Wireless and Cellular Communication [18EC81]

MODULE-4
Key Enablers for LTE 4G
LTE = Long Term Evolution
Orthogonal Frequency Division Multiplexing (OFDM)
 OFDM has emerged as a technology of choice for achieving high data rates.
 It is the technology used by a variety of systems including Wi-Fi and WiMAX.
 OFDM is a multicarrier modulation technique; where the subcarriers are selected such
that they are all orthogonal to one another over the symbol duration, thereby avoiding the
need to have non overlapping subcarrier channels to eliminate inter carrier interference.
Advantages of OFDM are:
1) Elegant Solution to Multipath Interference
 Multicarrier modulation divides a given high bit rate data stream into several parallel
lower bit rate streams and modulate each streams on separate carriers called subcarriers or
tones.
 Splitting data stream into many parallel stream increases the symbol duration of each
stream such that the multipath delay spread is only a small fraction of symbol duration.
 Any residual Inter Symbol Interference (ISI) can be eliminated by using guard intervals
between OFDM symbols that are larger than the expected multipath delay.
 By making guard interval larger than expected multipath delay spread, ISI can be
completely eliminated.
2) Reduced Computational Complexity
 The Computational Complexity of OFDM is O(BlogBT m) where O is the order of
complexity, B is the bandwidth and Tm is the delay spread.
 It is much lower than the complexity of time domain equalizer based system O(B2Tm).
 Reduced complexity simplifies receiver processing and thus reduces mobile device cost
and power consumption.
3) Graceful degradation of performance under excess delay
 The Performance of OFDM system degrades gracefully as the delay spread exceeds the
value designed for.
 Greater coding and low constellation sizes can be used to provide feedback rates that are
significantly more robust against delay spread.
 OFDM is well suited for Adaptive Modulation and Coding.

Tilak, Dept. of ECE, GAT Page 1

 Wireless and Cellular Communication [18EC81]

4) Exploitation of frequency diversity

 OFDM facilitates coding and interleaving across sub carriers in the frequency domain
which can provide robustness against burst errors.
 OFDM allows for channel bandwidth to be scaled without impacting the hardware design
of the base station and mobile station.
5) Enables efficient multi access scheme
 OFDM can be used as a multi access schemeby partitioning different sub carriers among
multiple users which is referred to as OFDMA and is exploited in LTE.
 OFDMA offers the ability to provide fine granularity in channel allocation, which can be
exploited to achieve significant capacity improvements, particularly in slow time varying
channels.
6) Robust against narrowband interference
 OFDM is relatively robust against narrowband interference since such interference affects
only a fraction of sub carriers.
7) Suitable for coherent demodulation
 It is relatively easy to do pilot based channel estimation in OFDM systems, which renders
them suitable for coherent demodulation schemes that are more power efficient.
8) Facilitates use of MIMO
 OFDM converts a frequency selective broadband channel into several narrowband flat
fading channels where MIMO models and techniques work well to improve the system
capacity.
9) Efficient support of broadcast services
 By synchronizing base stations to timing errors well within OFDM guard interval, it is
possible to operate an OFDM network as a Single Frequency Network (SFN) which
allows broadcast signals from different cells to combine over the air to significantly
enhance the received signal power, thereby enhancing higher data broadcast
transmissions for a given transmit power.

SC-FDE and SC-FDMA

 Single Carrier Frequency Domain Equalization (SC-FDE) is similar to OFDM but instead
of transmitting the Inverse Fast Fourier Transform (IFFT) of actual data symbols, the data
symbols are sent as a sequence of QAM symbols with a cyclic prefix added and the IFFT
is added at the end of the receiver.

Tilak, Dept. of ECE, GAT Page 2

 Wireless and Cellular Communication [18EC81]

 SC-FDE has advantages such as multipath resistance and low complexity but have a low
Peak to Average Ratio of 4-5dB.
 Multi User version of SC-FDE is called Single Carrier Frequency Division Multiple
Access (SC-FDMA).
 SC-FDMA allows multiple users to use parts of frequency spectrum.
 SC-FDMA can be thought of as DFT precoded OFDMA.
 SC-FDMA increases the complexity of transmitter and receiver.

Channel Dependent multi User Resource Scheduling

 OFDMA allows for allocation in both time and frequency and it is possible to design
algorithms to allocate resources in a flexible and dynamic manner to meet arbitrary
throughput, delay and other requirements.
 The standard supports dynamic Channel Dependent Scheduling to enhance overall system
capacity.
 Frequency Selective multi user Scheduling calls for focusing transmission power in each
users best channel portion.
 In OFDMA, Frequency Selective Scheduling can be combined with multi user Time
Domain Scheduling, which calls for Scheduling users during the crests of their individual
fading channels.
 For high mobility users, OFDMA can be used to achieve Frequency Diversity and
Frequency Diversity Scheduling is best suited for control signalling and delay and delay
sensitive services.
 By coding and interleaving across subcarriers in the Frequency Domain using a uniform
random distribution of subcarriers over the whole spectrum, the signal can be made more
robust against Frequency Selective fading or burst errors.

Multi Antenna Techniques

 Advanced multi antenna solutions are implemented to improve link robustness, system
capacity and spectral efficiency.
1) Transmit Diversity
 It is a technique to combat Multipath Fading in wireless channel.
 Copies of same signal are sent, coded differently over multiple transmit antennas.

Tilak, Dept. of ECE, GAT Page 3

 Wireless and Cellular Communication [18EC81]

 This technique is based on Space Frequency Block Coding (SFBC) techniques

complemented with Frequency Shift Time Diversity (FSTD) when 4 transmit antenna are
used.
 It is primarily intended for common downlink channels.
 It is applied to user transmissions such as low data rate VoIP.
 It increases system capacity and cell range.
2) Beam Forming
 Multiple antennas are used to transmit the same signal appropriately weighted for each
antenna element.
 The effect is to focus the transmitted beam in the direction of the receiver and away from
interference, thereby improving received Signal to Interference Ratio.
 It provides improvements in coverage range, capacity, reliability and battery life.
 It is useful in providing angular information for user tracking.
 It is supported in the downlink.
3) Spatial Multiplexing
 Multiple independent streams can be transmitted in parallel over multiple antennas and
can be separated at the receiver using multiple receive chains through the appropriate
signal processing.
 It provides data rate and capacity gains proportional to the number of antennas used.
 It works well under good SNR and light load conditions.
 Spatial Multiplexing is supported with up to 4 transmit antennas and 4 receive antennas
by the LTE standard and is supported in the downlink.
4) Multi User MIMO (MU-MIMO)
 Multi User MIMO allows multiple users in the uplink, each with a single antenna to
transmit using the same frequency and time resource.
 The signal from different MU-MIMO users are separated at the base station receiver
using accurate Channel State Information of each user obtained through uplink reference
signals that are orthogonal between users.

IP based Flat Network architecture/ flat LTE SAE architecture

 Flat implies fewer nodes and less hierarchical structure for network.
 Fewer nodes implies lower infrastructure cost, fewer interfaces and protocol related
processing and reduced interoperability testing.

Tilak, Dept. of ECE, GAT Page 4

 Wireless and Cellular Communication [18EC81]

 Fewer nodes also allow better optimization of radio interface, merging of some control
plane protocols and short session start up time.
 3GPP Release 6 architecture has 4 network elements in the data path: the base station or
Node-B, Radio Network Controller (RNC), Serving GPRS Service Node (SGSN) and
Gateway GPRS Service Node (GGSN).
 Release 7 introduced a direct tunnel option from the RNC to GGSN which eliminated
SGSN from the data path.
 LTE has only 2 network elements in the data path: the enhanced Node-B (eNode-B) and
System Architecture Evolution Gate Way (SAE-GW).
 LTE merges the base station and Radio Network Controller functionality into single unit.
 The control path includes a functional entity called Mobility Management Entity (MME)
which provides control plane functions related to subscriber, mobility and session
management.
 The MME and SAE-GW could be collocated in a single entity called access Gate Way
(a-GW).
 All services including voice are supported on IP packet network using IP protocols.
 LTE envisions only a single Evolved Packet-Switched Core (EPC).
 Due to backward compatibility of LTE, GPRS tunnelling protocol and Packet Data
Convergence Protocol still exist.

Tilak, Dept. of ECE, GAT Page 5

 Wireless and Cellular Communication [18EC81]

LTE Network architecture/Evolved Packet Core architecture

 The Core Network designed presented in 3GPP Release 8 to support LTE is called
Evolved Packet Core (EPC).
 EPC is designed to provide a high capacity, all IP, reduced latency, flat architecture that
dramatically reduces cost and supports advanced real time and media rich services.
 EPC is designed not only to support new Radio Access Networks such as LTE but also
provide interworking with legacy 2G GERAN and 3G UTRAN networks connected via
SGSN.
 EPC provides functions such as access control, packet routing and transfer, mobility
management, security, radio resource management and network management.

 The EPC includes 4 new elements: Serving Gate Way (SGW), Packet Data Network
(PDN) Gate Way (PGW), mobility Management Entity (MME) and Policy Charging
Rules Function (PRCF).
1) Serving Gate Way (SGW)
 It terminates the interface towards the 3GPP radio access networks (RAN)
 It acts as a demarcation point between the RAN and Core network and management user
plane mobility.
 It serves as mobility anchor when terminals move across areas served by different node-B
elements in E-UTRAN- as well as acrossother3GPPradio networks such as GERANand
UTRAN.

Tilak, Dept. of ECE, GAT Page 6

 Wireless and Cellular Communication [18EC81]

 SGW does downlink packet buffering and initiation of network triggered service request
procedures
 Other function include packet routing and forwarding transport level packet marking in
the uplink and downlink support for per user and inter operator charging.
2) Packet Data Network Gate Way (PGW)
 It acts as the termination point of the EPC toward other Packet Data Networks (PDN)
such as the internet private IP network or the IP Multimedia Subsystem (IMS) network
providing end user services.
 It serves as an anchor point for sessions toward external PDN and provides functions such
as user IP address allocation, routing, policy enforcement, packet filtering and charging
support.
3) Mobility Management Entity (MME)
 It performs Signaling and Control functions to manage user terminal access to network
connections, assignment of network resources, mobility management functions such as
idle mode location tracking, paging, roaming and handover.
 It provides security functions such as providing temporary identities for user terminals,
interacting with Home Subscriber Server (HSS) for authentication and negotiation of
ciphering and integrity protection algorithms.
 It is responsible for selecting appropriate Serving and PDN Gate Ways and selecting
legacy Gate Ways for handovers to other GERAN or UTRAN networks.
 It is the point at which lawful interception of signalling is made.
4) Policy and Charging Rules Function (PCRF)
 It is a concatenation of Policy Decision Function (PDF) and Charing Rules Function
(CRF).
 It interfaces with the PDN Gate Way and supports service data flow detection, policy
enforcement and flow based charging.
 Release 8 enhanced PCRF functionality to include support for non 3GPP access such as
Wi-Fi of fixed line access to the network.

Tilak, Dept. of ECE, GAT Page 7

 Wireless and Cellular Communication [18EC81]

Multicarrier Modulation
The Multicarrier Concept
 Multicarrier modulation divides the high rate transmit bit stream into L lower rate
substreams, where L is chosen so that each of the subcarriers has effective symbol time
and is hence effectively ISI free.
 The individual substreams can then be sent over L parallel subcarriers, maintaining the
total desired data rate.
 Typically, the subcarriers are orthogonal under ideal propagation conditions, in which
multicarrier modulation is often referred to as Orthogonal Frequency Division
Multiplexing (OFDM).
 The data rate on each of the subcarriers is much less than the total data rate, and so the
corresponding subcarrier bandwidth is much less than the total system bandwidth.
 The number of substreams is chosen to ensure that each subcarrier has a bandwidth less
than the coherence bandwidth of the channel, so the subcarriers experience relatively flat
fading. Thus the ISI on each subcarrier is small.
 In the digital implementation of OFDM, the ISI can be completely eliminated through the
use of cyclic prefix.

Problem
 A certain wideband wireless channel has a delay spread of 1μsec. In order to overcome
ISI, assume a requirement that Ts ≥ 10.
1. What is the maximum bandwidth allowable in this system if the ISI constraint is to be
met without using multicarrier modulation?
2. If multicarrier modulation is used and we desire a 10MHz bandwidth, what is the
required number of subcarriers?
Solution:
 Given delay spread =  = 1μsec
 Therefore Ts ≥ 10 * 1 μsec = 10μsec
 Maximum bandwidth B = 1/ Ts = 1/ 10μsec = 100kHz
 If Multicarrier modulation is employed, the symbol time goes to T = LTs and Maximum
bandwidth B ≥ 1/ LTs .
 For Bandwidth B = 10MHz, Ts = 1/B = 0.1 μsec,
 Therefore, for B = 100kHz and Ts = 0.1 μsec
 L ≥ 1/B Ts ≥ 1/(100kHz * 0.1 μsec) ≥ 100
 Hence, the number of subcarriers L is chosen to be 128 to allow full 10MHz bandwidth to
be used with negligible ISI.

Tilak, Dept. of ECE, GAT Page 8

 Wireless and Cellular Communication [18EC81]

An Elegant Approach to ISI

 Multicarrier modulation in its simplest form divides the wideband incoming data stream
into L narrow band substreams, each of which is then transmitted over a different
orthogonal frequency subcarrier.
 A simple illustration of a multicarrier transmitter and receiver is shown.
 A high rate data signal of rate R bps and with a passband bandwidth B is broken into L
parallel substreams each with rate R/Land passband bandwidth B/L.

 The mutually orthogonal signals can be individually detected.

Tilak, Dept. of ECE, GAT Page 9

 Wireless and Cellular Communication [18EC81]

 After passing through the channel H(f), the received signal would appear as shown.

 In the time domain, the symbol duration of each subcarrier has increased to , so by
letting L grow larger, it can be assured that the symbol duration exceeds the channel delay
spread i.e. , which is a requirement for ISI-free communication.
 In the frequency domain, the subcarriers have bandwidth , which assures flat
fading, the frequency domain equivalent to ISI-free communication.

Shortcomings of multicarrier modulation

 A large bandwidth penalty will be inflicted since the subcarriers can’t have perfectly
rectangular pulse shapes and still be time limited.
 A very high quality (and hence expensive) low pass filters will be required to maintain
the orthogonality of the subcarriers at the receiver.
 The scheme requires L independent RF units and demodulation paths.

OFDM Basics
Block Transmission with Guard Intervals
 Grouping of "L” data symbols into a block is known as OFDM symbol.
 OFDM symbol lasts for duration of T seconds, where T = LT s and Ts is the symbol time.
 To keep each OFDM symbol independent of others after going through a wireless
channel, we introduce a guard time in between each OFDM symbol.

 After receiving a series of OFDM symbols, as long as T g » τ, each symbol will interfere
with only itself.

Tilak, Dept. of ECE, GAT Page 10

 Wireless and Cellular Communication [18EC81]

 OFDM transmissions allow ISI within an OFDM symbol, but by including a sufficiently
large guard band, it is possible to guarantee that there is no interference between
subsequent OFDM symbols.

Circular Convolution and DFT

 Circular Convolution is given by
where
 DFT of channel output is given by

 In Frequency Domain,
 The L-point DFT is given by
 The IDFT is defined as

 The estimate

The Cyclic Prefix

 If a cyclic prefix is added to the transmitted signal, then this creates a signal that appears
to be and .

 If the maximum channel delay spread has a duration of v+1 samples, then by adding a
guard band of atleast v samples between OFDM symbols, each OFDM symbol is made
independent of those coming before and after it and so just a single OFDM symbol can be
considered.
 OFDM symbol is represented in Time Domain as a length L vector given by
.

Tilak, Dept. of ECE, GAT Page 11

 Wireless and Cellular Communication [18EC81]

 After applying a cyclic prefix of length v, the actual transmitted signal is

 The output of channel is .

where h is a length v+1 vector describing impulse response of channel.
 has (L+v) + (v+1) – 1 = L + 2v samples.
 The first v samples contain interference from the preceding OFDM symbol and so
discarded.
 The last v samples disperse into the subsequent OFDM symbol and so discarded.

 Due to cyclic prefix, depends on and the circularly wrapped values .

 That is,

 A cyclic prefix that is atleast as long as the channel duration allows the channel output yto
be decomposed into a simple multiplication of the channel frequency response
and the channel frequency domain input .
 Due to cyclic prefix, the bandwidth increases from B to and hence carries a
power penalty of in addition to the bandwidth penalty.
 Therefore, Rate Loss = Power Loss = .
 Since the cyclic prefix length is fixed, least rate loss is incurred for longest block size and
most rate loss is incurred for shortest block size.

Tilak, Dept. of ECE, GAT Page 12

 Wireless and Cellular Communication [18EC81]

Problem
 Consider a channel where the maximum delay spread has been determined to be
τ = 2μsec. What is the minimum and maximum data rate loss due to the cyclic prefix?
What is the minimum theoretical loss if a shorter cyclic prefix could be used?
Solution
 The normal Cyclic Prefix (CP) length / Guard Time T g = 144 * Ts ≈ 4.7μsec which is
clearly longer than delay spread τ = 2μsec and so sufficient for this channel.
 Since CP length is fixed, least overhead (rate loss) will be incurred for the longest block
size and most overhead (rate loss) would be incurred for the shortest block size.
 With maximum block size Lmax= 2048 sub carriers, and block size L=128 sub carriers:
 CP length = v = 144 *Ts = 144 * L / Lmax = 144 * 128 / 2048 = 9 data symbols.
 Rate Loss factor = L / (L+v) = 128 / (128 + 9) = 128 / 137 = 0.934 ≈ 93% and hence
a rate loss of less than 7%.
 With maximum block size Lmax = 2048 sub carriers, and block size L=2048 sub carriers:
 CP length = v = 144 * Ts = 144 * L / Lmax = 144 * 2048 / 2048 = 144 data symbols.
 Rate Loss factor = L / (L + v) = 2048 / (2048 + 144) = 2048 / 2192 = 0.934 ≈ 93%
and hence a rate loss of less than 7%.

 If the Cyclic Prefix was even shorter, the rate loss would be less.
 For block size L = 128 sub carriers, with Tg = 4.7μsec, there are 9data symbols.
 With τ = 2μsec = Tg, there are (2μsec * 9) / 4.7μsec = 3.82 ≈ 4 data symbols = v
 Rate Loss factor = L / (L + v) = 128 / (128 + 4) = 128 / 132 = 0.969 ≈ 97% and
hence a rate loss of less than 3%.
 For block size L = 2048 sub carriers, with Tg = 4.7μsec, there are 144 data symbols.
 With τ = 2μsec = Tg, there are (2μsec * 144) / 4.7μsec = 61.27 ≈ 62 data symbols = v
 Rate Loss factor = L / (L + v) = 2048 / (2048 + 62) = 2048 / 2110 = 0.970 ≈ 97%
and hence a rate loss of less than 3%.

An OFDM Block Diagram

Tilak, Dept. of ECE, GAT Page 13

 Wireless and Cellular Communication [18EC81]

 A wideband signal of Bandwidth B is broken into L narrowband signals (sub carriers)

each of bandwidth B/L.
 In order to use a single wideband radio instead of L independent narrowband radios, the
sub carriers are created digitally using an IFFT operation.
 In order for the IFFT/FFT to decompose the ISI channel into orthogonal sub carriers, a
cyclic prefix of length v must be appended after the IFFT operation. The resulting L+v
symbols are then sent in serial through the wideband channel.
 At the receiver, cyclic prefix is discarded, and the L received symbols are demodulated
using an FFT operation which results in L data symbols each of the form
for sub carrier l.
 Each sub carrier can then be equalized via an FEQ by simply dividing by the complex
channel gain for that carrier. This results in .

OFDM in LTE

 The L independent QAM symbols (the vector X) act as input and are treated as separate
sub carriers.
 These L data bearing symbols are created from a bit stream by a symbol mapper and
serial to parallel converter (S/P).
 The L-point IFFT then creates a time domain L vector x that is cyclic extended to have
length L(1+G) where G is fractional overhead.
 The longer vector is then parallel to serial (P/S) converted into a wide digital signal that
can be amplitude modulated with a single radio at a carrier frequency of .
 The raw data rate of LTE system is .

Tilak, Dept. of ECE, GAT Page 14

 Wireless and Cellular Communication [18EC81]

Problem
Find the raw data rate of LTE system, if 16 QAM was used (M=16) with normal cyclic prefix
for the given specifications:
Bandwidth = B=10MHz, Number of sub carriers = L=1024, data sub carriers = L d =600 and
Guard Fraction = G=0.07.
Solution
= (107 / 1024) * [(600 * log2 (16)) / (1 + 0.07)] = 21.9Mbps.

Timing and Frequency Synchronization

 The timing offset of the symbol and the optimal timing instants needs to be determined
which is referred to as Timing Synchronization.
 The receiver must align its carrier frequency as closely as possible with the transmitted
carrier frequency which is referred to as Frequency Synchronization.
 The primary and secondary sync signals occupy the 31 tones on either side of the DC
sub carrier in the 5th and 6th OFDM symbols respectively, of the 0th and 10th slots in every
radio frame.
 The 5 adjacent tones on top and bottom of sync signals are not used so 62 tones are used
or a total of 62*15kHz = 930kHz ≈ 1MHz bandwidth.
 Timing Synchronization offset is 0.1/ Bs = 0.1μsec which is less than cyclic prefix length
of 0.5μsec.

Tilak, Dept. of ECE, GAT Page 15

 Wireless and Cellular Communication [18EC81]

Timing Synchronization
 Since the time window is T=1μs and a rectangular window is used, the frequency
response of each sub carrier becomes a sinc function with zero crossings every 1/T =
1MHz.
 This can be confirmed by using Fourier Transform

where
 If the timing window is slid to left or right, a unique phase change will be introduced to
each of the sub carriers.
 In the frequency domain, if the carrier frequency synchronization is perfect, the receiver
samples at the peak of each sub carrier where the desired sub carrier amplitude is
maximized and ICI is zero.
 In the case, the perfect timing synchronization is not maintained, it is still possible to
tolerate a timing offset of τ sec without any degradation in performance as long as
o .

Tilak, Dept. of ECE, GAT Page 16

 Wireless and Cellular Communication [18EC81]

 As long as , the timing offset simply results in a phase shift per sub
carrier of which is fixed for all sub carriers.
 The acceptable range of τ is referred to as timing synchronization margin.
 The SNR loss is approximated as .
 The SNR decreases quadratically within the timing offset.
 Longer OFDM symbols are increasing immune from timing offset.
 Since , timing synchronization errors are not that critical as long as induced
phase change is corrected.

Frequency Synchronization
 As long as the frequency offset , there is no interference between the sub carriers.
 Frequency offset ≈ .
 The matched filter receiver corresponding to sub carrier l can be simply expressed for the
case of rectangular windows as where .
 Since LTs is the duration of the data portion of the OFDM symbol, .
 An interfering sub carrier m can be written as .
 If the signal is demodulated with a fractional frequency offset of ,, then
o .
 The ICI between sub carriers l and l+m using a matched filter is simply the inner product
between them given as
 and .
 The total average ICI energy per symbol on sub carrier l is given as

 is constant that depends on various assumptions and is the average symbol energy.
 The SNR loss induced by frequency offset is given by

 SNR decreases quadratically with the frequency offset.

 SNR decreases quadratically with the number of sub carriers.
 The loss in SNR is also proportional to SNR itself.

Tilak, Dept. of ECE, GAT Page 17

 Wireless and Cellular Communication [18EC81]

Peak to Average Ratio (PAR)

 OFDM signals have high PAR since in time domain, a multi carrier signal is the sum of
many narrow band signals.
 High PAR reduces efficiency and hence increases cost of RF power amplifier.
 When a high peak signal is transmitted through a nonlinear device such as a High Power
Amplifier (HPA) or Digital to Analog Converter (DAC), it generates out of band
energy (spectral regrowth) and in band distortion (constellation tilting and scattering)
which cause degradation and affect the system performance.
 The nonlinear behavior of HPA can be characterized by AM/AM and AM/PM responses.
 To avoid undesirable nonlinear effects, a waveform with high peak power must be
transmitted in the linear region of HPA by decreasing the average power of input signal
which is called Input Back Off (IBO) and results in a proportional Output Back Off
(OBO).
 High Back Off reduces the power efficiency of HPA and limits the battery life for mobile
applications.
 The input back off is defined as
where is the saturation power and is the average input power.

Tilak, Dept. of ECE, GAT Page 18

 Wireless and Cellular Communication [18EC81]

 The power efficiency of an HPA can be increased by reducing the PAR of transmitted
signal.
 The efficiency of class A amplifier is halved when the input PAR is doubled or the
operating point (average power) is halved.
 It is desirable to have average and peak values to be as close as possible in order to
maximize the efficiency of power amplifier.
 A high PAR requires high resolution for both the transmitter’s digital to analog convertor
and receiver’s analog to digital convertor since the dynamic range of signal is
proportional to PAR.
 High resolution D/A and A/D conversions plays an additional complexity, cost and power
burden on system.

Quantifying the PAR

 The amplitude of the output signal is .
 Since is complex Gaussian, the envelope is Rayleigh distributed with the
parameter .

Tilak, Dept. of ECE, GAT Page 19

 Wireless and Cellular Communication [18EC81]

 The output power is therefore which is exponentially

distributed with mean .
 Since the output amplitude and power are random, PAR is not a deterministic quantity.
 The PAR of the transmitted analog signal can be defined as .
 The discrete time PAR can be defined for IFFT output as
.
 Although the average energy of IFFT outputs x[n] is the same as the average energy of
the inputs X[m] and equal to , the analog PAR is not exactly the same as the PAR of
IFFT samples, due to the interpolation performed by the D/A convertor.
 The analog PAR is usually higher than digital (Nyquist sampled) PAR.
 The analog PAR determines the PA performance.
 An oversampling factor of M can be used to interpolate the digital signal in order to better
approximate its analog PAR, i.e. an OFDM symbol is sampled LM times rather than
simply L.The maximum possible value of PAR is L.
 The Complementary Cumulative Distribution Function (CCDF) of PAR for L ≥ 64 is
given as
where is the peak power level and is a pseudo-approximation of the oversampling
factor which is given empirically as = 2.8.
 PAR is and is the CDF of a single Rayleigh distributed sub carrier with
parameter .
 The CDF of the Nyquist sampled signal power can be obtained by
.

Tilak, Dept. of ECE, GAT Page 20

 Wireless and Cellular Communication [18EC81]

SC-FDE
 Similar to OFDM, SC-FDE maintains 3 benefits:
a) Low complexity even for severe multipath channels.
b) Excellent BER performance.
c) Decoupling of ISI from other types of interference.
 The only difference between OFDM and SC-FDE system is that the IFFT is moved to the
end of the receive chain rather than operating at the transmitter, to create a multi carrier
waveform.
 The transmitted signal is simply a sequence of QAM symbols which have low PAR on
the order of 4-5dB depending on constellation size.
 Because of the application of cyclic prefix, the received signal appears to be circularly
convolved i.e. .
 Therefore, .

 SC-FDE has a lower complexity transmitter and higher complexity receiver compared
to OFDM.
 While OFDM has high PAR and more subjected to clipping and cause spectral dispersion,
SC-FDE has low PAR and nominally more dispersive spectrum compared to OFDM.
 In a cellular system like LTE, uplink utilizes SC-FDE whereas downlink utilizes
OFDM.

The Computational Complexity Advantage of OFDM and SC-FDE

 Frequency Domain Equalization requires much lower Computational Complexity
especially for high data rates.
 A Time Domain Equalizer consists of a series of multiplications with several delayed
versions of the signal.

Tilak, Dept. of ECE, GAT Page 21

 Wireless and Cellular Communication [18EC81]

 The number of delay taps in an Equalizer depends on the symbol rate of the system and
the delay spread in the channel.
 The number of Equalizer taps is proportional to the bandwidth delay spread product
.
 An Equalizer with taps performs complex multiply and accumulate (CMAC)
operations per receives symbol.
 Therefore the complexity of an Equalizer is of the order .
 IFFT and FFT have a complexity of where ‘L’ is the FFT block size.
 The computational complexity for each OFDM symbol (SC-FDE block) is of the order
.
 There are B/L OFDM symbols sent each second and since this means there are
order OFDM symbols per second so the computational complexity in terms of
CMACs for OFDM is .
 The complexity of a Time Domain Equalizer grows as the square of the data rate since
both the symbol rate and number of taps increases linearly with the data rate.
 LTE uses SC-FDMA and the complexity of SC-FDMA scales as .

Tilak, Dept. of ECE, GAT Page 22