Professional Documents
Culture Documents
Part6 OFDM
Part6 OFDM
Part6 OFDM
(OFDM)
1. Why Do We Need OFDM?
• The channel is frequency selective fading channel if the bandwidth of the signal is
higher than the coherence bandwidth of the channel.
• Solution: Equalization
2. Equalization
3. Problem of Equalization
• In fast fading (changeable) environments and at high data rates (more time
dispersion), it is difficult to use fast adaptive equalization.
• The carrier center frequencies (c1, c2, c3, c4) are frequencies that do not have any
special relationship to each other (may not be orthogonal).
• Signals are moved further apart. But interference is not completely eliminated
and wasting in spectrum.
• Pulse shaping using Raised Cosine filter which is implemented by using Root
Raised Cosine (RRC) filter at Tx and Rx.
• The Main idea of OFDM is coming from the mathematical concept (Discrete-FFT and
Discrete-IFFT).
• First: the time-domain signal is treated as a frequency domain signal where the
frequency components are mapped (by serial to parallel process and then digital
modulation) into a known finite number of orthogonal frequencies.
• At Receiver side:
• First: the received time-domain signal is considered as an input to the FFT that will
regenerate the frequency components.
• Second: the frequency components are remapped (by parallel to serial process and then
digital demodulation) to generate the original signal.
• Let’s examine the following bit sequence we wish to transmit and show the
development of the ODFM signal using 4 sub-carriers. The signal has a symbol rate of 1
and sampling frequency is 1 sample per symbol, so each transition is a bit.
• First few bits are 1, 1, -1, -1, 1, 1, 1, -1, 1, -1, -1, -1, -1, 1, -1, -1, -1, 1,…
• Let’s now write these bits in rows of fours, since this demonstration will use only four
subcarriers.
• Each column represents the bits that will be carried by one sub-carrier. Let’s start with
the first carrier, c1. What should be its frequency?
• From the Nyquist sampling theorem, we know that smallest frequency that can convey
information has to be twice the information rate. In this case, the information rate per
carrier will be 1/4 or 1 symbol per second total for all 4 carriers. So the smallest
frequency that can carry a bit rate of 1/4 is 1/2 Hz. But we picked 1 Hz for convenience.
• We pick BPSK as our modulation scheme for this example. Note that we can pick any
other modulation method, QPSK, 8PSK 32-QAM or whatever. No limit here on what
modulation to use.
• In general for all integers n and m, sinmx, cosmx, cosnx, sinnx are all orthogonal to each
other (The integral of their product is zero) . These frequencies are called harmonics.
So, we use in this example 1 Hz, 2 Hz, 3 Hz, and 4 Hz.
• Carrier 1:
• We need to transmit 1, 1, 1 -1, -1, -1 which I show below superimposed on the BPSK
carrier of frequency 1 Hz. First three bits are 1 and last three -1.
• Carrier 2:
• Carrier 3:
• Carrier 3 frequency is equal to 3 Hz . The third carrier is modulated with -1, 1, 1, -1, -1, 1
• Carrier 4:
• The fourth carrier is of frequency 4 Hz. It is modulated with -1, -1, -1, -1, -1, -1, 1
• Now add all four of these modulated carriers to create the OFDM signal, often produced
by a block called the IFFT. IFFT quickly computes the time-domain signal instead of
having to do it one carrier at time and then adding.
• Delay spread is like the undesired splash you might get from the car adead of you. In
fading, the front symbol similarly throws a splash backwards which we wish to avoid.
• Solution: blank spaces between symbols. But, we cannot have blank spaces in signals.
This will not work for the hardware. So, we need something in this space.
• If we just extend the symbol, then the front of the symbol which is important to us since
it allows figuring out what the phase of this symbol is, is now corrupted by the splash.
• If we move the symbol back, then not only we have a continuous signal but can get
corrupted and we do not care since we just cut it out before demodulation.
• The whole process can be done only once to the OFDM signal, rather than doing it to
each and every sub-carrier.
• To avoid inter-symbol interference, cyclic prefix (guard interval) should be longer than
the maximum channel delay. The prefix is any where from 10% to 25% of the symbol
time.
• Disadvantage of CP:
• Answer:
• 3 symbols (each has bandwidth of 1 Rs) are sent in 4 Rs bandwidth (from -2Rs to 2Rs).
So, ¾ symbol per Hz. The large number of carrier is better. For 5 carriers, 5/6 symbol
per Hz.
3. Effective against channel distortion (that results due to selective frequency fading),
where each sub-channel is converted to flat fading channel. OFDM mitigates deep fading
effect.
(a) The signal we want to send and the channel frequency response are well matched.
(b) A fading channel has frequencies that do not allow anything to pass. Data is lost
sporadically.
(c) With OFDM, where we have many little sub-carriers, only a small sub-set of the data
is lost due to fading (with proper coding, it can be recovered) .
• The BER of an OFDM is only exemplary in a fading environment. We would not use
OFDM in a straight line of sight link such as a satellite link. OFDM signal due to its
amplitude variation does not behave well in a non-linear channel such as created by
high power amplifiers on board satellites.
• But in channels that are fading, the OFDM offers far better BER than a wide band signal
of exactly the same modulation. The advantage here is coming from the diversity of the
multi-carrier such that the fading applies only to a small subset.
7. Frequency diversity.
8. No need for filter bank at Rx. It uses discrete Fourier transform (DFT)/fast Fourier
transform (FFT) for implementation.
1. More sensitive to phase, timing and frequency offsets (due to doppler shift or hardware
of Tx and Rx), so this leads to ICI.
• Example: For an OFDM signal that has 128 carriers, each with normalized power of 1 w,
then the max PAPR can be as large as log (128) or 21 dB.
• The large amplitude variation increases in-band noise and increases the BER when the
signal has to go through amplifier non-linearities. Large back off is required in such
cases. This makes use of OFDM just as problematic as Multi-carrier FDM in high power
amplifier applications such as satellite links.
1. Clipping
• We can just clip the signal at a desired power level. This reduces the PAPR but
introduces other distortions and ICI.
2. Selective Mapping
• Multiply the data signal by a set of codes, do the IFFT on each and then pick the one with
the least PAPR. This is essentially doing the process many times using a CDMA like code.
3. Partial IFFT
• Divide the signal in clusters, do IFFT on each and then combine these. So that if we
subdivide 128 carrier in to a group of four 32 carriers, each, the max PAPR of each will
be 12 dB instead of 21 for the full. Combine these four sequences to create the transmit
signal.
• In OFDM, tight synchronization is needed. Often pilot tones are served in the subcarrier
space. These are used to lock on phase and to equalize the channel
• The sub-carriers are typically coded with Convolutional coding prior to going through
IFFT. The coded version of OFDM is called COFDM or Coded OFDM.
• Example 1: IEEE 802.11a uses OFDM with 52 subcarriers. Four of the subcarriers are
reserved for pilot tones, so effectively 48 subcarriers are used for data. Each data
subcarrier can be modulated in different ways. One example is to use 16 square QAM on
each subcarrier (which is 4 bits per symbol per subcarrier). The symbol rate in 802.11a
is 250k/sec. Find the data bit rate.
• Solution:
1. A single carrier system would have symbol period Ts = 1/20MHz or Ts = 50 ns. The
delay spread, 4 microseconds, is huge compared to 0.05 microseconds, so we definitely
need multicarrier modulation to mitigate the inter-symbol interference.
2. We want BN = B/N 0.1/ , that is, 25kHz. Note we also want N to be a power of 2 for
ease of implementation via the FFT. B = 20 MHz divided by 25 kHz is 800, so we choose
N = 1024, which results in a subcarrier bandwidth BN of 20 MHz /1024 = 19.5 kHz.
3. Calculating, μ, the overhead of the cyclic prefix, is μ = /Ts = 80.
4. The symbol period for OFDM is or TN + μTs = N/B + μ/B = (N + μ)/B = (1024 +
80)50 ns = 55.2μs.
5. Each carrier generates log2 4 = 2 bits per symbol period. There are 1024 carriers, for a
total of 2048 bits per symbol. Since one symbol equals 55.2 × 10−6s, the raw data rate
is 2048/(55.2×10−6 ) = 37.1 Mbps. Half of those bits (for rate 1/2 coding) are code
bits, so the real data rate is 111/2 = 18.5 Mbps.
• Subset of subcarriers are assigned to each user to allow users to send simultaneously
low data rate.
• WiMax
• 4G (LTE)
• OFDM tutorial:
http://www.csie.ntu.edu.tw/~hsinmu/courses/lib/exe/fetch.php?media=wn_11fall:of
dm_tutorial.pdf
• ECE 5325/6325: Wireless Communication Systems Lecture Notes, Fall 2011. Prof. Neal
Patwari, University of Utah Department of Electrical and Computer Engineering.