Determining the Sidebands of an FM Signal

using Bessel Functions

By

Susie E. Maestre

Presented to:
Prof. Michael S. Villame
Mindanao State University – Iligan Institute of Technology

As a Final Requirement for the Course

ES202 – Advanced Engineering Mathematics II

20 December 2019
Basic Principles of Frequency Modulation

In FM, the carrier amplitude remains constant and the carrier frequency is changed
by the modulating signal. As the amplitude of the information signal varies, the carrier
frequency shifts proportionately. As the modulating signal amplitude increases, the car-
rier frequency increases. If the amplitude of the modulating signal decreases, the carrier
frequency decreases. The reverse relationship can also be implemented. A decreasing mod-
ulating signal increases the carrier frequency above its center value, whereas an increasing
modulating signal decreases the carrier frequency below its center value. As the modulating
signal amplitude varies, the carrier frequency varies above and below its normal center, or
resting, frequency with no modulation. The amount of change in carrier frequency produced
by the modulating signal is known as the frequency deviation fd. Maximum frequency de-
viation occurs at the maximum amplitude of the modulating signal. The instantaneous
voltage of the FM wave (νF M ) is given by

νF M = Vc sin(ωc t + mf sin ωm t) (1)

where Vc is the peak amplitude of the carrier, ωc is the angular frequency of the carrier,
ωm is the angular frequency of the modulating signal and mf is the modulation index. The
product ωc t is the carrier phase while ωm t is the modulation phase in radians.

Figure 1: Visualization of frequency modulation

FM and Bessel Functions

Mindanao State University–Iligan Institute of Technology
1
Modulation Index and Sidebands

Any modulation process produces sidebands. When a constant-frequency sine wave

modulates a carrier, two side frequencies are produced. The side frequencies are the sum
and difference of the carrier and the modulating frequency. The sidebands are spaced from
the carrier fc and from one another by a frequency equal to the modulating frequency fm . If
the modulating frequency is 1 kHz, the first pair of sidebands is above and below the carrier
by 1000 Hz. The second pair of sidebands is above and below the carrier by 2 x 1000 Hz =
2000 Hz, or 2 kHz, and so on. Note also that the amplitudes of the sidebands vary. If each
sideband is assumed to be a cosine wave, with a frequency and an amplitude as indicated
in Figure 2, and all the cosine waves are added, then the FM signal producing them will be
created.

Figure 2: Frequency spectrum of an FM signal. Note that the carrier sidebands amplitudes shown
are just examples. The amplitudes depend upon the modulation index mf .

Bessel Functions and FM

The FM wave is expressed as a composite of cosine waves of different frequencies and

amplitudes that, when added, give an FM time-domain signal. The first term is the carrier
with an amplitude given by a Jn coefficient, in this case J0 . The next term represents a
pair of upper and lower side frequencies equal to the sum and difference of the carrier and
modulating signal frequency. The amplitude of these side frequencies is J1 . The other terms
represent additional side frequencies spaced from one another by an amount equal to the
modulating signal frequency.
Consider a cosine carrier wave with frequency fc . Adding a signal with amplitude β
and frequency fm results in the combination

x(t) = A cos (2πfc t + β sin 2πfm t) (2)

This resulting wave can be expressed as a set of cosines weighed by Bessel functions of β.

FM and Bessel Functions

Mindanao State University–Iligan Institute of Technology
2
Proof.

The sum identity for cosines is cos (a + b) = cos a cos b − sin a sin b. Applying this to (2),

cos(2πfc t + β sin 2πfm t) = cos(2πfc t) cos(β sin(2πfm t))

− sin(2πfc t) sin(β sin(2πfm t)) (3)

Two Bessel function identities are

∞
X
cos(z sin θ) = J0 (z) + 2 J2k (z) cos(2kθ) (4)
k=1

∞
X
sin(z sin θ) = 2 J2k+1 (z) sin((2k + 1)θ) (5)
k=0

Applying eq(4) to expand the first term on the right-hand side of eq(3) where z = β and
θ = 2πfm t,

cos(2πfc t) cos(β sin(2πfm t)) = ♥

 ∞
X 
= cos(2πfc t) J0 (β) + 2 J2k (β) cos(2k(2πfm t))
k=1
= J0 (β) cos(2πfc t)
X∞
+2 J2k (β) cos(2πfc t) cos(2k(2πfm t)) (6)
k=1

Applying the identity 2 cos a cos b = cos(a − b) + cos(a + b) to the product inside the sum-
mation,

1 
cos(2πfc t) cos(2k(2πfm t)) = cos(2πfc t − 2k · 2πfm t) + cos(2πfc t + 2k · 2πfm t)
2
1 
= cos(2π(fc − 2kfm )t) + cos(2π(fc + 2kfm )t) (7)
2

Therefore,
∞
X  
♥ = J0 (β) cos(2πfc t) + J2k (β) cos(2π(fc − 2kfm )t) + cos(2π(fc + 2kfm )t)
k=1
 
= J0 (β) cos(2πfc t) + J∓2 (β) cos(2π(fc − 2fm )t) + cos(2π(fc + 2fm )t)
 
+ J∓4 (β) cos(2π(fc − 4fm )t) + cos(2π(fc + 4fm )t)
 
+ J∓6 (β) cos(2π(fc − 6fm )t) + cos(2π(fc + 6fm )t) + ...
X
= Jn (β) cos(2π(fc + nfm )t) (8)
n even

where the sum runs over positive and negative even integers and J−n (z) = (−1)n Jn (z)
(Bessel function identity).

FM and Bessel Functions

Mindanao State University–Iligan Institute of Technology
3
Now applying eq(5) to expand the second term on the right-hand side of eq(3) where z = β
and θ = 2πfm t,
♥
sin(2πfc t) sin(β sin(2πfm t)) =
 X∞ 
= sin(2πfc t) 2 J2k+1 (β) sin((2k + 1)2πfm t)
k=0
∞
X
=2 J2k+1 (β) sin(2πfc t) sin((2k + 1)2πfm t) (9)
k=0

Applying the identity 2 sin a sin b = cos(a − b) − cos(a + b) to the product inside the sum-
mation,
1
sin(2πfc t) sin((2k + 1)2πfm t) = cos(2πfc t − (2k + 1)(2πfm t))
2 
− cos(2πfc t + (2k + 1)(2πfm t))
1
= cos(2π(fc − (2k + 1)fm )t)
2 
− cos(2π(fc + (2k + 1)fm )t) (10)

Therefore,
∞
X
♥  
= J2k+1 (β) cos(2π(fc − (2k + 1)fm )t) − cos(2π(fc + (2k + 1)fm )t)
k=0
 
= J∓1 (β) cos(2π(fc − fm )t) − cos(2π(fc + fm )t)
 
+ J∓3 (β) cos(2π(fc − 3fm )t) − cos(2π(fc + 3fm )t)
 
+ J∓5 (β) cos(2π(fc − 5fm )t) − cos(2π(fc + 5fm )t) + ...
X
=− Jn (β) cos(2π(fc + nfm )t) (11)
n odd

where the sum runs over positive and negative odd integers and again, J−n (z) = (−1)n Jn (z).
♥
Substituting the expressions for ♥ and to eq(3),
♥
cos(2πfc t + β sin 2πfm t) = ♥ −
X X
= Jn cos(2π(fc + nfm )t) + Jn (β) cos(2π(fc + nfm )t)
n even n odd
X∞
= Jk (β) cos(2π(fc + kfm )t)
k=−∞
(12)

It is to be noted that the product J0 Vc is then the amplitude of the carrier frequency
and Jn is the amplitude of the nth sideband where n ∈ Z+ .

FM and Bessel Functions

Mindanao State University–Iligan Institute of Technology
4
Table 1: Tabulated values of sideband amplitudes for different modulation indices of FM signals
based on Bessel functions (Jn ); β ≡ mf ≡ modulation index

Sidebands
β Carrier J0
J1 J2 J3 J4 J5 J6 J7 J8 J9 J10
0.0 1.0 – – – – – – – – – –
0.25 0.98 0.12 – – – – – – – – –
0.5 0.94 0.24 0.03 – – – – – – – –
1.0 0.77 0.44 0.11 0.02 – – – – – – –
1.5 0.51 0.56 0.23 0.06 0.01 – – – – – –
2.0 0.22 0.58 0.35 0.13 0.03 – – – – – –
2.5 -0.05 0.50 0.45 0.22 0.07 0.02 – – – – –
3.0 -0.26 0.34 0.49 0.31 0.13 0.04 0.01 – – – –
4.0 -0.40 -0.07 0.36 0.43 0.28 0.13 0.05 0.02 – – –
5.0 -0.18 -0.33 0.05 0.36 0.39 0.26 0.13 0.06 0.02 – –
6.0 0.15 -0.28 -0.24 0.11 0.36 0.36 0.25 0.13 0.06 0.02 –
7.0 0.30 0.00 -0.30 -0.17 0.16 0.35 0.34 0.23 0.13 0.06 0.02
8.0 0.17 0.23 -0.11 -0.29 0.10 0.19 0.34 0.32 0.22 0.13 0.06

Figure 3: Plot of the Bessel function data from Table 1

FM and Bessel Functions

Mindanao State University–Iligan Institute of Technology
5
Figure 4: Examples of FM signal spectra: (a) β = 0 (no modulation or sidebands), (b) β = 1, (c)
β = 2 and (d) β = 0.25

References

Frenzel, Louis E., Jr. 2016. Principles of Electronic Communication Systems 4th Ed.
New York, NY: McGraw-Hill Education.

Cook, John D. (2016, February 17). Analyzing an FM Signal.

Retrieved from https://www.johndcook.com/blog/2016/02/17/analyzing-an-fm-signal/

FM and Bessel Functions

Mindanao State University–Iligan Institute of Technology
6