Increasing Channel Capacity Using MIMO Spatial Modulation Technique

DOI : 10.17577/IJERTV2IS4739

Download Full-Text PDF Cite this Publication

Text Only Version

Increasing Channel Capacity Using MIMO Spatial Modulation Technique

Increasing Channel Capacity Using MIMO Spatial Modulation Technique

Dayanand Sahu

Lecturer, Department of Electronics & Telecommunication Engineering Govt. Polytechnic, Janjgir (C.G.) 495668, INDIA

Ranbir Kumar Paul & Nitin Jain

Asstt. Professor, Department of Electronics & Telecommunication Engineering Chouksey Engineering College, Bilaspur (C.G.) 495001, INDIA

Index Terms – Channel, Coding, communication, Diversity, Multi Antenna, Receiver, transmitter, Wireless.

MIMO communications systems can exploit spatial multiplexing (SM) approach to increase the channel capacity and improve spectral efficiency as well. Therefore, the MIMO SM-based system is one of currently promising techniques that can achieve high- speed wireless communications networks. In MIMO SM-based systems, independent data streams are transmitted from sufficiently- separated antennas. These results in a linear increase in the channel capacity proportional to the minimum number of receive and transmit antennas. However, MIMO SM- based system requires powerful signal processing procedures at the receiver to efficiently recover the signal transmitted

from the multiple antennas, and hence to explore the advantages of MIMO systems. Therefore, the potential advantages of MIMO system can be guaranteed and the wireless system will work in the best possible way. Some special detection techniques have been proposed in the literature in order to exploit the high spectral capacity offered by MIMO systems.

In this Paper, we consider a conventional MIMO SM system with Nt transmit antennas and Nr receive antennas where Nt Nr as shown in Figure 1. Independent data streams a, b, and c, are encoded and modulated before being

transmitted. Herein, consider a transmitted vector x = [x1, x2 xNt]T whose elements are drawn independently from a complex constellation set , e.g. Quadrature Amplitude Modulation (QAM) constellation. The vector is then transmitted

Figure 1 SM System Model Including Transmitter and Receiver

Where the elements of the vector n = [n1, n2 nsNr]T are drawn from independent and identically distributed (i.i.d.) circular symmetric Gaussian random variables. The system model of (1) is then given in the matrix form as following.

via a MIMO channel characterized by the channel matrix H whose element hi,j CN (0, 1)1 is the complex channel coefficient between the jth transmit and ith receive antennas. The received vector r = [r1, r2 rNr]T can then be given as following,

Successive Interference Cancellation (SIC) (for instance, V-BLAST decoder).

The Vertical Bell Laboratories Layered Space Time (V-BLAST) scheme was originally proposed by Foschini [1] and has been discussed in details in literature. The main idea behind the V-BLAST architecture (i.e., transmitter) is to demultiplex the data stream into several sub- streams and transmit them simultaneously. At the receiver side, each antenna observes all the transmitted signals, which are mixed due to the environment surrounding the wireless propagation channel. V-BLAST detection algorithm detects the signals one after another in an iterative way. The construction of the filtering matrix can still be based on any of the aforementioned linear criteria, i.e. ZF or MMSE.

r1 (k )

p1

pNt x1 (k )

n1 (k )

The V-BLAST algorithm utilizes the

already detected symbol xi, obtained by the ZF or MMSE filtering matrix, to generate a

rNr (k )

(1)

hNr1

hNrNt xNr (k )

nNr (k )

modified received vector with xi cancelled out. Thus the modified received vector becomes with fewer interferers and better

  1. Although linear detection techniques

    performance due to a higher level of diversity. The algorithm continues until all Nt symbols being detected.

    If we rewrite the system in (1) into a matrix form with Nt = Nr = 4,

    are easy to implement, they lead to high

    r1

    p1

    p2

    p3

    p4 x1

    n1

    degradation in the achieved diversity order

    r h h

    h h x

    n

    due to the linear filtering. Another approach

    2 21 22

    23 24

    2

    2 (2)

    that takes advantage of the diversity

    r3

    p1

    p2

    p3

    p4 x3

    n3

    potential of the additional receive antennas, uses nonlinear techniques such as

    r h h

    r h h

    4 41 42

    h h x

    h h x

    43 44 4

    n

    n

    4

    Then, using ZF or MMSE criterion, the estimate of xi can be calculated. Assuming that this symbol is correct, it is

    Euclidean norm, corresponds to the required component of x. That is,

    weighted with its corresponding channel coefficient and then subtracted from the received vector r. The new modified vector y becomes:

    k1 arg j

    min g 2

    ,

    ,

    j

    j

    ~x g

    r (1) ,

    y1

    p2

    p3

    p4

    n1

    k1 k1

    y h

    h h x2

    n

    3

    3

    2 22

    23 24 x

    2

    (3)

    y3

    p2

    p3

    p4

    n3

    and,

    y h

    h h x4

    n

    4 42

    43 44

    4

    xk1

    Q(~x ),

    k

    k

    1

    Iteratively, the nulling matrix is computed. The newly detected symbol xi is subtracted of the already modified received vector y to produce the following equations:

    where gj is the jth row of the filtering matrix G, Q ( . ) is the demodulation function, and the superscript is the iteration index. At the first iteration, r(1) = r and G(1) = H . At the

    y1

    y

    p3 p4

    h h x

    n1

    n

    end of the first iteration, the interference due

    to the k th component of x is cancelled out as

    2

    23 24

    3 2

    1

    (4)

    y

    h h

    x

    n

    follows:

    3

    33

    34 4

    3

    y4 mod

    h43

    h44

    n4

    r(2) r(1) x h ,

    k

    k

    k1 1

    k1 1

    k

    k

    1 1

    Definitely the diversity level is getting better at each stage of detection and the performance is improved because the

    H (2) H (1)k1 …,h

    1

    1

    , hk 1

    ,…

    equations become more than unknowns. This method of successive interference cancellation is continued until all Nt symbols are detected.

        1. The Zero-Forcing V-BLAST algorithm (ZF-VBLAST) is based on detecting the components of x one by one. For the first decision, the pseudo-inverse, i.e., G equals H , of the matrix H is obtained. Assume that the noise components are i.i.d. and that the noise is independent of

          x. Then, the row of G, with the least

          Doing so until detecting the last element of x. When the sorting step is discarded, the code is called Unsorted ZF- VBLAST or ZF-VBLAST.

          Obviously, incorrect symbol detection in the early stages will create errors in the following stages; i.e. error propagation. This is a severe problem with cancellation based detection techniques particularly when the number of transmit an receive antennas are the same. The first detected symbol's performance is quite poor as it has no diversity. To reduce the effect of error propagation and to optimize the performance of VBLAST technique, it has

          been shown in that the order of detection can increase the performance considerably. By detecting the symbols with largest channel coefficient magnitude first, the effect of the noise vector producing an incorrect symbol can be reduced, and reducing error propagation as result.

          In order to achieve best performance, it is optimal to start detecting the components of x that suffer the least noise amplification i.e the layer with the largest SNR. Then sorting step in the code will be activated. This algorithm is called sorted Zero-Forcing VBLAST (SZF-VB).

          ZF is the simple linear receiver with low computational complexity and suffers from noise enhancement. But it can works well at high SNR. However, in Zero- Forcing we can choose any row of Gi to null the signal from the ith transmit antenna, while in ZF-VBLAST it was shown that it is best to start with the signal that has the

          greatest signal to noise ratio (SNR) in which is known by ordering, which results in a better performance as seen above. The ZF solution in general is an easier solution but not optimum as it enhances the noise. Instead we have used the MMSE method, which gives us better performance.

    In section 3.1, it was shown that MMSE algorithm suppresses both the interference and noise components, whereas the ZF algorithm removes only the interference components. This implies that the mean square error between the transmitted symbols and the estimate of the receiver is minimized. Therefore, MMSE is superior to ZF in the presence of noise. The MMSE filtering strategy can be used with

    VBLAST, where the resulting detector is referred to as the MMSE-VBLAST detector.

    Also, we refer to the MMSE- VBLAST as the Unsorted MMSE-VB when the sorting stage is skipped. In this case, the components of x are detected in an ascending order.

    The MMSE-VB detection algorithm can be obtained by the MMSE criterion in constructing the filtering matrix.

    Figure 2 BER of VBLAST Detection Schemes

    The main drawback of the VBLAST detection algorithms lies in the computational complexity, because multiple calculations of the pseudo-inverse of the channel matrix are required .

    Figure 2 shows the performance of various VBLAST detection schemes that utilizing both ZF and MMSE criteria with and without using optimal ordering. Comparing the simulation results of ZF- VBLAST and MMSE-VBLAST separately, the sorted detection schemes achieve an improved performance in comparison to the unsorted ones. At a target BER of 10-3 the difference between ZF-VBLAST curves is

    about 4 dB and the difference between MMSE-VBLAST curves is about 7 dB. This demonstrates the impact of employing signal ordering. Note that the performance advantage of the MMSE is quite considerable in all cases. The sorted MMSEVBLAST lags the MLD curve by about 6.7 dB at a target BER of 10-4.

  2. Figure 3 ZF Receiving Technique Under Different Modulation

    The above graph represents the behavior of BPSK modulation, QPSK modulation,QAM16 modulation,QAM64 modulation scheme under ZF receiving algorithm, here comparison has been made between BPSK,QPSK,QAM16,QAM64

    ,and result shows that BPSK give better ber performance compare to QPSK,QAM16,QAM64 accordingly.BER is worst in QAM64.The simulation model is implemented for MIMO Spatial Multiplexing V BLAST technique.

    Figure 4: MMSE Receiving Technique Under Different Modulation

    The above graph represents the behaviour of BPSK modulation,QPSK modulation,QAM16 modulation,QAM64 modulation scheme under MMSE receving algorithm,here comparision has been made between BPSK,QPSK,QAM16,QAM64

    ,and result shows that BPSKgive better ber performance compare to QPSK,QAM16,QAM64 accordingly.BER is worst in QAM64.The simulation model is implemented for MIMO Spatial Multiplexing V BLAST technique.

    Figure 5 ML Receiving Technique Under Different Modulation

    The above graph represents the behavior of BPSK modulation, QPSK modulation,QAM16 modulation,QAM64 modulation scheme under ML receiving algorithm, here comparison has been made between BPSK,QPSK,QAM16,QAM64

    ,and result shows that BPSK give better ber performance compare to QPSK,QAM16,QAM64 accordingly.BER is worst in QAM64.The simulation model is implemented for MIMO Spatial Multiplexing V BLAST technique.

In recent years, MIMO wireless communication systems have exploited spatial multiplexing (SM) approach to increase the channel capacity and improve spectral efficiency as well. Therefore, the

MIMO SM-based system has been one of currently promising techniques that could realize Gbps high-speed wireless transmission for future communications networks. The main challenge of MIMO SM-based system resides in designing signal processing techniques, i.e., detection techniques. Those are capable of separating the parallel transmitted signals with acceptable computational complexity and achieved performance. An intensive work is being done in this field to investigate several MIMO SM detection techniques such linear, nonlinear and tree based detections. In this study, several MIMO detection techniques have been successfully described, analyzed and compared. In general, linear detection techniques such as ZF and MMSE have an efficient computational complexity; however, the BER performance plots of these techniques demonstrated their relatively poor performance. In an attempt to improve the poor performance of the linear detections, VBLAST have been proposed. It was shown that the ordering strategy over Successive Interference Cancellation (VBLAST) has important benefits. This strategy was applied to the general V- BLAST code and got a higher performance gain. However, performance improvement with SIC techniques is limited due to error propagation, particularly with the same number of transmitter antennas as receiver antennas. Additionally, the main drawback of the VBLAST detection algorithms lies in the computational complexity, because multiple calculations of the pseudo-inverse of the channel matrix are required. This involves expensive computational requirements and makes VBLAST algorithms enduring computational bottleneck.

  1. P. W. Wolniansky, G. J. Foschini, G.

    D. Golden, and R. A. Valenzuela, VBLAST: An architecture for realizing very high data rates over the rich-scattering wireless channel," URSI Int. Symp. on Signals, Systems and Electronics, (Pisa, Italy), pp. 295-300, Sep. 1998.

  2. G.J.Foschini. and M.J.Gans, On the Limits of Wireless Communication in a Fading Environment When Using Multiple Antennas, Wireless Personal Communications, vol. 6, no. 3, pp. 311-355, 1998.

  3. G. J. Foschini, Layered space-time architecture for wireless communication in a fading environment when using multi- element antennas," Bell Labs Tech. J., vol. 1, no. 2, pp. 41-59, 1996.

  4. Jeffrey G.Andrews, Arunabha Ghosh, and Rais Mohamed, Fundamentals of WiMAX: Understanding Broadband Wireless Networking, Prentice Hall, 2007.

  5. S. M. Alamouti, A simple transmit diversity technique for wireless communications, IEEE Journal on Select Areas in Communications, vol.16, pp.14511458, August 1998.

  6. V. Tarokh, H. Jafarkhani, and A. Calderbank, Space-time block codes from orthogonal designs, IEEE Trans. On Information Theory, vol. 45, no. 5, pp. 1456-1467, July 1999.

  7. A. Kaye and D. George, Transmission of multiplexed PAM signals over multiple channel and diversity systems, IEEE Trans. On

    Communication Technology, vol. 18, no. 5, pp. 520-526, October 1970.

  8. W. Van Etten, An optimum linear receiver for multiple channel digital transmission systems, IEEE Trans. On Communications, vol. 23, pp. 828-834, August 1975.

  9. Simon Haykin, and Michael Moher, "Modern Wireless

    Communications", Prentice-Hall Inc., 2005.

  10. Choo C. Chiau, "Study of the Diversity Antenna Array for the MIMO Wireless Communication Systems ", PHD thesis, University of London, 2006.

  11. E. Telatar, Capacity of multi- antenna Gaussian channels, European Trans. on Telecommunications, vol. 10, pp. 585-595, Dec. 1999.

  12. Ezio Biglieri et al.," MIMO Wireless Communications", Cambridge University Press, 2007.

  13. Z. Xu and R. D. Murch, Performance analysis of maximum likelihood detection in a MIMO antenna system, IEEE Trans. Commun., vol. 50, no. 2, pp. 187 191, Feb. 2002

  14. G. D. Golden, G.J. Foschini, R. A. Valenzuela, and P. W. Wolniansky, Detection algorithm and initial laboratory results using V- BLAST space- time communication architecture, Electron. Lett., vol. 35, pp. 14 16, Jan. 1999.

  15. D. Gore, R. W. Heath Jr., and A. Paulraj, On performance of the zero forcing receiver in presence of transmit correlation, in Proc. Int. Symp. Inf. Theory, 2002, p. 159.

Leave a Reply