📚 Chapter 5 — Transmission Line Limitations + Filtering (Asgn05 + Lab05)
🎯 Learning Objectives
By the end of this lesson, students should be able to:
- Relate pulse width, duty cycle, and baud rate to spectral content.
- Use Fourier reasoning to identify the first null and determine baud rate.
- Understand why Baud = 2·fc for a high‑order low‑pass filter.
- Identify MSI points and interpret eye patterns.
- Build and test a 3‑pole Sallen‑Key low‑pass filter.
1️⃣ Sampling and Nyquist for Digital Symbols
- The rate of transmission of electrical symbols… is equal to the Nyquist sampling rate — the data signal must be sampled at a minimum of once per electrical symbol.”
Oversampling must be an integer multiple (2×, 4×, 16×).
2️⃣ Fourier Analysis of Pulses
2.1 First Null = Baud Rate
“The first dropout is the inverse of the pulse width… the first dropout in the spectrum is the Baud Rate.”
Thus: $\mathrm{Baud}=\frac{1}{t_p}$
2.2 Duty Cycle ↔ Number of Spectral Components
“Duty cycle is equal to the inverse of the number of frequency components in a lobe.”
3️⃣Filter Characteristics and Sinc Response
3.1 Sine Cardinal (sinc) Output
High‑order filters produce a sinc‑like response:
“The shape of the output signal is a function of the filter’s characteristics… best modelled using a Sine Cardinal function.”
MSI points occur at zero crossings of the sinc.
3.2 Maximum Baud Rate Through Filter
\[\mathrm{Baud_{\mathnormal{\max }}}=2f_c\]“The maximum Baud rate through a filtered transmission system will be twice the cutoff frequency.”
4️⃣ Eye Patterns
“Since it looks like a series of open eyes, this is called an Eye Pattern… the middle of each diamond is a point of MSI.”
Students should learn:
- Wide open eye → good timing margin
- Narrow eye → intersymbol interference
- MSI sampling → clean symbol decisions
5️⃣ Lab 05 — Filtering and Spectral Analysis
5.1 Part A — Generate Bipolar Pulse
- 2 kHz
- 50 µs pulse width
- ±5 V
- FFT to find first null → Baud rate
5.2 Part B — Change Duty Cycle
Observe that:
- 10% and 90% duty cycles produce identical spectra (symmetry).
- 25% ↔ 75% also match.
5.3 Part C — Build 3‑Pole Sallen‑Key Filter
Measure:
- Actual cutoff frequency
- Pass‑band gain
- MSI spacing → compute fc
5.4 Part D — Eye Pattern
Feed UART data through the filter and observe:
- Eye opening
- MSI points
- Timing recovery
6️⃣ Assignment 5 Prep Questions
Should be able to compute:
- Minimum sampling rate for a given baud
- Oversampling rate
- First null frequency
- Duty cycle from spectral components
- Cutoff frequency from MSI spacing
- Maximum baud rate through a filter All directly supported by: “The inverse of the time between MSI points is twice the cutoff frequency.” “Baud = 2fc.”