Adaptive Multiantenna Technology

This work was supported by the National Science and Technology Major Project of the Ministry of Science and Technology of China under Grant No. 2009ZX03003.

    Multiantenna technology is wireless communication technology that deploys multiple antennas at transmitters and receivers. In recent years, multiantenna technology has become an important research issue. Using this technology, power, spatial diversity, spatial multiplexing, and array gain can be achieved while interference [1] is suppressed. Thus, a system’s coverage can be enlarged, and link stability and transmission rate can be improved without the need to increase cost too much.

    Multiantenna technology works in different ways, for example, Beamforming (BF) [2], Cyclic Delay Diversity(CDD)[3], Spatial Diversity(SD)[4]-[6], Spatial Multiplexing(SM)[7] and a combination of these. Different multiantenna technologies are fit for specific applications. Because the channel environment and location of the receiver change, a single multiantenna technology alone cannot maximize a system’s performance. So it is necessary to switch between multiantenna modes to adapt to the ever-changing channel environment.

1 Introduction to Multiantenna Technology Modes
    Every mode has its own characteristics: 
    (1) SD technology
    SD uses signal redundancy to achieve diversity. The transmitter gains diversity by sending orthogonal information set at two different timeslots from two different antennas.

    (2) SM technology
    In SM, different data is sent on the same time-frequency resource from different antennas so that spectrum efficiency is multiplied without expending more frequency resource. Fig. 1 illustrates the SM technology. SD and SM are usually referred to as Multi-Input Multi-Output (MIMO) technology.

    (3) BF technology
    BF is based on an adaptive antenna rationale and uses the antenna array and advanced signal processing algorithm to perform weighted processing on every physical antenna. As shown in Fig. 2, the transmitter performs weighted processing on data stream S1 and sends it out. As far as the receiver is concerned, the whole antenna array works like a virtual antenna. After weighted processing, the antenna array forms a narrow transmit beam aimed at the target receiver and forms a null point to the direction of the interference reception.

    (4) MIMO+BF technology
    Because only one data stream is transmitted at a time, multiplex is not gained with BF technology. Especially when channel quality is good, transmission rate is not obviously increased with BF. To further improve transmission rate, BF can be combined with MIMO [8], [9]. Combined SD and BF is called SD+BF, and combined SM and BF is called SM+BF. Fig. 3 shows one of the working principles. The four physical antennas at the transmitter are divided into two sub-arrays. On each sub-array, one virtual antenna or beam is formed with BF. Two beams constitute SD or SM.

    (5) CDD technology
    CDD is an often-used multiantenna transmit diversity scheme of an Orthogonal Frequency Division Multiplexing (OFDM) system [10]. In CDD, the same frequency domain data is sent on each physical antenna, and different cyclic delays are performed over the OFDM sign of the time domain so that diversity is gained in the frequency domain. Fig. 4 shows the transmitter. The time domain data stream S1 performs separate cyclic delay δi on each physical antenna and then sends them out. δi is the amount of cyclic delay, i =1,2,3,4, and δ1 is usually 0. At the end of CDD processing, the whole antenna array is seen at the receiver as one virtual antenna.

    (6) CDD + MIMO technology
    Because only one data stream is sent at a time, CDD can be combined with MIMO when the channel condition is good in order to boost transmission rate [11], [12]. SD and CDD can be combined as SD+CDD and SM and CDD can be combined as SM+CDD. Fig. 5 shows one of the working principles. The four physical antennas at the transmitter are divided into two sub-arrays.

2 Comparison of Multiantenna Technologies
    (1) Data transmission format
    Data sent over physical antennas with diffferent multiantenna technologies is different. Take IEEE 802.16e [4] four antennas for example. Table 1 lists the frequency domain data stream sent by every physical antenna in two neighbor signs and on the same data sub-carrier. SM uses BLAST coding [7], SD uses Alamouti coding, and redundancy is introduced between two Orthogonal Frequency Division Multiple Access (OFDMA) signs. k data sub-carriers on i transmit antennas correspond to BF with weighted value of wi (k), i =1,2,3,4. In addition, cyclic delay performed on time domain data is equivalent to the frequency domain data multiplied by a phase rotation

    where coefficient 0.5 is the power normalization factor, NF is the point number of Inverse Fast Fourier Transform (IFFT), k is the sub-carrier index, δi  is the cyclic delay amount of CDD, and i =1,2,3,4,.The data stream is S1,S2,S3,S...

    (2) Characteristics
    Usually BF, SD+BF, and SM+BF need to dynamically adjust weighted value depending on the channel status information, which belongs to closed-loop technology. It is also necessary to perform BF on pilot, and therefore a dedicated pilot should be supported. CDD, SD+CDD, and
SM+CDD can work when the status of the channel to the transmitter is unknown. This belongs to open-loop technology. SM+BF and SM+CDD can be used to transmit different data streams over different virtual antennas, and if the channel conditions are good, the system's transmission capability can be enhanced to support high speed data transmission. However, in BF, SD+BF, CDD, and SD+CDD redundancy must be introduced to the spatial dimension in order to gain diversity, which boosts link stability and coverage. In addition, every virtual antenna of SD+BF and SD+CDD can transmit one data stream and introduce redundancy into the time or frequency domains to gain spatial diversity. On average, one data stream is transmitted at a time. BF and CDD send one data stream at a time and apply for applications with higher channel correlation. They are simple to implement, are transparent to the user, and do not need to support MIMO. The most often-used antenna configuration is one in which the transmitter has four or eight antennas and the receiver has one or two antennas. These characteristics are shown in Table 2.

    (3) Application scenarios of different multiantenna modes
    CDD, SD+CDD, and SM+CDD gain diversity in the frequency domain when channel mulitpath delay is manually introduced. This can be done when the channel status information is unknown. However, in BF, SD+BF, and SM+BF the weighted value of BF needs to be estimated, and the user is required to feedback channel status information or make use of the reciprocity feature of the channel. Therefore, performance is to a great extent affected by the precision and promptness of the estimation of the weighted value. In normal cases where the weighted value is estimated precisely and promptly, performances of BF, SD+BF, and SM+BF are better than those of CDD, SD+CDD, and SM+CDD. If the weighted value is not precise enough because of channel changes or because the user moves too fast, performances may not be as good as those of CDD, SD+CDD, and SM+CDD. In scenarios where channel space correlation is low, SM+BF and
SM+CDD can send different data streams over different virtual antennas. However, where channel space correlation is high, BF, SD+BF, CDD, and SD+CDD can gain diversity. Table 3 summarizes the application scenarios of different multiantenna modes [13].

3 Adaptive Mode Switchover
    Since every multiantenna mode has its own characteristics and application scenarios, the wireless communication system must adaptively switch between the modes to suit the changing physical location, channel environment, moving speed, and service type of the user. In this way, system performance can be maximized and high-quality communication can occur [14].
In practice, switching between multiantenna modes can present challenges, and there are many factors that can affect the performance of a multiantenna mode [15].

    There are many types of mode switchover; switching between BF, SD+BF, SM+BF, CDD, SD+CDD, and SM+CDD involves at least 15 modes. This makes the switchover algorithm very complex, and further research is required to work out the differences and similarities between diffierent switchover types so as to simplify implementation algorithms.

    Based on research conducted into the properties of multiantenna technologies in various simulations, three major types of switchover are shown in Fig. 6: BF-related (BF, SD+BF, and SM+BF) switchover, CDD-related (CDD, SD+CDD, and SM+CDD) switchover, and switchover between BF-related technologies and CDD-related technologies.

    BF-related or CDD-related technologies are employed depending on the speed of the receiver and the correlation of two neighboring weighted values. For BF-related technology, spectrum efficiency under the SM+BF, SD+BF, and BF modes is calculated, and the data transmission mode with the maximum spectrum efficiency is selected. For CDD-related technology, the spectrum efficiency under SM+CDD, SD+CDD, and CDD is calculated, and the data transmission mode with the maximum spectrum efficiency is selected.

4 Conclusion
    This paper introduces the concept of multiantenna modes and analyzes the performance, influential factors, and application scenarios of these modes. It also discusses the algorithm for multiantenna mode switchover. ZTE has not only realized different multiantenna technology modes but has also studied and simulated factors affecting performance. A multiantenna technology mode can be selected depending on the application scenario or channel environment in order to boost system performance to the greatest possible extent.

References
[1] ZTE Corporation, Combination of MIMO and Beamforming Technology for WIMAX, [Online]. Available: http://wwwen.zte.com.cn/en/solutions/wireless/wimax/200912/t20091218_178788.html
[2] M. Kobayashi, G. Caire, D. Gesbert, “Transmit diversity versus opportunistic beamforming in data packet mobile downlink transmission,” IEEE Trans. Commun., vol. 55, no. 1, pp.151-157. 2007.
[3] A. Sadeque and M. Saquib, “Delay diversity for correlated MIMO channel,” Proc. 2009 International Waveform Diversity Design Conf. (WD&D’09), Orlando, FL, 2009, pp.302-306.
[4] IEEE Standard for Local and Metropolitan Area Networks, Part 16: Air Interface for Fixed Broadband Wireless Access Systems, IEEE Std 802.16eTM rev2-D8, 2008.
[5] L. Zhang and D. Tse, “Diversity and multiplexing: A fundamental tradeoff in multiple antenna channels,” IEEE Trans.Inf. Theory, vol. 49, no. 5, pp.1073-1096, 2003.
[6] S. Alamouti, “A simple transmit diversity technique for wireless communications,” IEEE J. Sel. Areas Commun., vol. 16, no. 8, pp.1451-1458, 1998.
[7] G. Golden, C. Foschini, R. Valenzuela et al., “Detection algorithm and initial laboratory results using V-BLAST space-yime communication architecture,” Electronics Letters, vol. 35, no. 1, pp.14-16, 1999.
[8] W. Meng, L. Gu, and C. Li, “The Combined beamforming and space-time block coding technique for downlink transmission,” Proc.2005 Int.Conf. Wirelss Networks,Commun.,and Mobile Comput.(WirelessCom’05), Maui, HI, vol. 1, pp.481-486.
[9] I. Kim, K. Lee, and J. Chun, “A MIMO antenna structure that combines transmit beamforming and spatial multiplexing,” IEEE. Trans. Wireless Commun., vol. 6, no. 3, pp. 775-779, 2007.
[10] X.Tong,T.Luo, OFDM Principle and Applications of Mobile Communication Technology, Beijing: POSTS & TELECOM PRESS. 2000.
[11] C. Yuen, Y. Wu, and S. Sun, “Four transmit diversity schemes for coded OFDM systems with four transmit antennas,” J. Commun., vol. 3, no. 4, pp.1-7, 2008.
[12] M. Rahman, S. Das, E. de Carvalho et al., “Spatial multiplexing in OFDM systems with cyclic delay diversity,” Proc. 65th Vehicular Tech. Conf. (VTC-Spring’07), Dublin, Ireland, 2007, pp.1491-1495.
[13] Y. Hou,X.Tao. “Architecture of Green Communication-Oriented Base Stations,” ZTE Communications.vol.16, no.6, PP.16-19, 2010.
[14] W.Zhang, H.Hu,Y.Yang. “Green Wireless Network Technologies,” ZTE Communications. vol. 16, no. 6, pp. 4-7, 2010.
[15] P.Gao,X.Chen,J.Wang. “ZP-CI/OFDM：A Power Efficient Wireless Transmission System, ” ZTE Communications. vol. 16, no. 6, pp. 20-23, 2010.

Biographies

Huahua Xiao (xiao.huahua@zte.com.cn) received an M.S. degree in computer software and theories from Sun Yat-Sen University, China. He currently works with ZTE Corporation as a senior engineer in the field of antenna algorithm pre-research. Xiao has applied for more than 30 Chinese and foreign patents in the multiantenna field.

Dengkui Zhu (zhu.dengkui@zte.com.cn) received an M.S. degree in communication and information from Southeast University, China. Zhu currently works with ZTE Corporation as a pre-research manager for the MIMO project in the wireless pre-research department of the R&D system. Zhu has many years of research experience in broadband wireless communications. He has been a member of the pre-research and product implementation team for ZTE’s WiMAX and LTE projects. Zhu is responsible for system simulation and key technology research and has a very good understanding of key OFDM and MIMO technologies.

Liujun Hu (hu.liujun@zte.com.cn)graduated from Harbin Engineering University. He now works with ZTE Corporation as a senior engineer and director of the wireless pre-research department. Hu’s research interests include mobile communication networks and their key technologies. He has published more than 80 papers and patents.

[Abstract] Multiantenna technology can be implemented in several modes. These modes have varying characteristics and are used in different scenarios. This paper introduces Beamforming (BF), Cyclic Delay Diversity (CDD), Spatial Diversity (SD), Spatial Multiplexing (SM), and other multiantenna technologies. It also analyzes various technical features and their application scenarios. An adaptive multiantenna switching algorithm is proposed that chooses a suitable mode for sending data according to the scenario or wireless channel conditions. This switching algorithm improves multiantenna technology and enhances the quality of wireless network communications.

[Keywords] BF; CDD; SD; SM; adaptive mode switch