Skip to main content

Detection gain of DSB-AM is less than 1

Short answer

Detection gain is the ratio of output SNR to input SNR of a demodulator. For DSB-AM with 100% single-tone modulation, the detection gain is 2/3 ≈ 0.667, less than 1 because the carrier power contributes to input SNR but not to the output message.

What is detecting gain? Prove that for 100% modulation of (DSB-AM), the detection gain is less than 1.

Detection gain is the ratio of the output signal-to-noise ratio of a demodulator to its input (pre-detection) SNR:

γd=SNRoSNRi\gamma_d=\frac{\text{SNR}_o}{\text{SNR}_i}

It shows whether the detector improves (γd>1\gamma_d>1) or degrades (γd<1\gamma_d<1) the SNR.

Receiver model. The received signal s(t)s(t) plus white noise (N0/2N_0/2) passes through a band-pass (IF) filter of bandwidth BTB_T, then a demodulator and a low-pass filter of bandwidth WW (message bandwidth).

 s(t)  +   +-------+    +-------+   +-----+
 ---->(+)-->|  BPF  |--->| demod |-->| LPF |--> y(t)
       ^    |  B_T  |    |       |   |  W  |
      w(t)  +-------+    +-------+   +-----+

The filtered noise is narrowband and is written in in-phase/quadrature form

n(t)=nc(t)cos⁡ωct−ns(t)sin⁡ωctn(t)=n_c(t)\cos\omega_ct-n_s(t)\sin\omega_ct

where ncn_c and nsn_s are low-pass, each with the same power as n(t)n(t): nc2‾=ns2‾=n2‾=N0BT\overline{n_c^2}=\overline{n_s^2}=\overline{n^2}=N_0B_T.

DSB-FC (conventional AM). s(t)=Ac[1+kam(t)]cos⁡ωcts(t)=A_c[1+k_am(t)]\cos\omega_ct, BT=2WB_T=2W. (Coherent and envelope detection give the same result at high SNR.)

  • Signal power =Ac22(1+ka2P)=\frac{A_c^2}{2}(1+k_a^2P); noise in 2W2W =2N0W=2N_0W:
SNRi=Ac2(1+ka2P)4N0W\text{SNR}_i=\frac{A_c^2(1+k_a^2P)}{4N_0W}
  • Coherent detection (or envelope detection at high SNR) and DC blocking:
y(t)=Ac2kam(t)+12nc(t) ⇒ SNRo=Ac2ka2P/42N0W/4=Ac2ka2P2N0Wy(t)=\frac{A_c}{2}k_am(t)+\frac12n_c(t)\ \Rightarrow\ \text{SNR}_o=\frac{A_c^2k_a^2P/4}{2N_0W/4}=\frac{A_c^2k_a^2P}{2N_0W} γDSB-FC=SNRoSNRi=2ka2P1+ka2P,FOM=ka2P1+ka2P\gamma_{DSB\text{-}FC}=\frac{\text{SNR}_o}{\text{SNR}_i}=\frac{2k_a^2P}{1+k_a^2P},\qquad \text{FOM}=\frac{k_a^2P}{1+k_a^2P}

The carrier uses power but carries no information, so the gain is reduced.

Proof that γd<1\gamma_d<1. To avoid over-modulation, ∣kam(t)∣≤1\lvert k_am(t)\rvert\le1, so ka2P=ka2m2(t)‾≤1k_a^2P=\overline{k_a^2m^2(t)}\le1, with equality only for a square-wave message at 100% modulation. Hence

γd=2ka2P1+ka2P≤2(1)1+1=1\gamma_d=\frac{2k_a^2P}{1+k_a^2P}\le\frac{2(1)}{1+1}=1

For any practical message (sinusoid, speech) ka2P<1k_a^2P<1 and γd<1\gamma_d<1.

Single tone, 100% modulation. m(t)=Amcos⁡ωmtm(t)=A_m\cos\omega_mt, μ=kaAm=1\mu=k_aA_m=1, so ka2P=μ22=12k_a^2P=\frac{\mu^2}{2}=\frac12:

γDSB-FC=2(μ2/2)1+μ2/2=2μ22+μ2=23≈0.667FOM=μ22+μ2=13\begin{aligned} \gamma_{DSB\text{-}FC} &= \frac{2(\mu^2/2)}{1+\mu^2/2}=\frac{2\mu^2}{2+\mu^2}=\frac{2}{3}\approx0.667 \\ \text{FOM} &= \frac{\mu^2}{2+\mu^2}=\frac13 \end{aligned}

So for 100% single-tone DSB-AM the detection gain is 2/3<12/3<1: the detector output SNR is about 1.761.76 dB lower than the input SNR. The reason is that two-thirds of the transmitted power is in the carrier, which is counted in the input signal power but contributes nothing to the output message. (Compare DSB-SC, γd=2\gamma_d=2.)

✓ Answer checked by Claude Opus 5.5 (high effort)

All Communication Systems chapter 9 questions →