See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/284502697

Single Carrier FDMA for Next Generation Communication Systems: An Overview

Conference Paper · December 2015

CITATIONS READS

0 689

3 authors:

Madhusudan Sinha Vinay Kumar Trivedi

Indian Institute of Technology Madras Indian Institute of Technology Patna
3 PUBLICATIONS   0 CITATIONS    21 PUBLICATIONS   24 CITATIONS   

SEE PROFILE SEE PROFILE

P. Kumar
Indian Institute of Technology Patna
78 PUBLICATIONS   262 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Waveform Design View project

Multi-user Diversity Gain for different Multiple access schemes View project

All content following this page was uploaded by Vinay Kumar Trivedi on 24 November 2015.

The user has requested enhancement of the downloaded file.

 Single Carrier FDMA for Next Generation
Communication Systems: An Overview
Madhusudan Kumar Sinha∗ , Vinay Kumar Trivedi, StudentMember, IEEE † , Preetam Kumar, Member, IEEE ‡
Department of Electrical Engineering
Indian Institute of Technology Patna
Bihar 801118, India
Email: {∗ madhusudan.mtcm14, † vinay.pee14, ‡ pkumar}@iitp.ac.in

Abstract—SC-FDMA (Single carrier frequency division mul- delay spread of the channel is less than the interval between
tiple access) also known as DFT spread FDMA is a modified two consecutive pulses then there is no interference between
form of OFDMA (Orthogonal frequency division multiple access). consecutive pulses. However when the pulse duration is less
3GPP (Third Generation Partnership Project) has chosen SC-
FDMA technology for LTE uplink transmission due to its lower than the delay spread of the system consecutive pulses start
PAPR (Peak to average power ratio) compared to OFDMA. interfering with each other, degrading the performance of
SC-FDMA is being actively studied for other areas where the communication system. Usually delay spread of wireless
longer battery life is desirable like in underwater acoustic channel is given by
communication, sensor networks etc. In this paper we give an
overview of SC-FDMA with areas and significance of research. Dmax
OFDMA was well utilized for achieving high spectral efficiency τmax = µs (1)
0.3
in communication system. SC-FDMA is introduced recently and
it became handy candidate for uplink multiple access scheme in where Dmax is the maximum distance between transmitter and
LTE system, that is under 3GPP. The Multiple Access Scheme receiver [1]. Due to inter symbol interference delay spread
in Advanced Mobile radio system has to meet the challenging of the channel and symbol rate plays a significant role in
requirements, for example high throughput, good robustness,
efficient Bit Error Rate (BER), high spectral efficiency, low system design. We have no control over wireless channel
delays, low computational complexity, low Peak to Average Power and hence the delay spread of the channel, however we do
Ratio (PAPR), low error probability etc. In this paper focus areas have control over bitrate of transmission. Hence in order to
of research activities have been identified and a summary of the combat ISI (intersymbol interference) multicarrier schemes
ongoing activities and future trends has been presented. were introduced. MultiCarrier modulation is the principle of
Index Terms—SC-FDMA, OFDMA, CDS, PAPR, Subcarrier
mapping transmitting intended data by dividing the incoming bitstream
into several parallel bitstreams, each of which has lower bit
rates, and by using these substreams to modulate several
I. I NTRODUCTION
carriers.
Every technological advancement is driven by the need of Earlier FDM was in use, where different carriers were used
that time, similarly communication technologies arent any ex- to modulate substreams of data and guard frequency band
ception. Earlier needs of communication systems were limited were used to isolate each subcarriers in frequency. Due to
by low bitrate and lower number of users. With increase in inefficiency of FDM in bandwidth utilization, OFDM was
bitrate and number of users, multicarrier schemes and multiple introduced. In OFDM, we use orthogonal frequency carriers
access techniques were introduced. Multicarrier modulation is to modulate substreams without the need of any guard band.
the principle of transmitting intended data by dividing the However the subcarriers are being transmitted in parallel
incoming bit stream into several parallel bit streams, each which causes high PAPR in OFDM. In order to improve
of which has lower bit rates and these sub streams were PAPR performance of OFDM, a modification was made in
used to modulate several carriers. Multiple access technique OFDM system and hence SC-FDMA was introduced, where
or channel access scheme is a method of sharing a common subcarriers were transmitted in series. SC-FDMA has similar
channel among multiple users so that they can communicate BER performance as OFDMA but has significantly lower
simultaneously without interference from each other. PAPR than OFDMA.
Due to multiple path propagation nature of wireless commu- With time, the demand for high bitrate systems is increasing
nication channel, the impulse response consists of shifted and and with increasing bitrate, multiple access technique has
scaled impulse functions. This leads to multiple copies of changed from TDMA to CDMA(code division multiple ac-
same signal shifted and scaled randomly to be received at cess) and now to FDMA(OFDMA) [2]. Also with increasing
the receiver. Time duration over which significant amount of bit rate, bandwidth of transmission is increasing and frequency
signal energy is contained within the signal copies is called selectivity of channel is becoming prominent [3] and thus our
delay spread (τ). Due to statistical nature of delay spread focus is shifting towards multicarrier schemes. OFDMA is
rms value of delay spread, τrms is used to quantify it. If the a multicarrier scheme which utilizes orthogonal subcarriers
 Figure 1: Transmitter and receiver structure of SC-FDMA.

transmitted in parallel to achieve higher spectral efficiency frequency and time domain multicarrier approach for multiple
and lower ISI. A big advantage of OFDMA is its robustness access has been adopted by 3GPP. For downlink, OFDMA is
in presence of multipath propagation, which comes from the selected whereas for uplink SC-FDMA is selected. SC-FDMA
fact that an OFDMA system transmits information on M uses single carrier modulation at the transmitter and frequency
orthogonal frequency carriers, each operating at M1 times the domain equalization at the receiver and has similar complexity
bitrate of the information signal [4]. However OFDM has a big and performance as compared to OFDMA. SC-FDMA has one
disadvantage in terms of high PAPR. SC-FDMA has similar major advantage over OFDMA in terms of lower PAPR which
performance and structure as that of OFDMA except for the makes it technology of choice for uplink transmission. [3] and
fact that in SC-FDMA, a DFT precoded signal is used and [4] presents basics of SCFDMA with study of basic results of
all the subcarriers are transmitted in series in contrast to the PAPR and throughput.
parallel transmission of subcarriers in OFDMA. Because of The block of M complex modulation symbols are generated at
the fact that symbols are transmitted in serial fashion only a rate of R symbols /second after that M − point DFT produces
one subcarrier is present at any particular time and hence it M frequency domain symbols that modulate M subcarriers out
gives lower PAPR characteristics. However relatively complex of N orthogonal subcarriers. The Channel transmission rate is
frequency domain equalization is needed at the receiver. This Rchannel and band spreading Q is therefore written as
makes SC-FDMA system more suitable for the uplink because N
power consuming computation are shifted to the receiver Rchannel = R (2)
M
at the base station where power constraint is not as hard
as in mobile terminal. [2] presents detailed comparison of Rchannel N
BER performance of OFDMA and SC-FDMA systems in Q= = (3)
R M
general fashion, where it has been shown that SC-FDMA In figure 2 schematic of SC-FDMA transmitter, xm (m =
performs better at low SNR but sometimes at higher SNR, 0, 1, ., M − 1) represents M samples of the modulated source
OFDMA can perform better. [10] presents PAPR and spectral symbols. Xk (k = 0, 1, ., M − 1) represents M samples of DFT
efficiency comparison between OFDMA and SC-FDMA for of xm . The l th sample after DFT operation is
100 Mbps and 1 Gbps systems in local area IMT-A scenarios
M−1
is presented. In [5] an efficient blind CFO estimation algorithm l

for SC-FDMA scheme is presented. The remaining part of

Xk (l) = ∑ xm (u)e− j2πm N (4)
m=0
this paper is organised as follows: Section II presents single
carrier FDMA system model. Section III presents conclusion Where xm (u) represents the modulated source symbols from
and scope of future work in this area. uth user. Xk (l) now modulates the orthogonal subcarriers out of
N. Yl (l = 0, 1, ., N − 1) is the frequency domain samples after
II. S INGLE C ARRIER FDMA subcarrier mapping. These frequency domain samples are now
Single Carrier Frequency Division Multiple Access (SC- passed to IDFT block which gives yn (n = 0, 1, .N − 1). The
FDMA) also known as DFT spread OFDM has been selected process of subcarrier mapping is also referred as scheduling
by third generation partnership project (3GPP) for high bitrate and more accurately channel dependent scheduling as shown in
uplink cellular communication for long term evolution (LTE). figure 1. The purpose is to map the symbols to a set of subcar-
To achieve high spectral efficiency and efficient scheduling in riers with most favourable transmission characteristics which
in turn is dependent on channel model. Different subcarriers directly to modulate subcarriers however in SC-FDMA fre-
mapping formats give different characteristics over different quency components obtained by DFT blocks are used to
channels, the task is to choose a subcarrier mapping scheme modulate the subcarriers. In one sense DFT operation can be
which best suites the characteristics of the channel and give seen as providing nonuniform weighted average of the input
favourable results. Only multiuser schemes such as OFDMA symbols. This means a lot more averaging functions can be
and SC-FDMA enjoy this property as spatially dispersed users defined and studied to replace the DFT block with desired
have different fading channels. properties. For example in [10] it is shown that DCT SC-
Next a Parallel to serial convertor places y0 , y1 , ....yN−1 in a FDMA system has better BER performance than DFT SC-
sequence which is suitable for modulating a carrier. Transmit- FDMA, where N modulated symbols are used to generate
ter also adds a cyclic prefix (CP) which is inserted at the N frequency domain symbols, which are then mapped on
beginning of each symbol in order to prevent intersymbol available subcarrier space. The advantage of DFT precoding
interference, length of cyclic prefix is always taken greater is that in case of frequency nulls data is not completely lost
than the delay spread of the channel. Since CP is a copy of as data is extracted by IDFT, an averaging function over all
last parts of a symbol it converts linear convolution to circular frequency components.
convolution, therefore transmitted data through channel can
C.Subcarrier Mapping
be modelled as circular convolution between channel impulse
response and transmitted data (point wise multiplication). Subcarrier mapping assigns frequency domain modulation
symbols to available subcarriers. N frequency domain symbols
A.Baseband Modulation corresponding to one user are mapped over M > N subcarriers,
where M is the total number of subcarriers assigned to the SC-
The incoming bit stream is a conceptual entity and we need FDMA system for all users. Subcarrier mapping schemes are
physical realization of data for modulation. In most of the primarily of two types, Localised and Distributed. We will
cases the incoming data is in form of on off keying (OOK) consider a total of four main subcarrier mapping schemes
generated from computer/clocked source but is generally not which are LFDMA, IFDMA, DFDMA, BIFDMA.
the most efficient representation of data in terms of energy and
bandwidth utilization. Thus we need to represent the incoming
data more efficiently depending on system requirements. The
input data stream is modulated to a two dimensional signal
before applying to input DFT block of the SC-FDMA. The
type of modulation scheme (BPSK, QPSK or M-QAM) has
prominent effect on PAPR and BER performance of the system
as it dictates the amplitude variation and euclidian distance of
the received signal constellation. [6] presents PAPR, BER and
Spectral study of the application of Amplitude Phase Shift
Keying (APSK) in nonlinearly distorted SC-FDMA uplink
system and compares the result with the widely accepted M-
QAM modulation scheme. It has been shown that MAPSK
Figure 2: Different Subcarrier mapping schemes in frequency domain
provides high robustness against nonlinear amplification and
its application is specially reasonable in highly nonlinearly
LFDMA stands for localized SCFDMA. In this mapping
disturbed scenarios. [7] and [8] analyses effect of input and
scheme each user uses only a part of the total bandwidth and
output block sizes on PAPR and BER of DFT SC-FDMA and
contiguous frequency subcarriers are being used for transmis-
DCT SC-FDMA and tries to find the optimal block size for
sion of data. Block structure provides robustness to carrier
significantly lower PAPR and BER. Owing to the fact that SC-
frequency offsets as well as provides high multi user diversity
FDMA is a multiple access technique, IQ imbalances originat-
in case of adaptive resource allocation. LFDMA shows lesser
ing from different analog devices in different transmitters can
frequency diversity as subcarriers allotted to each user is
severely degrade the performance of the system. [9] studies the
localized within a part of the bandwidth.
effect of IQ imbalance in different transmitter stations and its
DFDMA stands for distributed SC-FDMA. In this mapping
effect on PAPR and BER of the whole system for SC-FDMA
scheme each user gets noncontiguous subcarriers. DFDMA
as well as OFDMA.
have following two main subtypes of subcarrier mapping that
are of more interest than others. IFDMA stands for Interleaved
B.DFT Precoding
SC-FDMA, in which data from DFT block is mapped on
Currently DFT spread OFDM is popularly known as SC- subcarriers which are equally spaced over entire bandwidth.
FDMA, however other precoding techniques are also being Unoccupied places are initialized with zero. IFDMA provides
studied such as DCT precoded FDMA in [10] and hadamard larger frequency diversity as compared to LFDMA as subcarri-
precoded SC-FDMA in [11]. One major difference between ers are distributed over entire bandwidth. PAPR performance
FDMA and SC-FDMA is that in FDMA data bits are used gets improved but BER degrades as compared to LFDMA.
IFDMA provides flexibility in subcarrier allocation. Other iterative equalization techniques are also being studied
BIFDMA stands for Block Interleaved SC-FDMA. BIFDMA like turbo equalization [16] which uses similar algorithm as
is considered as hybrid symbol mapping technique which pos- used in turbo coding. In SC-FDMA equalization is done in
sesses both the qualities of IFDMA and LFDMA. In BIFDMA frequency domain and the equalized frequency components
unlike IFDMA we take a block of data, which is mapped on are then transformed into time domain symbols where data
adjacent group of subcarriers. Rest of the unoccupied positions detection is performed. Sometimes length of cyclic prefix
is initialized to 0. It has better robustness to carrier frequency is less than channel delay spread and causes problem with
offset as compared to IFDMA as symbols are mapped in group linear frequency domain equalization. [17] presents frequency
which provides advantages of LFDMA. As the symbols are domain equalization when CP is less than channel delay
spread over entire bandwidth so it provides better frequency spread.
diversity as compared to LFDMA.

D.Channel Estimation F.Channel Dependent Scheduling

Communication system design depends on channel charac- Static resource scheduling has its advantage of being simple
teristics, which at any particular time can be precisely specified and doesn’t require any feedback. However in order to max-
by the impulse response of the channel. However the channel imize the mean throughput (and thus total throughput) and
is ever changing and thus we must use statistical estimates of minimize the variance of throughput we need to allocate time
the channel which can be considered constant over a small and frequency resource of the system according to channel
and finite duration of time. For a system to perform better, conditions. These two optimization problems are interlinked
good channel estimates are needed as well as the duration over with each other and have complex tradeoffs. Channel depen-
which these estimates are relatively constant is important. The dent scheduling (CDS) requires the system to monitor channel
time duration in which channel characteristics are relatively conditions as a function of subcarriers at each terminal, and
constant is known as coherence time and plays a significant modify the subcarrier mapping to achieve optimization. For a
role in communication system design. It determines how often typical SCFDMA system with 256 or 512 subcarriers checking
we need to test our channel for its properties using pilot every combination for best possible solution is not practi-
signals, and thus dictates the maximum length of an SC- cally feasible. We thus divide the the total subcarrier space
FDMA data block. [12] propose an interpolation algorithm into chunks with the condition that a particular optimization
based on adaptive order polynomial fitting for LTE uplink solution views a chunk as a single entity and assigns it
channel estimation to mitigate ICI in high Doppler spread with to a particular user. Two main chunk assignment include
better performance compared to conventional technique. [13] interleaved and localised subcarrier assignment. IFDMA with
presents iterative estimation algorithms with hard and soft fre- CDS can offers the advantage of frequency diversity while
quency replacement to achieve optimal frequency replacement. LFDMA with CDS can potentially offer multiuser diversity
[4]. Results shows that CDS increases throughput by up to
E.Frequency Domain Equalization 80% relative to static scheduling for LFDMA but by only up
At receiver some frequencies may be in deep fade while to 26% for IFDMA. In [18] authors investigate the impact of
others may be in good condition. This deformation in fre- imperfect channel state information (CSI) on CDS. It is also
quency domain causes time domain symbols to spread out, showed that localized mapping is very sensitive to the quality
that cause intersymbol interference. This can be clearly seen in of CSI (Channel state information) and capacity gain quickly
single carrier modulation scheme SC-FDMA, which is bound decreases when the channel changes rapidly. For high mobility
to have ISI and to combat that we need frequency domain users, distributed mapping with static round robin scheduling
equalization. The reason behind adopting frequency domain is more suitable.
equalization and not considering time domain equalization is
that in case of tap delay equalization, we need much longer G. Peak to Average Power Ratio (PAPR) and Pulse Shaping
delay extending to multiple symbols and that results into lot
of computations. In frequency domain equalization we start Peak to average power ratio (PAPR) is a performance mea-
with zero forcing equalization, which is a form of non adap- sure for efficiency of the transmitter. A positive PAPR in semi
tive linear equalization technique, which multiplies inverse logarithmic scale indicates that we require a power back off
of channel frequency response to the received signal. [14] to operate in the linear region of transmitter power amplifier.
presents details of zero forcing equalization over nakagami In this section we analyse the PAPR of SC-FDMA signal.
channel. Zero forcing algorithm causes noise components PAPR is determined by the combination of Modulation used,
to get amplified for deep fade subcarriers and thus doesnt Constellation of signal and pulse shaping used. High PAPR
help much with SNR. Thus MMSE is prefered over zero reduces power efficiency since the amplifier must operate with
forcing equalization due to its robustness to noise amplifi- large back off and more linearity. Pulse shaping is adopted in
cation. MMSE equalizer minimizes the mean square error cellular communications and is used extensively [19]. However
(MSE) of the estimated value to the actual transmitted value. its use in underwater communications is still relatively new.
[15] presents iterative MMSE equalization for SC-FDMA. Without pulse shaping, symbol rate sampling gives the same
PAPR as the continuous case which is given as [2] J. Gazda, P. Drotár, P. Galajda, and D. Kocur, “Comparative evaluation
of ofdma and sc-fdma based transmission systems,” in Applied Machine
2
max |xen | Intelligence and Informatics (SAMI), 2010 IEEE 8th International
PAPR = 2
(5) Symposium on. IEEE, 2010, pp. 177–181.
1
N ∑n=N−1
n=0 |xen | [3] H. G. Myung, “Introduction to single carrier fdma,” in Proceedings of the
15th European Signal Processing Conference (EUSIPCO07). Citeseer,
Where |xen |, n = 0, 1, .N − 1 are the time domain samples 2007, pp. 2144–2148.
after IDFT block. The two conflicting requirements for any [4] H. G. Myung, J. Lim, and D. J. Goodman, “Single carrier fdma for
communication systems are demand for high data rates and uplink wireless transmission,” Vehicular Technology Magazine, IEEE,
vol. 1, no. 3, pp. 30–38, 2006.
need for narrow bandwidth. Narrow bandwidth channels with [5] V. Trivedi, V. Sagar, and P. Kumar, “An efficient deterministic scheme
wide symbols causes ISI. One possible solution is to use ideal for blind cfo estimation in sc-fdma,” in International Symposium on
low pass filter (LPF) which is unrealizable and difficult to ap- Wireless Personal Multimedia Communications (WPMC15) (accepted),
2015 IEEE, 2015, pp. 1–5.
proximate. Approximate LPF cause jitter which is detrimental [6] J. Gazda, D. Dupák, and D. Kocur, “M-apsk modulation for sc-fdma
but if the frequency characteristics have odd symmetry at the communication systems,” in Radioelektronika (RADIOELEKTRONIKA),
cut-off frequencies the impulse response will have zeros at 2011 21st International Conference. IEEE, 2011, pp. 1–4.
[7] M. Pervej, M. Sarkar, and M. T. Islam, “Impact analysis of different
uniformly spaced intervals and for this purpose pulse shaping inputs and outputs block sizes of dft-scfdma system,” in Electrical
is used. Engineering and Information Communication Technology (ICEEICT),
For simulation and experimental purpose we have used raised 2015 International Conference on. IEEE, 2015, pp. 1–6.
[8] M. Pervej, M. Sarkar, T. Roy, and M. Koli, “Impact analysis of input and
cosine pulse shape, defined as follows in time domain, output block size of dct-scfdma system,” in Computer and Information
Technology (ICCIT), 2014 17th International Conference on. IEEE,
t cos( παt
T ) 2014, pp. 440–445.
r(t) = sinc(Π ) (6)
T 1 − 4α 22t 2 [9] Y. Yoshida, K. Hayashi, H. Sakai, and W. Bocquet, “Analysis and com-
T pensation of transmitter iq imbalances in ofdma and sc-fdma systems,”
Where α is roll off factor of the filter. Roll off factor is Signal Processing, IEEE Transactions on, vol. 57, no. 8, pp. 3119–3129,
2009.
the measure if the excess bandwidth of the filter that is [10] F. S. Al-Kamali, M. Dessouky, B. M. Sallam, F. Shawki, F. E. A. El-
1
occupied beyond the Nyquist bandwidth of 2T , where T1 Samie et al., “A new single carrier fdma system based on the discrete
is the symbol rate. As roll off factor increases eye in eye cosine transform,” in Computer Engineering & Systems, 2009. ICCES
2009. International Conference on. IEEE, 2009, pp. 555–560.
diagram opens up and hence improves the performance but [11] H.-G. Yeh and H. Mutahir Abdul, “Hadamard scfdma-a modified uplink
decreases the bandwidth efficiency. Smaller roll off gives transmission scheme with low papr and ser,” in Systems Conference
narrower bandwidth [19]. (SysCon), 2015 9th Annual IEEE International. IEEE, 2015, pp. 711–
715.
[12] B. Karakaya, H. Arslan, and H. Cirpan, “Channel estimation for lte
III. C ONCLUSION AND F UTURE W ORK uplink in high doppler spread,” in Wireless Communications and Net-
This paper reviewed recent developments in multiple ac- working Conference, 2008. WCNC 2008. IEEE, March 2008, pp. 1126–
1130.
cess technologies. Single Carrier Frequency Division Multiple [13] D. Kim, H.-M. Kim, and G.-H. Im, “Iterative channel estimation with
Access (SC-FDMA), is a novel method of radio transmission frequency replacement for sc-fdma systems,” Communications, IEEE
under consideration for deployment in future cellular systems. Transactions on, vol. 60, no. 7, pp. 1877–1888, July 2012.
[14] J. J. Sánchez-Sánchez, M. C. Aguayo-Torres, and U. Fernández-
It has similar performance and essentially the same overall Plazaola, “Ber analysis for zero-forcing sc-fdma over nakagami-m fading
complexity as orthogonal frequency division multiple access channels,” Vehicular Technology, IEEE Transactions on, vol. 60, no. 8,
(OFDMA). One prominent advantage over OFDMA is that pp. 4077–4081, 2011.
[15] S. Okuyama, K. Takeda, and F. Adachi, “Iterative mmse detection
the SC-FDMA signal has better PAPR characteristics because and interference cancellation for uplink sc-fdma mimo using harq,” in
of its inherent single carrier structure. SC-FDMA has drawn Communications (ICC), 2011 IEEE International Conference on. IEEE,
great attention as an attractive alternative to OFDMA, espe- 2011, pp. 1–5.
[16] J. Zhang, L.-L. Yang, and L. Hanzo, “Frequency-domain turbo equali-
cially in the uplink communications where better peak power sation in coded sc-fdma systems: Exit chart analysis and performance,”
characteristics greatly benefit the mobile terminal in terms of in Vehicular Technology Conference (VTC Fall), 2012 IEEE. IEEE,
transmit power efficiency and manufacturing cost. Because of 2012, pp. 1–5.
[17] Q. Wang, C. Yuan, J. Zhang, and Y. Li, “Frequency domain soft-
its merits, SC-FDMA has been chosen as the uplink multiple decision feedback equalization for sc-fdma with insufficient cyclic
access scheme in 3GPP Long Term Evolution (LTE). An prefix,” International Journal of Computer Science (IJCSI), vol. 9, no. 6,
attempt has been made to present a compact yet exhaustive pp. 103–108, 2012.
[18] H. G. Myung, K. Oh, J. Lim, and D. J. Goodman, “Channel-dependent
literature survey that will serve as standard reference for scheduling of an uplink sc-fdma system with imperfect channel informa-
researchers working in this area. tion,” in Wireless Communications and Networking Conference, 2008.
Further different precoding techniques may be checked for WCNC 2008. IEEE. IEEE, 2008, pp. 1860–1864.
[19] H. Myung, J. Lim, and D. Goodman, “Peak-to-average power ratio of
better performance. Multi input multi output (MIMO) may be single carrier fdma signals with pulse shaping,” in Personal, Indoor
used with SC-FDMA to improve throughput of the system. and Mobile Radio Communications, 2006 IEEE 17th International
Improved scheduling algorithm may be used for further per- Symposium on, Sept 2006, pp. 1–5.
formance enhancement.
R EFERENCES
[1] H. G. Myung and D. J. Goodman, Single Carrier FDMA: A New Air
Interface for Long Term Evolution. Wiley Publishing, 2008.

View publication stats