NivaarExam PrepOfficial exam papers ↗

22-Elec-A3 Signals and Communications · December 2016

Question 5 of 6: Spectra of DSB and SSB Modulation, Coherent Detection and Frequency Translation

Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)

Notes on this paper

Paper format. Professional Engineers Ontario / Engineers Canada national examination 07-Elec-A3 — Signals and Communications, December 2016. Three hours, closed book; a standard non-programmable calculator is the only permitted aid. Six questions of equal value (20 marks each); any five constitute a complete paper. All six are solved below, because this set is a study resource rather than an examination script.

Reference texts. B. P. Lathi and Z. Ding, Modern Digital and Analog Communication Systems, 4th ed. (Ch. 2–4 Fourier analysis, Ch. 4 amplitude modulation, Ch. 5 angle modulation, Ch. 6 sampling and pulse-code modulation); A. V. Oppenheim and A. S. Willsky, Signals and Systems, 2nd ed. (Ch. 3 Fourier series, Ch. 10 the z-transform); J. G. Proakis and D. G. Manolakis, Digital Signal Processing, 4th ed. (Ch. 3 z-transform, Ch. 7 realisation structures); S. Haykin and M. Moher, Communication Systems, 5th ed. (Ch. 3 amplitude modulation, Ch. 4 angle modulation, Ch. 7 digital transmission).

Notation used throughout. The normalised sinc function is $\operatorname{sinc}(x)=\dfrac{\sin \pi x}{\pi x}$, so that $\operatorname{sinc}(0)=1$ and the zeros sit at the non-zero integers. Average power of a periodic signal is $P=\frac{1}{T_0}\int_{T_0}|x(t)|^{2}\,dt$, and the exponential Fourier series is $x(t)=\sum_{n=-\infty}^{\infty} c_n e^{j n \omega_0 t}$ with $P=\sum_n |c_n|^{2}$ (Parseval).

Question 5: Spectra of DSB and SSB Modulation, Coherent Detection and Frequency Translation (20 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

Given. $m(t) = \cos^{3}(2\pi f_m t) + \operatorname{sinc}(f_m t)$, where the second term is written in the paper as $\sin(\pi f_m t)/(\pi f_m t)$. Carrier $A\cos(2\pi f_c t)$ with $f_c = 20 f_m$.

Find. The message spectrum and bandwidth; the DSB and lower-sideband SSB spectra and their bandwidths; a coherent DSB demodulator; and a frequency translator that moves the DSB signal to a carrier of $25 f_m$.

Approach. Expand the cubed cosine with the triple-angle identity to expose its discrete lines, transform the sinc term to a rectangle, take the larger of the two as the message bandwidth, and then apply the modulation and frequency-shifting properties of the Fourier transform.

  1. Expand the cubed cosine. Using $\cos^{3}\theta = \tfrac34\cos\theta + \tfrac14\cos 3\theta$, $$\cos^{3}(2\pi f_m t) = \tfrac34\cos(2\pi f_m t) + \tfrac14\cos(2\pi\,3f_m t),$$ which contributes impulses of weight $\tfrac38$ at $f = \pm f_m$ and $\tfrac18$ at $f = \pm 3f_m$. (A numerical projection of $\cos^{3}$ onto the two cosines returns exactly $0.75$ and $0.25$, confirming the identity.)
  2. Transform the sinc term. With $\operatorname{sinc}(f_m t) = \dfrac{\sin \pi f_m t}{\pi f_m t}$, the standard pair gives $$\operatorname{sinc}(f_m t) \;\longleftrightarrow\; \frac{1}{f_m}\,\Pi\!\left(\frac{f}{f_m}\right),$$ a rectangle of height $1/f_m$ extending over $|f| \lt f_m/2$. This term is therefore band-limited to $f_m/2$ — half the fundamental, and well inside the discrete lines.
  3. Message spectrum and bandwidth (part a). The complete spectrum is the flat rectangle on $|f| \lt f_m/2$ plus four impulses at $\pm f_m$ and $\pm 3f_m$. The highest frequency present is the third harmonic, so $$\boxed{B_m = 3f_m}$$ The sinc term does not set the bandwidth here; the cubed cosine does.
f / fm|M(f)|-3-1-0.50.5130.1250.3750.3750.125flat sinc term
Figure Q5.1 — message spectrum |M(f)|: impulses of weight 3/8 at ±fₘ and 1/8 at ±3fₘ, plus the flat sinc rectangle on |f| < fₘ/2.
  1. DSB spectrum (part b). Multiplying by the carrier translates the message spectrum to $\pm f_c$ and halves it: $$s_{\text{DSB}}(t) = A\,m(t)\cos(2\pi f_c t) \;\longleftrightarrow\; \frac{A}{2}\left[M(f - f_c) + M(f + f_c)\right].$$ Each copy occupies $f_c \pm B_m$, i.e. from $17f_m$ to $23f_m$ about the positive carrier, and the transmission bandwidth is $$\boxed{B_{\text{DSB}} = 2B_m = 6f_m}$$ The impulses land at $f_c \pm f_m$ ($19f_m$ and $21f_m$) and $f_c \pm 3f_m$ ($17f_m$ and $23f_m$); the flat rectangle occupies $f_c \pm f_m/2$.
f / fm|S(f)|1720 (fc)23DSB, width 6 fm
Figure Q5.2 — DSB spectrum about the positive carrier: 17fₘ to 23fₘ, bandwidth 6fₘ.
  1. Lower-sideband SSB (part c). Retaining only the sideband below the carrier deletes everything above $f_c$, leaving the band from $f_c - B_m = 17f_m$ up to $f_c = 20f_m$ (and its mirror image at negative frequencies). Hence $$\boxed{B_{\text{SSB}} = B_m = 3f_m}$$ which is exactly half the DSB requirement — the spectral-efficiency argument for SSB. Note the spectrum is now asymmetric about the carrier, which is why an envelope detector cannot demodulate it.
f / fm|S(f)|1720 (fc)20LSB, width 3 fm
Figure Q5.3 — lower-sideband SSB spectrum: 17fₘ to 20fₘ, bandwidth 3fₘ.
  1. Exact recovery from DSB (part d). Because DSB-SC carries no carrier line, the receiver must supply a local oscillator that is coherent in both frequency and phase. Multiplying the received signal by $2\cos(2\pi f_c t)$, $$2\,s_{\text{DSB}}(t)\cos(2\pi f_c t) = A\,m(t)\left[1 + \cos(2\pi\,2f_c t)\right] = A\,m(t) + A\,m(t)\cos(4\pi f_c t).$$ The first term is the message at baseband; the second is centred on $2f_c = 40f_m$. An ideal low-pass filter of cut-off $3f_m$ (anywhere between $3f_m$ and $37f_m$ would do) removes the high-frequency term and leaves $A\,m(t)$ exactly. Because the argument never used any property of $m(t)$ beyond its bandwidth, the receiver works for arbitrary messages of bandwidth $3f_m$, as the question requires.
DSB inxIdeal LPFcut-off 3 fmm(t) (scaled)Local osc.2 cos(2 pi fc t)coherent (synchronous) detector
Figure Q5.4 — coherent (synchronous) detector for DSB-SC (part d).

Check: the local oscillator is assumed phase-locked to the transmitter. A constant phase error $\phi$ scales the output by $\cos\phi$ (total loss at $\phi = 90^{\circ}$), and a frequency error produces a beat — which is why practical DSB receivers add a Costas loop or a pilot tone. That recovery machinery is outside what this question asks for.

  1. Frequency translation to $25 f_m$ (part e). Mixing the DSB signal with a local oscillator at $f_{\text{LO}}$ produces sum and difference bands centred at $f_c + f_{\text{LO}}$ and $f_c - f_{\text{LO}}$. Choosing $$f_{\text{LO}} = 25f_m - f_c = 25f_m - 20f_m = \boxed{5f_m}$$ places the wanted band at $25f_m$ (spanning $22f_m$ to $28f_m$) and the unwanted image at $15f_m$ (spanning $12f_m$ to $18f_m$). The two are separated by $4f_m$ of clear spectrum, so an ordinary band-pass filter centred at $25f_m$ with a passband $6f_m$ wide isolates the wanted output. The result is a genuine DSB signal on a $25f_m$ carrier, since mixing preserves the two-sideband structure.
DSB at fcxBand-pass filtercentre 25 fm, 6 fm wideDSB outOscillator5 fmimage at the difference frequency is rejected by the BPF
Figure Q5.5 — frequency translator moving the DSB signal from 20fₘ to 25fₘ (part e).
QuantityResult
Message lines$\pm f_m$ (weight $3/8$), $\pm 3f_m$ (weight $1/8$), plus a flat rectangle on $|f| \lt f_m/2$
(a) Message bandwidth$B_m = 3f_m$
(b) DSB band / bandwidth$17f_m$ to $23f_m$; $B = 6f_m$
(c) LSB SSB band / bandwidth$17f_m$ to $20f_m$; $B = 3f_m$
(d) DemodulatorMultiply by $2\cos(2\pi f_c t)$, low-pass at $3f_m$
(e) TranslatorMix with $f_{\text{LO}} = 5f_m$, band-pass at $25f_m$ (width $6f_m$)