2017 International Conference on Intelligent Computing and Control (I2C2)

Design of Efficient Massive MIMO for 5G

Systems - Present and Past: A Review
M.Ramesh C.Geetha Priya V.P.M.B.Aarthi Alais
ECE Department, ECE Department, Ananthakirupa
Kamaraj College of Kamaraj College of ECE Department,
Engineering and Technology, Engineering and Technology, Mepco Schlenk Engineering
Virudhunagar, India Virudhunagar, India College, Sivakasi, India
mrameshme@gmail.com geethapriyaece@kamarajengg.edu.in vpmb2aarthi@ gmail.com

applications and the controller.

Abstract— In this survey paper, the ground work on Massive b. Cognitive Radio Network (CRN) based architectures: The
Multiple Input – Multiple Output (mMIMO) is concentrated by cognitive radio nodes named secondary users (SUs) exploit the
studying the various existing techniques in the system design for existing spectrum opportunistically by having the LEIRA
fifth generation wireless communications. The service quality of (learning, efficiency, intelligence, reliability, and adaptively)
5G systems can be enhanced in terms of throughput, latency, peak
property in the absence of licensed primary users (PUs)
data rate along with spectral and energy efficiency. Though there
exist, new methodologies, techniques and architectures including c. Device-to-device (D2D) communication architectures: D2D
Cognitive Radio Networks, D2D communications for designing communications allow close proximity user equipments (UEs)
and implementing 5G networks, the mMIMO utilizes spatial to communicate among one another on a licensed cellular
multiplexing for capacity improvement and remains robust across bandwidth without involving a Mobile Base Station (MBS) or
intentional jamming thereby playing a key aspect in the 5G with a very controlled involvement of a MBS. The challenged
system development. The efficient design of mMIMO still
here are Interference Management, Resource Allocation and
remains an open problem in the present research. This paper
analyses the existing techniques in the development of mMIMO delay-sensitive processing.
specifically concentrating on energy & spectral efficiency and d. Machine-to-machine (M2M) communication architectures:
discusses the challenges in the system design. It indicates the communication amidst devices without human
Index Terms— mMIMO, spectral efficiency, 5G, energy intervention. Examples include monitoring of buildings,
efficiency. intelligent transportation systems and health measurement.
Challenges are connectivity of massive devices, bursty data,
I. INTRODUCTION zero latency, scalability etc.,
e. Millimeter Waves (mmWave) communication: The current

T HE 5G communications aims at supporting infinite

capability of networking, big data bandwidth along with
extensive signal coverage in order to provide a good range of
band is unable to support huge UEs and so 30-300 GHz
frequency bands where mmWave communication, is proposed
for achieving high-speed data transfer. mmWave has several
highǦquality personalized services to the end users. Towards challenges at physical, MAC and network layers.
this aim, 5G communications will integrate multiple existing f. Massive MIMO (mMIMO): mMIMO systems as depicted in
advanced technologies with innovative new techniques. Fig. 1 are named as very large MIMO, large-scale antenna
The various approaches and techniques for 5G networks [1] systems, hyper-MIMO and full-dimension MIMO.
are discussed below.
a. Software-Defined Networking (SDN): SDN architectures
partition the network control functions and data forwarding
functions, in which network control functions are
programmable. SDN architectures consist of three modules,
as (i) software controller which holds functions such as
network management, APIs, network operating system. (ii) the
southbound part provides a protocol and an interface amidst
the controller and SDN-enable infrastructure. (iii) the
northbound part provides interface across various SDN
Fig. 1. Structure of MIMO Systems
Massive MIMO systems use arrays of antennas with hundreds
in number of antennas at MBS for parallely serving many UEs
with a single antenna. So expensive equipments are fixed on
 2017 International Conference on Intelligent Computing and Control (I2C2)

MBS. A mMIMO system is based on spatial multiplexing,

which again depends on the channel information at MBS, for
both Up Link (UL) and Down Link (DL) channels. A mMIMO
system minimizes energy and latency, develops a low complex
MAC layer, remains powerful against jamming, and improves
the capacity as a result of spatial multiplexing. However the
mMIMO has various implementation challenges [2] such as
energy/spectral efficiency, pilot overhead in channel
estimation, channel feedback, channel reciprocity,
architectures, hardware impairments etc., In this survey paper
we mainly concentrate on energy and spectral efficiency of
mMIMO systems.
This survey is organized as follows: Section II deals with the Fig. 2. Rate adaptive mode – IR System Model
related works on the energy efficient design of mMIMO
systems and their design issues. Section III discusses the The rate adaptive mode as depicted in Fig. 2 attains
several techniques used for achieving energy efficient improvement in the spectral efficiency and also maintains BER
mMIMO systems. Finally, section IV discusses the conclusion
reliability whereas in energy-conservative mode given in Fig. 3
with further scope for the researchers.
achieves a significant reduction in symbol-level reception
energy requirements.
II. RELATED WORKS

The categorization of MIMO schemes [3,4] includes two

approaches namely Spatial Multiplexing (SM) and Space Time
Block Coding (STBC). SM [3] can provide high data rates
whereas STBC [4] is having high transmission fidelity. Hence
there remains a pact between capacity and transmission
reliability extremeness. In [5,6], a switching criterion is
derived to decide a scheme which gives least probability of
error [5] or to chose a scheme which provides higher spectral
efficiency [6]. A Hybrid SM/STBC [7] is used as a MIMO
technique that blends STBC and SM by considering the
benefits of two schemes.
Another scheme called Incremental Redundancy (IR) is a Fig. 3. Energy conservative mode – IR System Model
Hybrid Automatic Repeat Request (HARQ) which provides
error control in wireless networks [8,9]. In this, upon receiving In conventional linear precoding and detection schemes,
negative acknowledgement (NACK) which indicates erroneous increase in count of antennas can harm energy efficiency (EE)
packet, and so the transmitter sends IR symbols to ensure due to the necessity of more radio frequency (RF) chains for
robustness of previously transmitted data and is repeated until the system. Hence SM [11-14] is suggested in which only a
a positive acknowledgement (ACK) is received. This IR can portion of antennas are simultaneously active in the system.
be adopted within MIMO systems where the system This in turn reduces the achievable rates. A normalized
incrementally shifts from SM to STBC or hybrid SM/STBC compressive sensing (CS) detection algorithm used for space
[10]. When negative acknowledgement is found after shift keying (SSK) and generalized space shift keying (GSSK)
decoding, the system incrementally switch towards diversity was proposed in [15] where the system encodes information
mode. The incremental diversity symbols are combined till a only in active antenna locations and hence a normalization of
positive acknowledgement is received. Incremental diversity channel matrix before applying greedy compressive detectors
can be operated in two modes: Rate-Adaptive and Energy can improve the system performance. Detection algorithms
Conservative. applied to compressive sensing scenarios are proposed in [16].
A message passing detection algorithm [17] is conditioned by
the frequency of message transfer between nodes. In addition it
is adopted to obtain reconstruction of sparse signals with high
complexity in compressive sensing. Hence a low complex
greedy CS algorithms [15] are used for large system dimension
for signal reconstruction.

Recent researches concentrate on Scaling Laws to develop

 2017 International Conference on Intelligent Computing and Control (I2C2)

energy efficient MIMO systems. Thus, analyzing the scaling

laws to provide EE by considering power consumption, pilot
contamination and errors in channel estimation with regard to
the number of antennas involved in each base station. Scaling
laws for Spectral Efficiency (SE) with regard to number of
antennas M are discussed in [18]. Transmit power is taken to
consideration [19] in MIMO to achieve higher EE than single
antenna system for the same data rate. Massive MIMO systems
provide energy efficiency than traditional MIMO [20]
considering the circuit power of RF links along with
complicated signal processing schemes. A large array gain and
multi user multiplexing gain [21] reduces the transmit power
for large number of antennas [22]. The energy efficiency
results [23] of mMIMO holds good when average channel gain
and power consumption is low. On considering the circuit
power consumption, the EE results of [24-27] shows a slighter Fig. 5. Efficient mMIMO System Design
increment and then decreases with increase in antennas M.
Scaling laws are used to show potential gain of mMIMO This paper studies the various design methodologies for
systems [28-30]. It mainly investigates uplink transmit power developing an efficient mMIMO system that fits for the 5G
scaling laws with reference to a given data rate. The environment along with the major challenges in its
experimentations of Ricean and Rayleigh channel models [28] implementation.
for the case with imperfect CSI is discussed. A single cell
downlink mMIMO supporting Zero forcing Beam forming
IV. CONCLUSION AND FUTURE SCOPE
(ZFBF) with perfect CSI is considered for energy efficient
design [31]. The analysis of closed form equations of
transmitted power along with maximal EE including their This paper provides an exhaustive survey on the existing EE
scaling laws with respect to M antennas proves that the pilot mMIMO system design with their preliminaries, state-of-art to
contamination is the limitation of mMIMO systems. Hence it explore an energy efficient MIMO system for 5G
can be mitigated with pilot decontamination methods [32]. communications. Additionally, it throws light on open
challenges in the development process. In future, research on
EE needs to be incorporated for multi-user and multi-cell cases
III. DISCUSSION
by considering the practical challenges like circuit energy
consumption, which is of great significance to the practical
Though several techniques and methods are available for 5G system design. As for the advanced techniques that will be
system design as given in Fig. 4, mMIMO is a promising used in the wireless communications, such as OFDMA,
technology to ensure the various requirements to be satisfied MIMO, and relay, existing research has proved that larger EE
with 5G systems. can be attained through energy-efficient design. However,
most work is still in the initial stage and more effort is needed
to investigate potential topics such as those listed in this
article.
Energy-Efficient MIMO Schemes in Multi-User and Multi-
Cell Scenarios: In multi-user and multi-cell environments, the
existence of inter-user and inter-cell interference complicates
the design of energy efficient MIMO systems. The idea of
maximizing EE by utilizing the spatial resource while
suppressing the interference is worth investigating. Complexity
reduction plays a major role in any system development.
Hence effective but simple algorithms are required to obtain a
Fig. 4. 5G System Design – Technologies good pact between complexity and performance.

In order to have an efficient mMIMO system, Spectral and/or ACKNOWLEDGMENT

Energy Efficiency may be concentrated much more than the We thank our teachers and mentors who directly or
others as given in Fig. 5. indirectly inculcated technical enthusiasm within us.
 2017 International Conference on Intelligent Computing and Control (I2C2)

REFERENCES [16] Narasimhan T., Raviteja P. and Chockalingam A., “Large-scale

multiuser SM-MIMO versus massive MIMO,” Information
Theory and Applications Workshop (ITA), pp. 1–9, Feb 2014.
[17] Donoho D. L., Maleki A. and Montanari A., “Message-passing
[1] Nisha Panwar, Shantanu Sharma and Awadhesh Kumar Singh, algorithms for compressed sensing,” Proceedings of the
“A survey on 5G: The next generation of mobile National Academy of Sciences, vol. 106, no. 45, pp. 18914–
communication,” Physical Communication, vol. 18, pp. 64-84, 18919, 2009.
2016. [18] Rusek F., Persson D., Lau B.K., Larsson E.G., Marzetta T.L.,
[2] Daniel C. Araujo, Taras Maksymyuk, Andre L.F. de Almeida, Edfors O. and Tufvesson F., “Scaling up MIMO: Opportunities
Tarcisio Maciel, Joao C.M. Mota, Minho Jo, ‘’ Massive MIMO: and challenges with very large arrays,” IEEE Signal Processing
Survey and Future Research Topics,” IET Communications. Magazine, vol. 30, no. 1, pp. 40–60, Jan. 2013.
[3] Chae C., Katz M., Suh C. and Jeong H., “Adaptive Spatial [19] Cardoso F.D. and Correia L.M., “MIMO gain and energy
Modulation for MIMO-OFDM,” IEEE Wireless efficiency in LTE,” Proc. IEEE WCNC, 2012.
Communications and Networking Conference, vol. 1, pp. 87-92, [20] Xu Z., Yang C., Li G.Y., Zhang S. and Xu S., “Energy-efficient
2004. configuration of spatial and frequency resources in MIMO-
[4] Lopez-Vicario J. and Anton-Haro C., "Adaptive Switching OFDMA systems,” IEEE Trans. Commun., vol. 61, no. 2, pp.
between Spatial Diversity and Multiplexing: A Cross-Layer 564–575, Feb. 2013.
Approach," IST Mobile Communications Summit, 2005. [21] Ngo H.Q., Larsson E.G. and Marzetta T.L., “Energy and
[5] Heath R.W. and Paulraj A.J., "Switching between Diversity and spectral efficiency of very large multiuser MIMO systems,”
Multiplexing in MIMO Systems," IEEE Transactions on IEEE Trans. Commun., vol. 61, no. 4, pp. 1436–1449, Apr.
Communications, vol. 53, pp. 962-968, 2005. 2013.
[6] Choi Y. and Alamouti S.M., "A Pragmatic PHY Abstraction [22] Larsson E.G., Edfors O., Tufvesson F. and Marzetta T.L.,
Technique for Link Adaptation and MIMO Switching," IEEE “Massive MIMO for next generation wireless systems,” IEEE
Journal on Selected Areas in Communications, vol. 26, pp. 960- Commun. Mag., vol. 52, no. 2, pp. 186–195, Feb. 2014.
971, 2008. [23] Mohammed S.K., “Impact of transceiver power consumption on
[7] Wu Y. and Tellambura C., "Low-Complexity Optimal Detection the energy efficiency of zero-forcing detector in massive MIMO
for Hybrid Space-Time Block Coding and Spatial systems,” IEEE Trans. Commun., vol. 62, no. 11, pp. 3874–
Multiplexing," Vehicular Technology Conference, Montreal, pp. 3890, Nov. 2014.
1-4, 25-28 September 2006. [24] Ha D., Lee K. and Kang J., “Energy efficiency analysis with
[8] Elshabrawy T. and Elsharkawy M. "Computationally Efficient circuit power consumption in massive MIMO systems,” Proc.
Implementation for Incremental Redundancy within Wireless IEEE PIMRC, 2013.
Networks," IEEE International Symposium on Personal Indoor [25] Xin Y., Wang D., Li J., Zhu H., Wang J. and You X., “Area
and Mobile Radio Communications, Toronto, pp. 1051-1055, spectral efficiency and area energy efficiency of massive MIMO
11-14 September 2011. cellular systems,” IEEE Trans. Veh. Technol., vol. 65, no. 5, pp.
[9] El-Shabrawy T. and Wahed, S.A., "Adaptive Modulation and 3243–3254, May 2016.
Coding for Broadcast DVB-H Systems," International [26] Kong C., Zhong C., Matthaiou M. and Zhang Z., “Performance
Symposium on Personal, Indoor and Mobile Radio of downlink massive MIMO in ricean fading channels with ZF
Communications, Tokyo, pp. 1292-1296, 13-16 September precoder,” Proc. IEEE ICC, 2015.
2009. [27] Yang H. and Marzetta T. L., “Energy efficient design of massive
[10] Elshabrawy T. and Nafae S. "An Adaptive MIMO System Using MIMO: How many antennas?” Proc. IEEE VTC, 2015.
Incremental Diversity," Vehicular Technology Conference, Las [28] Di Renzo M. and Haas H., “Bit error probability of space-shift
Vagas, pp. 1-5, 2013. keying MIMO over multiple-access independent fading
[11] Rusek F., Persson D., Lau B.K., Larsson E., Marzetta T., Edfors channels,” IEEE Trans.on Vehicular Technology, vol. 60, no. 8,
O. and Tufvesson F., “Scaling up MIMO: Opportunities and pp. 3694–3711, Oct 2011.
challenges with very large arrays,” IEEE Signal Processing [29] Narasimhan T., Raviteja P., and Chockalingam A., “Large-scale
Magazine, vol. 30, no. 1, pp. 40–60, 2013. multiuser SM-MIMO versus massive MIMO,” Information
[12] Wagner s., Couillet R., Debbah M.and Slock D.T.M., “Large Theory and Applications Workshop (ITA), pp. 1–9, Feb 2014.
system analysis of linear precoding in correlated MISO [30] Donoho D.L., Maleki A. and Montanari A., “Message-passing
broadcast channels under limited feedback,” IEEE Transactions algorithms for compressed sensing,” Proceedings of the
on Information Theory, vol. 58, no. 7, pp. 4509–4537, 2012. National Academy of Sciences, vol. 106, no. 45, pp. 18914–
[13] Masouros C., Sellathurai M. and Ratnarajah T., “Large-scale 18919, 2009.
MIMO transmitters in fixed physical spaces: The effect of [31] Bjornson E., Sanguinetti L., Hoydis J. and Debbah M., “Optimal
transmit correlation and mutual coupling,” IEEE Transactions design of energy-efficient multi-user MIMO systems: Is massive
on Communications, vol. 61, no. 7, pp. 2794–2804, July 2013. MIMO the answer?” IEEE Trans. Wireless Commun., vol. 14,
[14] Ha D., Lee K. and Kang J., “Energy efficiency analysis with no. 6, pp. 3059–3075, June 2015.
circuit power consumption in massive MIMO systems,” IEEE [32] Candes E. J. and Wakin M. B., “An introduction to compressive
24th International Symposium on Personal Indoor and Mobile sampling,” IEEE Signal Processing Magazine, vol. 25, no. 2, pp.
Radio Communications (PIMRC), pp. 938–942, Sept 2013. 21–30, 2008.
[15] Yu C.M., Hsieh S.H., Liang H.W., Lu C.S., Chung W.H., Kuo
S.Y. and Pei S.c., “Compressed sensing detector design for
space shift keying in MIMO systems,” IEEE Communications
Letters, vol. 16, no. 10, pp. 1556–1559, 2012.