal changes not between bits but during each bit. Like NRZ-L, a +ve voltage means 1 and a βve voltage means 0, but unlike NRZ-L, half way through each bit interval, the signal returns to zero. A 1 bit is represented by positive to zero and a 0 is represented by negative to zero transition. The only problem with RZ encoding is that it requires two signal changes to encode one bit and therefore occupies more bandwidth. But of the three alternatives discussed, it is most effective.
Biphase Encoding
This is the best existing solution to the problem of synchronization. Signal changes at the middle of the bit interval but does not stop at zero; instead it continues to the opposite pole.
Manchester
Manchester encoding uses inversion at the middle of each bit interval for both synchronization and bit representation. Negative-to-Positive Transition = 1; Positive-to-Negative Transition = 0. By using a single transition for a dual purpose, Manchester achieves the same level of synchronization as RZ but with only two levels of amplitude.
Differential Manchester
In Differential Manchester encoding, inversion at the middle of the bit interval is used for synchronization, but the presence or absence of an additional transition at the beginning of the bit interval is used to identify a bit. A transition at the beginning means binary 0 and no transition at the beginning means binary 1. It requires 2 signal changes to represent binary 0 but only one to represent binary 1.
Bipolar Encoding
Like RZ, it uses three voltage levels. Unlike RZ, zero level is used to represent binary 0. Binary 1βs are represented by alternate positive and negative voltages.
Alternate Mark Inversion (AMI)
This is the simplest type of Bipolar Encoding. βMarkβ comes from Telegraphy, meaning 1. Alternate Mark Inversion means Alternate β1β Inversion. By inverting on each occurrence of 1, AMI accomplishes 2 things: the DC component is zero, and long sequences of 1βs stay synchronized. However, there is no mechanism of ensuring synchronization for a long stream of 0βs. Two variations are developed to solve this problem: B8ZS used in North America and HDB3 used in Europe & Japan. Both modify the original pattern of AMI only in the case of a long stream of zeroes.
π Definition β B8ZS: Convention adopted in North America to provide synchronization for a long string of zeros. The difference between AMI and B8ZS occurs only when 8 or more consecutive zeros are encountered. It forces artificial signal changes called violations. Each time eight 0βs occur, B8ZS introduces changes in the pattern based on the polarity of the previous 1 (the β1β occurring just before the zeros).
π Definition β HDB3: Alteration of AMI adopted in Europe and Japan. It introduces changes into AMI every time four consecutive zeros are encountered instead of waiting for eight zeros as in the case of B8ZS. As in B8ZS, the pattern of violations is based on the polarity of the previous 1 bit. Unlike B8ZS, HDB3 also looks at the number of 1βs that have occurred since the last substitution.
β Key Takeaways
For the exam, you must remember that RZ uses three voltage levels (positive, negative, zero) but requires double bandwidth. Manchester and Differential Manchester are biphase schemes that embed synchronization within each bit using mid-bit transitions. Manchester uses the transition direction to represent bits, while Differential Manchester uses the presence or absence of an initial transition. Bipolar encoding (AMI) uses zero for binary 0 and alternates voltage for binary 1, ensuring zero DC component but failing for long zero sequences. B8ZS (North America) solves this for 8+ zero runs by introducing violations based on previous 1 polarity, whereas HDB3 (Europe/Japan) acts on 4+ zero runs and also considers the count of 1βs since the last substitution.
π§ Quick Revision Questions
- What is the primary problem that RZ encoding solves compared to unipolar and NRZ schemes?
- How does Manchester encoding achieve synchronization while using only two amplitude levels?
- In Differential Manchester encoding, what does a transition at the beginning of a bit interval represent?
- What are the two key advantages of AMI encoding, and what is its main limitation?
- What is the key difference between B8ZS and HDB3 in terms of when they force violations?
π Lecture 16 β Conversions
π Overview: This lecture covers the process of converting analog signals into digital data for transmission, focusing on techniques like Pulse Amplitude Modulation (PAM) and Pulse Code Modulation (PCM). It explains why digitization is necessary and how to maintain signal quality during conversion, which is essential for applications like voice communication over long distances.
ποΈ Topics Covered
The lecture begins by briefly revisiting types of digital-to-digital encoding (unipolar, polar, bipolar) and an example using B8ZS encoding. It then dives into the core topic of analog-to-digital conversion, explaining the need for CODECs. The lecture details the two main steps: Pulse Amplitude Modulation (PAM) and Pulse Code Modulation (PCM), which includes quantization and binary encoding. Finally, it covers the Nyquist theorem for sampling rate and the calculation of bit rate.
π Lecture Summary
Types of Digital-to-Digital Encoding
This section briefly recaps the three main categories of digital-to-digital encoding: Unipolar, Polar, and Bipolar. An example is provided using B8ZS (Bipolar with 8-Zero Substitution), a line coding scheme used to handle long strings of zeros in bipolar encoding.
π Definition β B8ZS (Bipolar with 8-Zero Substitution): A line coding scheme that substitutes eight consecutive zeros with a special code containing deliberate violations of the bipolar rule to maintain synchronization.
π Example: Using B8ZS, encode the bit stream 10000000000100. Assume that the polarity of the first 1 is positive.
- Step 1: The stream starts with
1(positive). Then we have eight zeros (00000000), followed by000100. - Step 2: The eight consecutive zeros are substituted according to the B8ZS rule. The standard substitution for eight zeros is
000VB0VB, where V is a violation and B is a bipolar pulse. - Step 3: With the first
1being positive, the encoded sequence becomes:+1(first bit)000+-0-+(for the 8 zeros)00-+00(for the final000100). - (Note: The exact
000VB0VBpattern depends on the polarity of the preceding non-zero pulse.)
Analog-to-Digital Conversion
We sometimes need to digitize an analog signal. To send human voice over a long distance, we need to digitize it, since digital signals are less prone to Noise. This is called Analog-to-Digital Conversion or Digitizing an Analog Signal. This type of conversion requires a reduction of the potentially infinite number of values in the analog signal so that it can be converted to a digital bit stream with minimum loss of information. A CODEC (Coder-Decoder) is used for this purpose. Digital signals can take any of the forms discussed previously (unipolar, polar, bipolar). The problem is how to convert an analog signal from an infinite number of values to a discrete number of values without scarifying quality.
Pulse Amplitude Modulation (PAM)
This is the first step in Analog-to-Digital Conversion. This technique takes an Analog signal, Samples it, and Generates a series of Pulses based on the results of Sampling. Sampling means measuring amplitudes of signal at equal intervals. The original signal is sampled at equal intervals. PAM uses a technique called Sample & Hold which means at a given moment, the signal level is read, then held briefly. The pulses are of any amplitude (still analog, not digital). To make them digital, we need PCM.
π Definition β PAM (Pulse Amplitude Modulation): A technique that samples an analog signal at equal intervals and generates a series of pulses whose amplitudes represent the amplitude of the signal at each sampling instant. π‘ Why this matters: PAM is the first crucial step, but it is not yet digital. The pulse amplitudes are still analog values (continuous), so a further step is needed to convert these into discrete digital numbers.
Pulse Code Modulation (PCM)
PCM modifies the pulses created by PAM into a complete digital signal. It involves four separate processes:
- PAM
- Quantization
- Binary Encoding
- Digital/Digital Encoding
Quantization
PCM's first step is Quantization. βQuantization is a method of assigning integral values in a specific range to sampled instancesβ. The figure (in the lecture) shows a simple method of assigning sign and magnitude values to quantized samples.
π Definition β Quantization: The process of mapping the infinite number of analog (PAM) amplitudes to a finite set of discrete integer values (quantization levels).
Results of Binary Encoding
Each quantized value is translated into its seven-bit binary equivalent. The eighth bit indicates the sign (positive or negative). This creates a fully digital representation of the analog sample.
Result of PCM
The figure in the lecture shows the result of PCM of the original signal encoded finally into a unipolar signal. Only the first 3 values are shown. The final output is a series of digital pulses ready for transmission.
Sampling Rate
The accuracy of any digital reproduction of an analog signal depends upon the number of samples taken. The fundamental question is: How many samples are sufficient?
The Nyquist theorem states: The sampling rate must be at least twice the highest frequency present in the signal.
If the signal has a maximum frequency of f_max, then the sampling rate f_s must satisfy: f_s >= 2 * f_max.
π Formula: Nyquist Theorem β f_s >= 2 * f_max (Sampling rate must be at least twice the highest frequency of the analog signal to avoid aliasing).
π‘ Why this matters: Sampling at a rate lower than the Nyquist rate causes "aliasing," where high-frequency components are misinterpreted as lower frequencies, leading to an unrecoverable loss of information.
Bit Rate
The sampling rate is given by the Nyquist Theorem. The number of bits per sample is chosen according to the precision needed at the receiver end. The overall bit rate is the product of these two factors.
π Formula: Bit Rate β BitRate = SamplingRate * (bits per sample)
(Note: BitRate is the number of bits per second produced by the digitizer, SamplingRate is the number of samples per second, and bits per sample is the number of bits used to represent each sample's value.)
β Key Takeaways
A student must remember that the fundamental purpose of analog-to-digital conversion is to make analog signals robust for long-distance digital transmission by using a CODEC. The core process is PCM, which is built upon PAM, where the analog signal is sampled at discrete time intervals. The samples must then be quantized (converted to integer values) and binary-encoded to produce the final digital stream. The single most critical rule governing this process is the Nyquist theorem, which dictates that the sampling rate must be at least twice the highest frequency in the original signal to ensure perfect reconstruction. Finally, the total bit rate of the digital stream is calculated by multiplying the sampling rate by the number of bits used per sample.
π§ Quick Revision Questions
- What does CODEC stand for, and what is its primary function?
- What is the key difference between the pulses produced by PAM and the final digital signal produced by PCM?
- Explain the process of quantization in the context of PCM.
- State the Nyquist theorem and explain why it is crucial for analog-to-digital conversion.
- If an analog signal has a maximum frequency of 4 kHz and we use 8 bits per sample, what is the required sampling rate and the resulting bit rate of the digital stream?
π Lecture 17 β Conversion
π Overview: This lecture covers the process of converting digital data into analog signals for transmission over analog media like telephone lines. It explains the fundamental mechanisms of digital-to-analog modulation, the distinction between bit rate and baud rate, and provides detailed analysis of Amplitude Shift Keying (ASK) as the first modulation technique.
ποΈ Topics Covered
The lecture begins with an introduction to digital-to-analog conversion and the three characteristics of sine waves that can be modified. It then covers the mechanisms for modulating digital data into analog signals: ASK, FSK, PSK, and QAM. The important distinction between bit rate and baud rate is explained with an analogy. Carrier signals are introduced, and finally, a detailed examination of Amplitude Shift Keying (ASK) including its noise susceptibility, On-Off Keying, and bandwidth requirements is provided.
π Lecture Summary
Digital To Analog Conversion
Digital-to-analog conversion is the process of changing one of the characteristics of an analog signal based on the information in a digital signal. This is necessary when transmitting data from one computer to another using a public telephone line because telephone wires carry analog signals. The original digital data must be modulated on an analog signal that has been manipulated to look like two distinct values corresponding to binary 1 and binary 0.
π Definition β Digital-to-Analog Conversion: The process of changing one of the characteristics of an analog signal (amplitude, frequency, or phase) based on the information in a digital signal.
Variation in Characteristics of Sine Wave
A sine wave is defined by three characteristics: Amplitude, Frequency, and Phase. By changing one aspect of a simple electrical signal back and forth, we can use it to represent digital data. When we vary any one of these characteristics, we create a second version of that wave. If we then say that the original wave represents binary 1, the variation can represent binary 0, or vice versa.
Mechanisms for Modulating Digital Data to Analog Signals
Any of the three characteristics of a sine wave can be altered to modulate digital data into analog signals, giving at least three mechanisms:
- Amplitude Shift Keying (ASK)
- Frequency Shift Keying (FSK)
- Phase Shift Keying (PSK)
A fourth mechanism combines changes in both amplitude and phase called Quadrature Amplitude Modulation (QAM). QAM is the most efficient of these options and is the mechanism used in all modern modems.
Aspects of Digital to Analog Conversion
Before discussing specific methods of digital-to-analog modulation, two basic issues must be defined: Bit/Baud rate and Carrier signal.
Bit Rate & Baud Rate
- Bit rate: The number of bits transmitted during one second.
- Baud rate: The number of signal units per second that are required to represent that bit.
In discussions of computer efficiency, bit rate is more important. In data transmission, however, we are more concerned with how efficiently we can move data from place to place. The fewer signal units required, the more efficient the system and less bandwidth required to transmit more bits, so we are more concerned with baud rate. The baud rate determines the bandwidth required to send the signal.
Relationship between bit rate and baud rate
- Bit rate equals the baud rate times the number of bits represented by each signal unit.
- The baud rate equals the bit rate divided by the number of bits represented by each signal shift.
- Bit rate is always greater than or equal to baud rate.
π Definition β Bit Rate: The number of bits transmitted during one second. π Definition β Baud Rate: The number of signal units per second that are required to represent that bit. π Formula: Bit Rate = Baud Rate Γ Number of bits per signal element β The total data throughput depends on both how many signal changes happen per second and how many bits each change represents. π Formula: Baud Rate = Bit Rate / Number of bits per signal element β The rate of signal changes is equal to the bit rate divided by the number of bits encoded in each change.
Analogy for Bit rate & Baud rate In transportation, a band is analogous to a car, and a bit is analogous to a passenger. A car can carry one or more passengers. If 1000 cars go from one point to another carrying only one passenger (only driver), then 1000 passengers are transported. However, if each car carries four passengers, then 4000 passengers are transported. The number of cars, not the number of passengers, determines the traffic and therefore the need for wider highway. Similarly, the baud determines the required bandwidth, not the bit rate.
π Example 5.6: An analog signal carries 4 bits in each signal element. If 1000 signal elements are sent per second, find the Baud Rate and Bit Rate? Solution:
- Baud Rate = Number of Signal Elements = 1000 bauds/second
- Bit Rate = Baud Rate Γ Number of bits per signal element = 1000 Γ 4 = 4000 bps
Carrier Signals In analog transmission, the sending device produces a high frequency signal that acts as a basis for the information signal. This base signal is called the Carrier Signal or Carrier Frequency. The receiving device is tuned to the frequency of the carrier signal that it expects from the sender. Digital information is then modulated on the carrier signal by modifying one or more of its characteristics (Amplitude, Frequency, Phase). This kind of modification is called Modulation, and the information signal is called a Modulating Signal.
π Definition β Carrier Signal: A high-frequency signal produced by the sending device that acts as a basis for the information signal in analog transmission.
Amplitude Shift Keying (ASK)
In ASK, the strength of the carrier signal is varied to represent binary 1 or 0. Both frequency and phase remain constant, while the amplitude changes. Which voltage represents 1 and which represents 0 can be chosen by the system designer. A bit duration is the period of time that defines one bit. The peak amplitude of the signal during each bit duration is constant, and its value depends on the bit (1 or 0). The speed of transmission during ASK is limited by the physical characteristics of the transmission medium.
Effect Of Noise on ASK ASK is highly susceptible to noise interference. Noise is defined as unintentional voltages introduced onto a line by various sources such as heat or electromagnetic radiation from other sources. These unintentional voltages combine with the signal to change the amplitude. A 1 can be changed to 0 and a 0 to a 1. ASK relies solely on amplitude for recognition. Since noise usually affects the amplitude, ASK is the modulating method that is most affected by noise.
π Definition β Noise: Unintentional voltages introduced onto a line by various sources such as heat or electromagnetic radiation from other sources.
On-Off Keying (OOK) A popular ASK technique is On-Off Keying (OOK). In OOK, one of the bit values is represented by no voltage. The advantage is the reduction in the amount of energy required to transmit information.
Bandwidth for ASK
The bandwidth of a signal is the total range of frequencies occupied by that signal. When we decompose an ASK modulated signal, we get a spectrum of many simple frequencies. The most significant ones are those between fc - Nbaud/2 and fc + Nbaud/2 with carrier frequency fc at the middle.
The bandwidth requirements for ASK are calculated using the formula: BW = (1 + d) Γ Nbaud Where:
- BW = Bandwidth
- Nbaud = Baud Rate
- d = factor related to condition of line (minimum value = 0)
π‘ Why this matters: Understanding the bandwidth formula is critical because it directly determines how much frequency spectrum is needed to transmit a signal, which is a limited and often expensive resource.
π Example 5.8: Find the minimum bandwidth required for an ASK signal transmitted at 2000 bps. Transmission mode is half-duplex. Solution:
- In ASK, Baud Rate = Bit Rate
- Therefore, Baud Rate = 2000
- Also, ASK requires a minimum bandwidth equal to its Baud Rate
- Therefore, Minimum BW = 2000 Hz
β Key Takeaways
Digital-to-analog conversion is essential for transmitting digital data over analog infrastructure like telephone lines, and it works by modulating a carrier signal using one of its three characteristics: amplitude, frequency, or phase. The distinction between bit rate and baud rate is fundamental: bit rate is total bits per second, while baud rate is the number of signal changes per second, and the relationship is governed by how many bits each signal unit carries (Bit Rate = Baud Rate Γ bits per signal element). Among the modulation methods, Amplitude Shift Keying (ASK) varies amplitude to represent bits, but it is highly susceptible to noise because noise directly alters amplitude. A common variant of ASK is On-Off Keying (OOK), where one bit value is represented by no voltage, saving energy. Finally, the bandwidth required for ASK is calculated using the formula BW = (1+d) Γ Nbaud, and in its simplest form, the minimum bandwidth equals the baud rate.
π§ Quick Revision Questions
- What are the three characteristics of a sine wave that can be modified for digital-to-analog conversion?
- Explain the difference between bit rate and baud rate, and provide the formula that relates them.
- Why is ASK considered the modulation method most affected by noise?
- What is the primary advantage of using On-Off Keying (OOK) over standard ASK?
- Using the bandwidth formula for ASK, what is the minimum bandwidth required to transmit a signal at 5000 bps in the simplest case (d=0)?