Skip to main content

SSB vs full carrier AM; ring modulator

Short answer

SSB transmits only one sideband with carrier suppressed, using half the bandwidth of conventional AM and saving 83% power. A ring modulator switches the message with a carrier square wave via diodes, producing DSB-SC which is filtered to retain the component at the carrier frequency.

How is SSB different from conventional full carrier AM? Describe how ring modulator can be used to generate DSB-SC.

SSB versus conventional AM

SSB-SC transmits only one sideband with the carrier suppressed, while conventional AM (DSB-FC) sends the carrier and both sidebands.

PointDSB-FC AMSSB-SC
Spectrum sentCarrier + USB + LSBOne sideband only
Bandwidth2W2WWW
Power (μ=1\mu=1)1.5Pc1.5P_c0.25Pc0.25P_c (83.3% saving)
Efficiency≤ 33.3%100%
DemodulationEnvelope detectorCoherent detector needed
ReceiverSimple, cheapComplex (carrier recovery)
Noise/fadingMore affectedLess selective fading
EquationAc[1+μm(t)]cos⁡ωctA_c[1+\mu m(t)]\cos\omega_ctAc2[mcos⁡ωct∓m^sin⁡ωct]\frac{A_c}{2}[m\cos\omega_ct \mp \hat m\sin\omega_ct]

Ring modulator for DSB-SC

A ring modulator uses four diodes connected in a ring between two centre-tapped transformers. The carrier is applied between the centre taps and acts as a switch; the message enters at the input transformer.

        T1      D1       T2
 m(t) ))|(( a----|>|----b ))|(( s(t)
        |(( \           / ))|
        |((  D4       D2  ))|
        |(( /           \ ))|
        )|( d----|<|----c ))|
          |      D3       |
          +----( c(t) )---+
       (carrier between centre taps)

Operation

  1. The carrier amplitude is much larger than m(t)m(t), so the carrier alone decides which diodes conduct.
  2. Positive carrier half cycle: D1 and D3 conduct, D2 and D4 are off. The message passes to the output with its original polarity.
  3. Negative carrier half cycle: D2 and D4 conduct, D1 and D3 are off. The connections are crossed, so the message reaches the output inverted.
  4. So the output is m(t)m(t) multiplied by a bipolar square wave c(t)c(t) of frequency fcf_c:
c(t)=4π∑n=1∞(−1)n+12n−1cos⁡[2πfc(2n−1)t]c(t) = \frac{4}{\pi}\sum_{n=1}^{\infty}\frac{(-1)^{n+1}}{2n-1}\cos[2\pi f_c(2n-1)t] v(t)=m(t)c(t)=4πm(t)cos⁡ωct−43πm(t)cos⁡3ωct+…v(t) = m(t)c(t) = \frac{4}{\pi}m(t)\cos\omega_ct - \frac{4}{3\pi}m(t)\cos 3\omega_ct + \dots
  1. The square wave has no DC term, so no carrier and no baseband m(t)m(t) appear at the output. A band-pass filter centred at fcf_c (bandwidth 2W2W) keeps only
s(t)=4πm(t)cos⁡ωcts(t) = \frac{4}{\pi}m(t)\cos\omega_ct

which is DSB-SC. The circuit is balanced, so the carrier leakage is very small when the diodes are matched. The condition fc>Wf_c > W (in practice fc≫Wf_c \gg W) keeps the bands around fcf_c and 3fc3f_c separate.

✓ Answer checked by Claude Opus 5.5 (high effort)

All Communication Systems chapter 2 questions →