F Pass6

4.
M-ary Modulation Techniques and OFDM

A. M-ary PSK Scheme: The phase of carrier takes on one of M possible values, namely,
= ( 2 i 1) / M ,
i = 1 ,..., M
An M-ary signal set is represented as

si (t ) = 2E T cos(2f ct + ( 2i 1) ), i = 1,...,M , 0 t T M
where T is the symbol duration and E is the signal energy per symbol. The carrier frequency where n c is a fixed integer: f c = n c / T
@G. Gong 1
Similar as we did for QPSK, each signal si(t) can be represented by the following two orthogonal functions with unit energy:
1 (t ) = 2 2 cos(2f ct ) and 2 (t ) = sin(2f ct ) T T
Thus, the signal constellation of M-ary PSK is two-dimensional. The M messages are equally spaced one circle of radius and center at the origin, see Figure 1 for an example of octa-PSK.
Figure 1. Signal Constellation for octaPSK (M = 8). The decision boundaries are shown as dashed lines.
@G. Gong
The coordinates of the received signal given si(t) was transmitted is

(2i 1) xI = E cos + nI , M (2i 1) xQ = E cos + nQ , i = 1,..., M M
where nI and nQ are Gaussian random variables with zero mean and variance N0/2 (why ?). Remark. The probability of correct reception is to integrate the shaded area. This probability can be bounded by some bound. Therefore, for large values of E/N0, the probability of symbol error is approximately given by
2E Pe,M PSK 2Q sin , (A) N M 0 M 4
@G. Gong 3
/ M si / M
B. M-ary FSK
In an M-ary FSK scheme, the transmitted signals are defined by
2E si (t ) = cos 2 [ f 0 + (i 1)f ]t , T
0 t T , i = 1,.., M
(*)
where f 0T is taken as an integer for convenience and ( f ) min = 1 /( 2T ) is the minimum frequency spacing such that adjacent signals are orthogonal (recall this result from MSK). For coherent M-ary FSK, the optimum receiver consists of a bank of M correlations or matched filters. At the sampling times t = kT, the receiver makes decisions based on the largest matched filter output. The probability of symbol error can be upper bounded by
E Pe,M FSK ( M 1)Q N 0
(B)
where E = Eb (log 2 M ) the symbol set.

@G. Gong
is the energy per symbol and M is the size of

4
C. Multicarrier Modulation and OFDM

Applications of M-ary FSK: multicarrier modulation and OFDM. Multicarrier modulation is a way to transmit digital data through bandlimited channel. For doing so, one can subdivide the available channel bandwidth into a number of equal-bandwidth subchannels, where the bandwidth of the subchannels are nearly ideal. |C(f)|
f Figure 1. Subdivision of the channel bandwidth W into narrow band subchannels of equal width f.
@G. Gong 5
Orthogonality: Each subchannel is associated a carrier given in (*), where f i = f 0 + (i 1) f , i = 1,..., K which is the mid-frequency in the ith subchannel. By selecting the symbol rate 1/Ts (recall the proof for minimum spacing for MSK) on each of the subchannels to be equal to the separation f of adjacent subcarriers, the subcarriers are orthogonal over the symbol interval Ts, independent of the relative phase relationship between subcarrier. In this case, we call orthogonal frequency-division multiplexing (OFDM). ISI Reduction: The symbol interval in the OFDM system is T = KTs where Ts is the symbol interval in the single-carrier system. Thus by selecting K to be sufficiently large, the symbol interval T can be make significantly larger than the time duration of the channel-time dispersion. Thus, ISI can be made arbitrarily small by selection of K. Implementation: by use of a parallel bank of filters based on the discrete Fourier transform. Disadvantage: high peak-to-average power (PAR) ratio.
@G. Gong 6
5. Comparison of Digital Modulation Systems

A. Bit Error Probabilities from Symbol Error Probabilities There are two approaches to define an equivalent bit error probability, Pb, or bit error rate (BER), from a symbol error probability, Ps. It depends on (1) structure of the signal space, and (2) the mapping of the signal space points into equivalent bit sequences.
@G. Gong
Definition 1. We assume that in going from one signal point to an adjacent signal point, only one bit in the binary word representing the signal changes. In this case,
Pb = Ps log
2
Remark. M-ary PSK, if a Gray code is employed and Mary QAM are of the case.
@G. Gong
Definition 2. Denote n = log 2 M. We assume that all symbol errors are equally likely. We define P b as the ratio of A, the average number of bit errors per n-bit symbols to n, number of bits per symbol. In the following, we will give an explicit formula for Pb. Notice that - Each symbol is in error in an M-ary system with probability
Ps ( M 1)
- For a given symbol error, suppose that k bits are in error. There are
Pb ways that this can happen, which results n A 1 n Ps M = = k 1 k ( M 1 ) = 2 ( M 1 ) P s P s / 2 n n k= n k
for large M. Remark. M-ary FSK is of this case.
B. Bandwidth Efficiencies of M-ary Digital Comm. Systems (DCS)

Goal: Consider the bandwidth efficiency in terms of bits per second per hertz (bps/Hz) of bandwidth of various digital modulation schemes, i.e.,
= Rb BM
,X
where
M ,X
represents bandwidth of modulation schemes X. (1) For a M-ary PSK, QAM, DPSK, the null to null bandwidth is
B
M ,X
For a M-ary DCS, let Rb denote the bit rate and Rs symbol rate. Then
R b = (log M )R
2Rb log 2 M
R B
M ,X 2
= 0 . 5 (log
@G. Gong
M ) (bps/Hz)
10
(2) For a M-ary FSK, consider the spacing between frequency is minimum. Then the bandwidth is
B coh
=
, MFSK
=
=
(M + 3)Rb 2 log 2 M
2 log 2 M M + 3 (bps/Hz)
Rb B
coh , MFSK
Table 1. Bandwidth Efficiencies of M-ary Signals

M : PSK DPSK QAM : FSK 2 0.5 1 4 1 1 8 1.5 .75 16 2 .5 32 2.5 .3125 64 3 .1875
@G. Gong
11
Remark: (1) M-ary PSK and M-ary QAM have 2-dimensional signal space and they are both bandwidth efficient (or called spectral efficient). (2) MFSK has M-dimensional signal space and it is bandwidth inefficient. Note. The other parameter used in comparing performance (power efficiencies) of different schemes is E/N0, the ratio of symbol energy to noise power spectral density. In other words, it is to make comparisons between different DSCs on the basis of the relative signal power needed to support a given received information rate assuming identical noise environment.
@G. Gong 12
Shannons system capacity C of an AWGN channel: C = W log2(1 + P/WN0) bits/s 16 R>C 8 4 2 -2 -1.6 -.1 1/2 1/4 6 4 2 4 8 16 12 18 M=2 16 16
R=C
= Rb /W
Bandwidth limited region R<C M-PSK M-QAM
24
30 E b
/ N
( dB )
M-FSK Power limited region
Figure 1. Band Width Efficiency Plane

@G. Gong
P e = 10
13

F Pass6

Uploaded by

Copyright:

Available Formats

F Pass6

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

F Pass6

Uploaded by

Copyright:

Available Formats

4.

M-ary Modulation Techniques and OFDM

An M-ary signal set is represented as

The coordinates of the received signal given si(t) was transmitted is

where E = Eb (log 2 M ) the symbol set.

is the energy per symbol and M is the size of

C. Multicarrier Modulation and OFDM

5. Comparison of Digital Modulation Systems

for large M. Remark. M-ary FSK is of this case.

B. Bandwidth Efficiencies of M-ary Digital Comm. Systems (DCS)

Table 1. Bandwidth Efficiencies of M-ary Signals

Bandwidth limited region R<C M-PSK M-QAM

M-FSK Power limited region

Figure 1. Band Width Efficiency Plane

You might also like