Exploiting spatial domain to increase spectrum efficiency for wireless communications from source to media based modulation

This paper is aimed to provide a comprehensive review on the spectrally efficient transmission techniques applied in multiple antenna systems. Specifically, three state-of-the-art techniques which make use of the spatial domain to convey information bits, including the Vertical Bell-Labs Layered Space-Time (V-BLAST), the source-based spatial modulation (SM) and the media-based modulation (MBM), will be surveyed and their critical advantages as well as limitations will be highlighted.

EXPLOITING SPATIAL DOMAIN TO INCREASE SPECTRUM

EFFICIENCY FOR WIRELESS COMMUNICATIONS FROM

SOURCE TO MEDIA-BASED MODULATION (INVITED PAPER)

Tran Xuan Nam*

Abstract: Spectrum is a scarce resource for wireless communications During

the past decades there have been great research efforts in proposing various

transmission techniques which can increase the spectral efficiency of the wireless

systems This paper is aimed to provide a comprehensive review on the spectrally

efficient transmission techniques applied in multiple antenna systems Specifically,

three state-of-the-art techniques which make use of the spatial domain to convey

information bits, including the Vertical Bell-Labs Layered Space-Time (V-BLAST),

the source-based spatial modulation (SM) and the media-based modulation

(MBM), will be surveyed and their critical advantages as well as limitations will be

highlighted Finally, technical challenges and open research problems will be

given as a guideline for possible future advancement

Keywords: Wireless communications, MIMO, Spatial Modulation, MBM, Spectral efficiency

1 INTRODUCTION

Together with the growth of the information-centric society, wireless applications have become more popular During the last few decades we have witnessed a very fast development of wireless communications Today it is easy to find a wireless network nearby such as Bluetooth, Wi-Fi, television and mobile cellular communications Many users are equipped with several electronic devices such as smart phones, laptop computers and digital cameras which all can be easily connected to a wireless network The Internet

of Things (IoT) is a new network concept which describes a huge network connecting several billions electronic devices which can connect with one another through ubiquitous wireless networks

Increase in wireless networks and devices would result in a scarcity of carrier frequencies for transmission due to limited frequency spectrum Various efforts have been paid on the road to search for transmission solutions which can utilize spectrum more efficiently Traditional carrier digital modulation such as Phase Shift Keying (PSK) or Quadrature Amplitude Modulation (QAM) has been known as a common method to embed information bits into carrier waves Theoretically, it is possible to increase the number of bits into a carrier wave in order to achieve high spectrum efficiency For example, for 8-PSK modulation it is possible to embed 3 bits into a transmitted symbol to attain the spectrum efficiency of 3 bpcu1 or 3 bits/symbol Using 16-QAM modulation can increase the spectral efficiency to 4 bpcu More high order modulation like 256-QAM which often seen in the digital television systems can achieve the spectrum efficiency of 8 bpcu In general the spectral efficiency of the signal modulation schemes is given by:

2

log M

  bpcu, where M is the modulation order The straight implication here is that

increasing the modulation order would lead to improved spectral efficiency However, due to constrain in the transmit power, increasing modulation order means the constellation points are packed closer to each other As a result the increase in spectral efficiency of the signal modulation schemes is significantly affected by their error performance

In the past several decades, there have been large efforts by the information theory and

communications research community in order to propose novel transmission techniques to

increase the spectral efficiency of the wireless communications systems The aim of this

paper is to provide a comprehensive technical overview of the state-of-the-art transmission

techniques which exploit the spatial domain to convey information We first introduce the

concept of a spatial multiplexing in the multiple-input multiple-output (MIMO) system

We then present the principle of the so-called spatial modulation which uses the antenna

indexes to convey information bits Next, we will focus on the most advanced modulation

technique which is referred to as the media-based modulation Technical challenges and

open problems will be highlighted for possible future research

2 INCREASING SPECTRAL EFFICIENCY USING MIMO TRANSMISSION

Traditionally, the capacity of a noisy communication channel with bandwidth B Hz

and the signal-to-noise power ratio SNR is given by Shannon theorem as follows

C  B log (12  SNR) (bps) (1)

Under a static fading wireless communication channel with path gain h the capacity can

be expressed as

2 2

(bps) (2) The spectral efficiency is then given by

2 2



(bps/Hz) (3) Note that this spectral efficiency increases with SNR of the channel in the logarithmic

scale and thus soon becomes saturated when the transmit power is large enough

The Shannon capacity was limited for a very long time until the late of the 20th century

when Foschini and Gans in [2] and Telatar in another independent work [3] found the

capacity of a rich-scattering MIMO fading channel Given a wireless communication

system with Nt transmit antennas and Nr receive antennas, using the result of [2][3] the

spectral efficiency of the MIMO channel Hunder invariant condition is given by

2

2

SNR

/

SNR

H

t H

t

N

N















Where:   min( N Nr, t)

When the number of antennas is large enough we have

MIMO, 2

t

r

N

N N









  (4)

This means for a MIMO system with a large number of antennas the spectral efficiency is linearly increased with the employed antennas For this reason the spectral efficiency of the MIMO channel has been considered a breakthrough which can break the limit by the Shannon theorem

In order to make the MIMO system realistic, the authors in [4] proposed a MIMO architecture called Vertical-Bell Labs Layered Space-Time (V-BLAST) and implemented

it in a tested The proposed system multiplexes parallel data streams over multiple transmit antennas and uses linear detection combined with successive interference cancellation (SIC) to estimate transmitted symbols from spatial layers The V-BLAST spatial multiplexing system was demonstrated to achieve spectral efficiencies of 40-50 bps/Hz at the SNRs from 24-34 dB with reasonable detection complexity The V-BLAST system was then adopted as an air-interface standard for various state-of-the-art wireless communication systems such as 3GPP LTE/LTE-Advanced, WiMAX and Wi-Fi Significant efforts were also paid to invent other MIMO systems such as space-time codes [5][6], beam forming [7], however, these approaches aim at improving the transmission

quality of MIMO systems and thus are out of scope of the current paper Assume that M-PSK/M-QAM modulation is used for signal mapping the spectral efficiency of the V-BLAST system in terms of bpcu is given by

V-BLAST  Nt log2M (5)

Although the MIMO V-BLAST system could achieve significant improvement in spectral efficiency, it faces some critical problems Firstly, since multiple data streams are transmitted simultaneously over multiple antennas the problem of inter-channel interference (ICI) should be carefully resolved Although the combined linear estimation and SIC detector proposed in V-BLAST could treat ICI sensibly, the residual interference after each detection iteration still exists, which makes the bit error rate (BER) performance

of the V-BLAST system suboptimal Secondly, the V-BLAST system is very sensitive with the problem of Inter-Antenna Synchronization (IAS) Apart from these limitations the V-BLAST system also requires more radio frequency (RF) chains and thus is not energy efficient All these disadvantages of the V-BLAST system have opened an

opportunity for a novel transmission system which is called spatial modulation

4 SPATIAL MODULATION USING ANTENNA INDEXES

Spatial modulation is a novel transmission technique for wireless communications, which was originally proposed in [8] and then further developed in [9][11] The basic idea of spatial modulation is to exploit spatial domain, i.e antenna indexes to convey information bits When information bits are carried only by antenna indexes we have space-shift keying or generalized space-shift keying Whereas when information bits are conveyed by both antenna indexes and signal symbols we have spatial modulation or generalized spatial modulation

4.1 Space-Shift Keying/Generalized Space-Shift Keying

The Space-Shift Keying (SSK) [10] and Generalized Space-Shift Keying (GSSK) [11]

use only spatial domain for modulation A typical system configuration of SSK/GSSK is illustrated in Figure 1

1



1

r

N

1

n

r

N

n

2

n

t

N



1, , ,2 m

1 2

ˆ ˆ, , ,ˆ

m

y H

Figure 1 A typical SSK/GSSK system

In the SSK/GSSK system the transmit bit sequence b b1 2, , , bmdoes not modulate the

carrier wave but selects the transmit antennas For SSK systems only one transmit antenna,

i.e N a 1, is activated at an instant time for transmission Similarly, in GSSK systems

there are Na  1,( Nt  Na  1), active antennas used for transmission For

implementation convenience, the number of transmit antennasNtshould be a power of

two As an illustrative example, let us assume that we have 8 bits 00110110 to be

transmitted at the spectral efficiency of 2 bpcu In the SSK system, we need to use 4

transmit antennas and map each combination of 2 bits to a specific antenna Specifically,

the first 2 bits 00 is mapped to the first antenna, 11 to the fourth, 01 to the second, and 10

to the third Since each time an antenna is activated there are a combination of two bits

transmitted, the spectral efficiency is clearly 2 bpcu In general, the spectral efficiency of

SSK is given by

SSK  log2Nt (6)

Note that when higher spectral efficiency is required more transmit antennas need to be

used and the SSK system demands a complex antenna system In such a case, using GSSK

with an appropriate number of Naallows the transmitter to reduce the number of transmit

antenna For a GSSK system with Nt transmit and Na active antennas we have  t

a

N N

different active antenna combinations For example, let Nt  5, Na  2 we have 10

combinations of active antennas and we can select 8 out of the 10 for transmission The

achieved spectral efficiency is 3 bpcu If SSK is used the transmitter need to have 8

antennas so it is clear that GSSK can save 3 transmit antennas at the cost of an additional

RF unit In general, the achievable spectral efficiency of GSSK is given by:

GSSK 2

RF

log N t

N

   



 

(7)

Where:    denotes rounding down to the nearest power of two

The receiver in the SSK/GSSK uses an SSK/GSSK decoder to detect the activated antennas during each transmission period Specifically, upon reception of the received vector y the decoder finds the transmitted vector ˆxas follows:

ˆ arg min k F2

k



Where: xk denotes the transmit vector which contains 1s in the elements corresponding to

activated antennas and 0s elsewhere

4.2 Spatial Modulation/Generalized Spatial Modulation

Although they were independently proposed, spatial modulation (SM) [8] and generalized spatial modulation (GSM) [9] can be regarded as extensions of SSK and GSSK, respectively Figure 2 shows a typical configuration of SM/GSM systems

x

1

 1

r

N

1

n

r

N

n

2

n

t

N



bits

m

bits

m

y H

bits

a

m

bits

s

m

Figure 2 A typical SM/GSM system

The difference between SM and SSK and between GSM and GSSK is that the activated antennas transmit modulated symbols in SM/GSM systems but not in SSK/GSSK As shown in Figure 3, at the input to the SM/GSM transmitter m  ma  ms data bits are divided into two branches in which ma bits are used for spatial modulation (antenna selection) as in SSK/GSSK systems while the remaining msbits are used for signal

modulation as in the conventional wireless communication systems Either PSK or

M-QAM can be used for signal modulation For example, given a SM/GSM system which transmits signal at the spectral efficiency of 7 bpcu If SM is used and the transmitter has 8 antennas, m a 3bits can be used for spatial modulation In order to achieve 4 bpcu by signal modulation one can use 16-QAM In case GSM is used with N a 2 antennas and

5

t

N  transmit antennas it is possible to achieve 3 bpcu by spatial modulation while the remaining spectral efficiency of 4 bpcu can be obtained by 16-QAM modulation

In general, the spectral efficiency of SM/GSM is given, respectively, by:

RF

t t

N

M N





   

   



(9)

At the receiver, in order to detect the transmitted bits the SM/GSM decoder needs to

perform joint estimation of both the activated antennas and modulated symbols Denote

,

k i the index of transmit a receive antennas, respectively Also, let qdenote the index of

a modulated symbol from the signal alphabet The joint estimation performed by the

decoder is given by

,

ˆ ˆ, arg max | , arg min r

N



Note that although they can achieve higher spectral efficiency, SM and GSM suffers

from error performance degradation due to joint detection compared with SSK and GSSK

4.3 High-Rate Spatial Modulation

High rate spatial modulation (HR-SM) [12] is an enhanced spatial modulation scheme

which achieves higher spectral efficiency over GSM systems In the HR-SM scheme all

transmit antennas are activated for transmission, i.e Na  Nt Configuration of the

HR-SM scheme is shown in Figure 3

x

1

 1

r

N

1

n

r

N

n

2

n

t

N



bits

m

bits

m

y H

bits

a

m

bits

s

m

s

x

Figure 3 Configuration of HR-SM system

The idea of the HR-SM scheme is not simply to send the transmit vector xbut

optimize it before transmission By decomposing the transmit as x   s x wherex

denotes the modulated symbol while s is a complex vector which is referred to as the

spatial constellation codeword In the GSM scheme each element of s has the following

properties:s  s={0,1}, sum( ) s  Na In the HR-SM scheme the authors design the

SC code words s such that s  s={ 1,   j }, sum( ) s  Na  Nt

Since the SC code words receive complex values for an HR-SM system with Nt

transmit antennas there are 4N t

different combinations of the SC vectors However, in order to achieve full diversity the first element of s is fixed to 1 As a result there are

1

4N  t

combinations ofs, which can be used for spatial modulation The authors proposed

to select 2N  t 1

combinations which satisfy the minimum Euclidean distance as the SC code words As a result, the spectral efficiency of the HR-SM is given by

HR-SM  2( Nt  1)  log2M (10)

It is noted that the achievable spectral efficiency of the HR-SM is linearly increased with the number of transmit antennas and thus much higher than that of the previous SSK, GSSK, SM, and GSM schemes

In order to detect the transmitted bits, the HR-SM decoder uses a reduced-complexity maximum-likelihood detector which can achieve optimal bit error rate performance

Figure 4 compares the spectral efficiency of various spatial modulation schemes including: V-BLAST, SSK, GSSK, SM, GSM and HR-SM for Na  2, M  16

Figure 5 Spectral efficiency of spatial modulation schemes:Na  2, M  16

We can see from the figure that the V-BLAST scheme achieves the highest spectral efficiency Next, the HR-SM scheme outperforms all remaining schemes due to the linear increase of spectral efficiency similar to V-BLAST Among the other schemes, SM and GSM are superior to SSK and GSSK respectively At N t 24 the spectral efficiency of

SM is shown to coincide with that of GSSK This is a special case where both the total spectral efficiencies due to spatial modulation and signal modulation of SM equal the spectral efficiency due to spatial modulation of GSSK In reality, there may be other similar cases Moreover, when designing an SM system for a given spectral efficiency it is possible to choose either of the above schemes according to the specific requirements and constrains such as the number of transmit antennas, the number of employed RF chains and signal modulation For example, given the spectral efficiency of 4 bpcu, one can flexibly choose: V-BLAST with N t 4and BPSK, SSK withN t 16, GSSK with

N  N  , SM with N t 4 and 4-QAM, GSM with Nt  4, Na  2 and 4-PSK, and HR-SM with Nt  2, Na  2and 4-QAM

4.4 Other spatial modulation schemes

Apart from the above mentioned schemes, many other works have also been successful

in inventing spatial modulation schemes with different types of merits such as obtaining

4 6 8 10 12 14 16 18 20 22 24 0

10 20 30 40 50 60 70 80 90 100

Number of Transmit Antennas

V-BLAST SSK GSSK SM GSM HR-SM

full diversity [13], low computational complexity, low power consumption A more

detailed review on these schemes can be found in a tutorial review in [14]

5 MEDIA-BASED MODULATION USING PERTURBED CHANNELS

Media-based modulation (MBM) [15]-[17] is the latest technology related to spatial

modulation The idea of MBM is similar to that of SM in which each unique channel is

utilized to convey signal symbols In SM systems each transmission channel corresponds

to the natural wireless channel between the receiver and an activated transmit antenna

Under assumption that the propagation channel is affected by rich scattering multipath

fading and that the space between transmit antennas is large enough the channels from

transmit antennas are independent and uncorrelated This allows the SM transmitter to

utilize unique channels to bear information bits in the form of antenna index In the MBM

system, however, antenna index is not used as the transmitter has only single antenna In

contrast, unique channels are created by perturbing propagation environment intentionally

by RF elements called RF mirrors Figure 5 presents a typical configuration of an MBM

system As shown in the figure the RF mirrors with K elements are placed around a dipole

antenna to generate unique channel states to convey information bits WithKmirrors

which can be turned on or off it is possible to generate 2Kchannel states to achieve

spectral efficiency ofK bpcu Theses channel states are selected by madata bits Together

with signal modulation the achievable spectral efficiency of the MBM system is given by

MBM  K  log2M (11)

It is also interesting to note that the spectral efficiency of MBM is linearly increased

with the number of RF mirrors The more RF mirrors are used the higher spectral

efficiency can be achieved However, the difficulty lies in designing the RF mirrors so that

they can generate unique channel states but not too complex in size and estimating channel

states A design example of a transmit antenna with RF mirrors can be found in [16] More

discussions about channel estimation for MBM are provided in [17]



1

1

n

r

N

n

2

2

n

bits

m

y

bits

a

bits

s

m

Figure 5 Configuration of MBM

6 TECHNICAL CHALLENGES AND OPEN RESEARCH PROBLEMS

While the V-BLAST system has been well researched and implemented in various

advanced wireless systems such as Wi-Fi, WiMAX or 3GPP LTE, the implementation of

the SM and MBM still requires further investigations For the SM systems including

transmission efficiency and error performance need further efforts For hardware implementation, RF switches which can smoothly switch from one antenna to another are expected to be soon completed For the MBM system, as cited above the design of the RF mirrors is challenging and needs further proposals Similar to the SM, channel estimation

is also an interesting topic for research

7 CONCLUSION

Increasing spectral efficiency is an important task for making efficient use of radio spectrum in order to meet the increasing demand in transmission rate of advanced wireless communications systems In this paper, various state-of-the-art transmission techniques which can increase the spectral efficiency have been introduced The V-BLAST system has the highest spectral efficiency, however, has significant problems with IAS and ICI The source-based SM systems can alleviate those problems faced by V-BLAST but at the cost of reduced spectral efficiency The MBM system is similar to the SM system but may face difficulty in fabricating the antenna system with RF mirrors Thus the employment of which system for practical implementation depends on the requirements for each specific case

REFERENCES

[1] T X Nam và L M Tuấn, “Xử lý tín hiệu không gian-thời gian,” NXB Khoa học Kỹ

thuật (2013)

[2] G J Foschini and M J Gans "On limits of wireless communications in a fading

environment when using multiple antennas." Wireless personal communications, Vol

6, No.3 (1998), pp 311-335

[3] E Telatar, “Capacity of Multi-antenna Gaussian Channels." European Trans on Telecom., Vol 10, No 6 (1999), pp 585-595

[4] P W Wolniansky, G J Foschini, G D Golden and R A Valenzuela, "V-BLAST: an

architecture for realizing very high data rates over the rich-scattering wireless channel," Proc 1998 URSI Int’l Symp on Signals, Systems, and Electronics, Pisa,

(1998), pp 295-300

[5] V Tarokh, N Seshadri, and A R Calderbank "Space-time codes for high data rate

wireless communication: Performance criterion and code construction," IEEE Trans

on Infor Theory, Vol 44, no.2 (1998), pp 744-765

[6] S M Alamouti, "A simple transmit diversity technique for wireless communications," IEEE J on Select Areas in Commun., Vol 16, No.8 (1998), pp 1451-1458

[7] D J Love, R.W Heath, and T Strohmer "Grassmannian beamforming for

multiple-input multiple-output wireless systems," IEEE Trans on Inform Theory, Vol.49,

No.10 (2003), pp 2735-2747

[8] R Y Mesleh, H Haas, S Sinanovic, C W Ahn and S Yun, "Spatial modulation,"

IEEE Trans on Vehicular Tech., Vol 57, No 4 (2008), pp 2228-2241

[9] A Younis, N Serafimovski, R Mesleh and H Haas, "Generalised spatial

modulation," in Proc 2010 Conference Record of the 44th Asilomar Conf on Signals, Systems and Computers, Pacific Grove, CA, (2010), pp 1498-1502

[10] J Jeganathan, A Ghrayeb, L Szczecinski and A Ceron, "Space shift keying

modulation for MIMO channels," IEEE Trans on Wireless Commun., Vol 8, No 7

(2009), pp 3692-3703

[11] J Jeganathan, A Ghrayeb and L Szczecinski, "Generalized space shift keying

modulation for MIMO channels," in Proc 2008 IEEE 19th Int’l Symp on Personal,

[12] T P Nguyen, M T Le, V D Ngo, X N Tran and H W Choi, "Spatial Modulation

for High-Rate Transmission Systems," in Proc 2014 IEEE 79th Vehicular Tech

Conf (VTC Spring), Seoul, (2014), pp 1-5

[13] M.-T Le, V.-D Ngo, H.-A Mai, X N Tran, and M Di Renzo, "Spatially modulated

orthogonal space-time block codes with non-vanishing determinants," IEEE Trans on

Commun., Vol 62, No 1, (2014), pp 85-99

[14] M Di Renzo, H Haas, A Ghrayeb, S Sugiura and L Hanzo, "Spatial Modulation for

Generalized MIMO: Challenges, Opportunities, and Implementation," Proceedings of

the IEEE, Vol 102, No 1 (2014), pp 56-103

[15] A K Khandani, "Media-based modulation: A new approach to wireless

transmission," in Proc 2013 IEEE Int’l Symp on Infor Theory, Istanbul, (2013), pp

3050-3054

[16] A K Khandani, "Media-based modulation: Converting static Rayleigh fading to

AWGN," in Proc 2014 IEEE Int’l Symp on Infor Theory, Honolulu, (2014), pp

1549-1553

[17] E Seifi, M Atamanesh, and A K Khandani "Media-Based MIMO: A New Frontier

in Wireless Communications," arXiv preprint arXiv:1507.07516 (2015)

TÓM TẮT

TĂNG HIỆU SUẤT SỬ DỤNG PHỔ NHỜ KHAI THÁC MIỀN KHÔNG GIAN

CHO THÔNG TIN VÔ TUYẾN TỪ ĐIỀU CHẾ TẠI NGUỒN ĐẾN ĐIỀU CHẾ

TẠI MÔI TRƯỜNG (BÀI BÁO MỜI)

Phổ tần là một nguồn tài nguyên khan hiếm cho thông tin vô tuyến Trong vài

thập kỷ vừa qua đã có nhiều nỗ lực trong việc tìm kiếm các kỹ thuật truyền dẫn có

khả năng tăng hiệu suất sử dụng phổ trong các hệ thống vô tuyến Bài báo này

cung cấp một đánh giá tổng quan về các kỹ thuật truyền dẫn hiệu quả về phổ tần

trong các hệ thống đa ăng-ten Cụ thể, ba kỹ thuật tiên tiến nhất sử dụng miền

không gian để chuyển tải các bit thông tin sẽ được phân tích gồm V-BLAST, kỹ

thuật điều chế không gian tại nguồn và kỹ thuật điều chế tại môi trường Các ưu

điểm cũng như các hạn chế quan trọng của các kỹ thuật này sẽ được làm sáng tỏ

Cuối cùng, các thách thức và vấn đề mở sẽ được đưa ra như một định hướng cho

các nghiên cứu tương lai

Từ khóa: Thông tin vô tuyến, MIMO, Điều chế không gian, Điều chế tại môi trường, Hiệu suất sử dụng phổ

Author affiliations:

Military Technical Academy ;

*Corresponding author: namtx@mta.edu.vn