Question 1 of 5: Fourier Series of a Rectangular Pulse Train — Harmonic Amplitude, Power and Duty-Cycle Optimisation
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Notes on this paper
Paper format. Engineers Canada national examination
16-Elec-A3 — Signals and Communications, May 2018. Closed book (one approved Casio or
Sharp calculator permitted), 3 hours, 4 pages. Five questions, all compulsory, all of equal value
(20 marks each). All five are solved in full below, every sub-part answered.
Reference texts. B. P. Lathi & Z. Ding, Modern Digital and Analog Communication
Systems, 5th ed. (AM modulation and detection, PCM quantisation, frequency conversion);
B. P. Lathi, Linear Systems and Signals, 2nd ed. (Fourier series of pulse trains, signal power);
A. V. Oppenheim & A. S. Willsky, Signals and Systems, 2nd ed. (discrete-time LTI systems,
convolution, DTFT); J. G. Proakis & D. G. Manolakis, Digital Signal Processing, 4th ed.
(difference equations, frequency response); S. Haykin, Communication Systems, 5th ed.
(envelope and coherent detection, mixers and image products).
Check: a note on the Question 3 figure. The plotted message and the printed equation agree: the plotted waveform has period 1 in normalised time with
peaks of $+1.5$ at $t = 0, \pm 1, \pm 2$ and troughs of $-1.5$ midway between them, which is exactly
$m(t) = \cos(2\pi f_m t) + \tfrac{1}{2}\cos(6\pi f_m t)$ evaluated with $f_m t$ as the abscissa. The
solution below therefore uses the printed equation, and the peak value $|m|_{\max} = 1.5$ read from it.
The paper's "$f_m = 4$ Kz" is read as 4 kHz.
Question 1: Fourier Series of a Rectangular Pulse Train — Harmonic Amplitude, Power and Duty-Cycle Optimisation (20 marks)
unipolar rectangular, $x = A$ on the pulse, 0 elsewhere
Find. The third-harmonic term of the Fourier series at $d = 0.25$; the average power of the
whole pulse train; the duty cycle that maximises the third-harmonic amplitude; and the average power carried
by the third harmonic at that optimum.
Figure Q1.1 — the periodic rectangular pulse train x(t): amplitude A, period T, on-time dT with duty cycle d = 25 %.
Approach. Expand the pulse train as a trigonometric Fourier series — whose
coefficients are the familiar $\operatorname{sinc}$ samples of the single-pulse spectrum — read off the
$n = 3$ term, obtain the total power directly from the definition of mean square, then maximise the
third-harmonic coefficient analytically over $d$ and convert that amplitude into power.
Write the Fourier series of the pulse train. Take the pulse centred on $t = 0$ so the
waveform is even and only cosines appear. With $\tau = dT$ the exponential coefficients are
$$c_n = \frac{1}{T}\int_{-\tau/2}^{\tau/2} A\,e^{-j2\pi n t/T}\,dt
= \frac{A\tau}{T}\cdot\frac{\sin(\pi n \tau/T)}{\pi n \tau/T}
= A\,d\;\operatorname{sinc}(nd),$$
where $\operatorname{sinc}(x) \equiv \sin(\pi x)/(\pi x)$. Because $c_{-n} = c_n$ here, the one-sided
(trigonometric) series is
$$x(t) = A d + \sum_{n=1}^{\infty} a_n \cos\!\left(\frac{2\pi n t}{T}\right),
\qquad a_n = 2 A d \operatorname{sinc}(nd).$$
The DC term $Ad$ is simply the time average of the waveform, as it must be.
Extract the third harmonic and evaluate it at $d = 0.25$. Setting $n = 3$,
$$a_3 = 2 A d \operatorname{sinc}(3d) = \frac{2A\sin(3\pi d)}{3\pi},$$
which is the general expression asked for "in terms of the signal parameters". Substituting $d = 0.25$ gives
$\sin(0.75\pi) = 0.70711$ and $3\pi = 9.42478$, so $a_3 = 2A(0.70711)/9.42478 = 0.15005\,A$. The third
harmonic therefore sits at $3/T$ hertz with amplitude
$$\boxed{\,x_3(t) = 0.1501\,A\,\cos\!\left(\frac{6\pi t}{T}\right)\ \text{V}\,}$$
(with the pulse centred on the origin; if the pulse instead starts at $t=0$ the same amplitude appears with
a phase lag of $3\pi d = 135^\circ$, which does not change any of the numbers below).
Compute the average power of the whole pulse train. Power is the mean square, and the
waveform is $A$ for a fraction $d$ of every period and zero for the rest, so the integral collapses to a
single rectangle:
$$P_x = \frac{1}{T}\int_{0}^{T} x^2(t)\,dt = \frac{1}{T}\left(A^2 \cdot dT\right) = A^2 d.$$
With $d = 0.25$,
$$\boxed{\,P_x = 0.25\,A^{2}\ \text{W (into 1 }\Omega)\,}$$
As a check, Parseval's theorem gives the same figure from the spectrum:
$P_x = c_0^2 + 2\sum_{n\ge1}|c_n|^2 = A^2d^2\left[1 + 2\sum \operatorname{sinc}^2(nd)\right]$, which
converges to $A^2 d$ — a useful reminder that the sinc-squared series carries all the power the time
domain says it must.
Maximise the third-harmonic amplitude over duty cycle. From step 2 the amplitude
depends on $d$ only through $\sin(3\pi d)$:
$$|a_3(d)| = \frac{2A}{3\pi}\,\bigl|\sin(3\pi d)\bigr| .$$
Note that the $d$ multiplying the sinc has cancelled against the $d$ in its denominator, so this is a pure
sine in $d$ — the maximum is reached whenever $|\sin(3\pi d)| = 1$, i.e. $3\pi d = \pi/2 + k\pi$:
$$d = \tfrac{1}{6},\ \tfrac{1}{2},\ \tfrac{5}{6}.$$
All three give the same magnitude,
$$\boxed{\,d = \tfrac{1}{6} = 16.67\%\ \Rightarrow\ |a_3|_{\max} = \frac{2A}{3\pi} = 0.2122\,A\,}$$
The smallest of the three, $d = 1/6$, is the natural design answer because it is the one that also keeps the
pulse narrow (the $d = 1/2$ square wave and the $d = 5/6$ nearly-DC waveform give the same third-harmonic
amplitude but very different DC and low-order content). Compared with the given 25% duty cycle the
improvement is $0.2122/0.1501 = 1.414$, i.e. 3.01 dB.
Convert that amplitude into average power. A real sinusoid of amplitude $C$ carries
mean-square value $C^2/2$, so the power in the third harmonic at the optimum duty cycle is
$$P_3 = \frac{|a_3|_{\max}^2}{2} = \frac{1}{2}\left(\frac{2A}{3\pi}\right)^{2} = \frac{2A^{2}}{9\pi^{2}}.$$
Evaluating, $9\pi^2 = 88.826$, hence
$$\boxed{\,P_3 = 0.02252\,A^{2}\ \text{W}\,}$$
At $d = 1/6$ the total power is $A^2/6 = 0.1667A^2$, so the third harmonic carries 13.5% of the signal's
power — a large share for a third harmonic, which is precisely why $d = 1/6$ is the classic choice when
a pulse train is used to drive a tripler.
Figure Q1.2 — magnitude of the third-harmonic coefficient |a3| = (2A/3π)|sin 3πd| against duty cycle d. The given d = 0.25 gives 0.1501 A; the maxima 0.2122 A occur at d = 1/6, 1/2 and 5/6.