Research Article Dynamic Beamforming for Three-Dimensional MIMO Technique in LTE-Advanced Networks

Transcription

1 Antennas and Propagation Volue 213, Article ID 76457, 8 pages Research Article Dynaic Beaforing for Three-Diensional MIMO Technique in LTE-Advanced Networks Yan Li, Xiaodong Ji, Dong Liang, and Yuan Li Wireless Signal Processing and Network Lab, Key Laboratory of Universal Wireless Counications (Ministry of Education), Beijing University of Posts and Telecounications, Beijing 1876, China Correspondence should be addressed to Yan Li; ly @gail.co Received 8 March 213; Revised 3 July 213; Accepted 4 July 213 Acadeic Editor: Feifei Gao Copyright 213 Yan Li et al. This is an open access article distributed under the Creative Coons Attribution License, which perits unrestricted use, distribution, and reproduction in any ediu, provided the original work is properly cited. MIMO syste with large nuber of antennas, referred to as large MIMO or assive MIMO, has drawn increased attention as they enable significant throughput and coverage iproveent in LTE-Advanced networks. However, deploying huge nuber of antennas in both transitters and receivers was a great challenge in the past few years. Three-diensional MIMO (3D MIMO) is introduced as a proising technique in assive MIMO networks to enhance the cellular perforance by deploying antenna eleents in both horizontal and vertical diensions. Radio propagation of user equipents (UE) is considered only in horizontal doain by applying 2D beaforing. In this paper, a dynaic beaforing algorith is proposed where vertical doain of antenna is fully considered and beaforing vector can be obtained according to UEs horizontal and vertical directions. Copared with the conventional 2D beaforing algorith, throughput of cell edge UEs and cell center UEs can be iproved by the proposed algorith. Syste level siulation is perfored to evaluate the proposed algorith. In addition, the ipacts of downtilt and intersite distance (ISD) on spectral efficiency and cell coverage are explored. 1. Introduction As is generally known, long-ter evolution (LTE) designed by the third generation partnership project (3GPP) helps operators to provide wireless broadband services with enhanced perforance and capacity. A variety of targets and requireents for LTE have been suggested by 3GPP, including higher peak data rates and ore UEs per cell as well as lower control plane latency than currently eployed 3G architectures [1, 2]. Based on orthogonal frequency division ultiple access (OFDMA), radio technology applies various scheduling and ultiantenna ethods. 3G LTE is further developed to eet requireents set for IMT-Advanced technologies.theroleofantennaparaeterselectioninevolution of 3G LTE-Advanced has been widely discussed currently [3]. AsoneofthekeytechniquesinLTE-Advancedsyste, ultiple input ultiple output (MIMO) [4 6] is considered as an effective way to obtain high data rates without sacrificing bandwidth and to iprove the perforance of both cell edge UE and the whole syste. A trend toward deploying larger nuber of antennas [7, 8]can be noted in the evolution of soe standards such as IEEE 82.11n/82.11ac and LTE. With Massive MIMO, huge nubers of eleents in antenna arrays are able to deployed in systes being built today [9]. Larger nubers of terinals can always be accoodated by cobining assive MIMO technology with conventional tie and frequency division ultiplexing via orthogonal frequency division ultiplexing (OFDM). Massive MIMO is a new research field in counication theory, propagation, and electronics and represents a paradig shift in the way of considering with regard to theory, systes, and ipleentation. Three-diensionalMIMO(3DMIMO)canbeseenas an effective ethod to approach assive MIMO without applying too uch antennas on transitter or receiver. Vertical diension will be utilized in the antenna odeling, and downtilt of the antennas will becoe significant channel paraeters. A typical two-diensional (2D) antenna is used to cover a sector of 12 degrees only in horizontal doain [1]. Copared with the 2D channel propagation, scatterer is no longer located in the sae plane with antennas and is supposed to distribute randoly in three-diensional space.

2 2 Antennas and Propagation We are considering the possibility of extending the current 2D antenna to the future 3D antenna, which eans that thedepartureandarrivalangleshavetobeodeledintwo directions, that is, horizontal and vertical [11 13]. Beaforing of linear array antenna eleents erely in horizontal diension does not give full free-space gain. This is due to the aziuth spread of the received signal as seen fro the base station (BS), which has been extensively investigated in previous studies. It is necessary to propose a novel beaforing algorith which includes the gain of the vertical diension. In this paper, we will present a dynaic beaforing algorith in which vertical difference of 3D antennas will be taken into account. In conventional 2D MIMO scenario, cell edge UEs suffer serious intercell interference fro neighboring cell due to their locations [14, 15]. Since the radio propagation fro a transission node to a UE is divided into horizontal direction and vertical direction, the power of intercell interference can be reduced largely, which results in an enhanceent of the signal to interference plus noise ratio (SINR). Copared with the existing 2D beaforing algorith, better syste perforance can be brought by dynaic beaforing algorith. The rest of this paper is organized as follows. Section 2 describes the syste odel of the downlink LTE-Advanced networks and the odeling of 3D MIMO channel. The principle and details of the dynaic beaforing algorith are introduced in Section 3, and perforance of the proposed algorith with different configuration paraeters is siulated and analyzed in Section 4.Theconclusionofthispaper is given in Section 5. Macro cell Macro UE Signal Strong interference Weak interference Macro BS Figure 1: Syste odel for downlink transission of LTE- Advanced networks. Antenna eleent 2. Syste Model D MIMO Syste. The downlink of a cellular network with M hexagonal cells is considered as depicted in Figure1, and each cell is partitioned into 3 sectors with K active UEs served within the coverage of each sector. The total nuber of BSs in the syste is 3M. Each BS which corresponds to one sector is equipped with N t transitting antennas, while each UE has N r receiving antennas. For the received signal at the kth UE served by BS,wehave y (k) = H(k) W(k) s(k) desired signal 3M K H (k) + n W(k) n s(w) n n =w=1 inter-sector interference + n (k), (1) where s (k) is the l k 1transitted vector for UE k served by BS and l k is the nuber of data strea. Denote s (k) = [ p k s (k),1, p ks (k),2,..., p ks (k),l ] T,andp k k is the transitting powerofeachdatastreaatuek. For 3D MIMO, due to the introduction of vertical diension in transitting antenna, the channel coefficient atrix is three-diensional, that is, the nuber of receiving antenna, the nuber of transitting antenna, and the nuber of elevation eleent. It is assued that each transitting antenna has N V eleents, The whole antenna Figure 2: Structure of 8 4rectangular 3D antenna array. and receiving antenna reains as a conventional 2D antenna without a different eleent in vertical diension. Thus, H (k) is the N r (N t N V ) channel atrix fro BS to UE k, and W (k) is the (N t N V ) l k precoding atrix. n (k) is the additive white Gaussian noise with zero ean and variance E(n (k) n(k)h )=σ2. The detailed analysis of detected SINR for thereceivedsignalisin[16, 17] D Antenna Modeling. Most geoetry-based stochastic radio channel odels are two-diensional (2D) in the sense that they use only geoetrical xy-coordinates or equivalent paraeters of distance and rotation angle. This has been sufficient until these days, while vertical diension of the arrays and the height of the BS are fully considered in 3D antenna [18, 19]. For instance, structure of 8 4arrays with rectangular eleentsisshowninfigure 2.

3 Antennas and Propagation 3 Vertical direction θ tilt Line of horizontal Macro BS Figure 3: Electrical downtilt. The whole antenna can be divided into several eleents which are controlled by different antenna port. Assuing that 14 db is the axiu direction gain of an antenna, equal power is given for each eleent as defined in G ax =14 1log 1 N V, (2) where G ax is the axiu directional gain of the array eleents and N V is the nuber of elevation eleent. In 3GPP LTE-Advanced siulations, we apply two forulas below for horizontal and vertical radiation patterns so that horizontal antenna gain and vertical antenna gain can be obtained, respectively: A H (φ) = in [12 ( φ ),A φ ], 3 db A =2dB, A V (θ) = in [12 ( θ θ tilt θ 3 db ),SLA V ], SLA V =2dB, where φ is defined as the angle between the direction of interest and the boresight of the antenna in horizontal diension, which is siilar to θ in vertical diension. φ 3dB and θ 3dB are the 3 db beawidth of the horizontal bea and the vertical bea, respectively. θ tilt is the downtilt angle in transitter. A is the front-to-back attenuation, and SLA V is side lobe attenuation. The 3D antenna gain is cobined as a su of horizontal pattern and elevation pattern antenna gain. Generation of the 3D pattern fro two perpendicular cross-sections aziuth and elevation patterns is denoted as below: (3) A (φ, θ) = in { [A H (φ) + A V (θ)],a }, (4) The angle of the ain bea of the antenna up/below the horizontal plane is called antenna tilt as shown in Figure 3. Positive and negative angles are referred to as downtilt and uptilt, respectively. In electrical downtilt, ain, side, and back lobes are tilted uniforly by adjusting phases of antenna eleents [2]. Figure 4: Conventional 2D MIMO beaforing. Angle θ (i,k) between horizontal line and LOS direction of connecting BS i and UE k in vertical plane is obtained by θ (i,k) = arctan ( h (i,k) d (i,k) ), (5) where d (i,k) is distance between UE k and BS i and h (i,k) is the height difference between the antenna of UE k and the antenna of BS i. 3. Algoriths for 3D Dynaic Beaforing D MIMO Beaforing. Conventional beaforing can be seen as a sort of 2D cell-specific beaforing. The UE data to be transitted is processed in accordance with the channel inforation only in the horizontal diension. The transitter fors a sall bea within the antenna radiation patterntopointattheuetobeserved.thiseansthatonly in the horizontal diension the UE channel state inforation is being tracked and utilized. Only in the case of several UEs within a cell, the horizontal antenna bea generated by beaforing can distinguish different UEs well. However, as the UEs in a cell increase substantially, several UEs that request for service at the sae tie ay be located in the sae aziuth angle as shown in Figure 4, which eans that 2D antenna gain fro the serving BS is equivalent. Once several UEs are close to each other and located in the cell edge area, they own the sae power of desired signal but will suffer serious intercell interference. In this case, transitting bea siply fored by horizontal diension beaforing is not enough to distinguish different UEs. For instance, if the bea is aligned with the UE near the base station, the reote UEs will be outside the range of radiation; otherwise, the bea aligned with the UEs far fro the base station will result in severe intercell interference to the neighbor cell Proposed Beaforing Algorith. In the sae frequency resource allocation networks, the intercell interference decreases the perforance of cell edge UE draatically.

4 4 Antennas and Propagation Figure 5: 3D dynaic beaforing in horizontal sight. Figure 6: 3D dynaic beaforing in vertical sight. 3D MIMO can be seen as a dynaic interference cancellation ethod, which ais to eliinate intercell interference coing fro neighbor cells and iprove the throughput of cell edge UE through 3D dynaic beaforing. For each UE, since the introduction of vertical diension, difference in height between BS and UE has been taken into account, and UE s position should be considered in three-diension. According to both horizontal and vertical directions of the specific UE, transitting bea generated by 3D antennas can be divided into both horizontal and vertical diensions, which eans it has uch less influence on other UEs. For instance, the two UEs in Figures 5 and 6 have the sae aziuth angle. When they are scheduled at the sae tie, identical horizontal bea direction is shared. Due to the introduction of dynaic beaforing algorith, transitter is able to for two different vertical beas through setting different vertical power weight on the antenna eleents. Thus, different axiu bea direction can be discriinated for the two specific UEs. Aong noncodebook-based precoding technique, the calculation of the 3D MIMO beaforing vector seriously affects the perforance of the interference cancellation, and the ethod of cobining suitable horizontal beaforing vector and vertical beaforing vector together should be a big proble needed to be solved. To ipleent precoding at BS, an iproved beaforing algorith is adopted. It is assued that one UE is served by all the antenna eleents, and data fro a BS is transitted in a single strea. Antenna nuber at Macro BS is N t,and UE has N r antennas. Firstly, copared with the conventional algoriths such as SVD algorith and BD algorith, equivalent process is done with the channel between BSs and UEs, andthentraditionalprecodingethodsareutilized.the ain procedure of dynaic beaforing is as follows. Step 1. Coputation of horizontal beaforing vector. The horizontal beaforing vector represents the transitting weight of the eleents fro different antenna in horizontal diension. Therefore, the diension of each horizontal beaforing vector is N t 1.ForthereareN V eleents in an antenna, the total nuber of horizontal beaforing vector should be N V.ThechannelfroN V layer antenna eleents in transitter to UE k is H k,1, H k,2,...,h k,nv,respectively,and H k,i denotes the N r N t channel atrix fro the ith layer transitting antenna eleents to UE k. V k,i SVD(H k,i ); V k,i is the V atrix fro the SVD decoposition of H k,i. v k,i is the first colun of atrix V i,k. The horizontal beaforing vector v horizontal for the whole antenna eleents can be expressed as v horizontal =[v T k,1, vt k,2,...,vt k,n V ], (6) Step 2. Coputation of vertical beaforing vector. The vertical beaforing vector represents the transitting weight of the eleents fro a single antenna in vertical diension. Therefore, the diension of each vertical beaforing vector is N V 1.ForthereisactualN t antenna in transitter, the total nuber of vertical beaforing vector should be N t. The channel fro vertical eleents of N t antennas in transitter to UE k is H k,1, H k,2,...,h k,n t, respectively; H k,i denotes the N r N V channel atrix fro the vertical eleents of the ith transitting antenna to UE k. V k,i SVD(H k,i ),wherev k,i is the V atrix fro the SVD decoposition of H k,i. v k,i is the first colun of atrix V k,i. The vertical beaforing vector v vertical for the whole antenna eleents can be expressed as v vertical = [(v k,1 )T, (v k,2 )T,...,(v k,n t ) T ], (7) Step 3. Cobination of horizontal beaforing vector and vertical beaforing vector. There are totally N V N t antenna eleents in transitter. H is a cobined channel and H = [H T k,1, HT k,2,...,ht k,n V ]. Each eleent has its own horizontal transitting weight as coputed in Step 1. Thus, the coplete horizontal beaforing vector s diension is N V N t 1,cobinedbyN V subvectors which own diension of N t 1. Siilar process is done with the coplete vertical beaforing vector that N t subvectors represent N t antennas vertical transitting weight, respectively. Thus, vertical beaforing vector is generated fro row cobination of subvector v k,1,...,v k,n t in Step 2. Finally,the dynaic beaforing vector is obtained fro the coplete horizontal beaforing vector and the coplete vertical

5 Antennas and Propagation 5 Table 1: Siulation configuration paraeters. Paraeters Assuptions Figure 7: 19 cells topological structure. beaforing vector, for dot production will be done with these two N V N t 1vectors. 3D beaforing vector w is available as follows: w = v horizontal v vertical, (8) where the diension of 3D dynaic beaforing vector is N V N t 1,representingN V N t antenna eleents dynaic power transitting weight. 4. Siulation Results In this section, syste level perforance of the proposed 3D dynaic beaforing algorith is evaluated in ters of average spectral efficiency, cell edge UE spectral efficiency, and SINR of the UEs. Consider a syste with 19 hoogeneous cells as shown in Figure 7.Eachcellconsistsof3Macro BSs, which is supposed to provide service to its own sector s UEs. It is worth noting that the 3 BSs are located in the center of the cell, which eans they share the sae coordinate point. In order to eliinate border effect, wrap-around technique is used to siulate interference generated by UEs fro at least 2 layers neighbor cells. Since the siulated scene is configured to Ua (Urban Macro-cell), only outdoor UEs exist during the siulation process. Other detailed siulation paraeters are listed in Table 1. Figure 8(a) is the average spectral efficiency of different downtilt angle deployents with sector radius of 288. The perforance of 3D MIMO without dynaic beaforing has 5%, 14%, 22%, and 24% enhanceent copared to 2D MIMO networks, respectively. Although array gain due to the introduction of 3D MIMO can be obtained in the syste, beaforing gain is not included in the perforance Cell type Cellular layout Sector radius Bandwidth Channel odel Shadowing standard deviation Antenna nuber Antenna configuration Antenna spacing Macro cell 19 cells, each with 3 sectors 288, 5 5 M WINNER II 8 db Transitter: 2Tx, Receiver: 2Rx Co-polarized linear array Transitter:.5λ, Receiver:.5λ Nuber of elevation eleent 2 Regular antenna gain Downtilt angle BS ax TX power BS height UE height UE distribution UE nubers UE speed Initial UE access criterion Scheduler CQI reporting interval Delay for scheduling and AMC Receiver Traffic odel 14 dbi 8,1,12,14 46 db Uniforly distributed 1 UEs per sector 3 k/h Access by location Proportional fair 5 s 5 s MMSE Full buffer iproveent.ascanbeseeninthefigure,theproposed dynaic beaforing algorith is able to offer substantial beaforing gain copared to static beaforing. About 14%, 24%, 32%, and 34% of the gain copared to 2D MIMO canbeachievedbythenovelbeaforingalgorithwith different downtilt angle paraeter. It can be concluded that the perforance of 14 outperfors the perforance of other 3 downtilt angles for the scenario with sector radius of 288. Figure 8(b) is the cell edge UE (5% CDF) spectral efficiency.theproposeddynaicbeaforingalgorithoffers large perforance gain copared to 2D MIMO networks as the intercell interference decreases apparently. For cell edge UEs, the UE spectral efficiency has been increased by about 21%, 51%, 79%, and 92%, respectively, and the perforance of coverage has been increased draatically. For different scenarios with sector radius of 5, Figure 8(c) shows the average spectral efficiency of different

6 6 Antennas and Propagation Cell average spectru efficiency (bit/s/hz) Downtilt Cell edge UE spectru efficiency (bit/s/hz) Downtilt (a) (b) Cell average spectru efficiency (bit/s/hz) Downtilt Cell edge UE spectru efficiency (bit/s/hz) Downtilt 2D MIMO 3D MIMO without dynaic beaforing 3D MIMO with dynaic beaforing 2D MIMO 3D MIMO without dynaic beaforing 3D MIMO with dynaic beaforing (c) (d) Figure 8: Syste level results of 2D MIMO, 3D MIMO without dynaic beaforing and 3D MIMO with dynaic beaforing: (a) cell average spectral efficiency with sector radius of 288 ; (b) cell edge UE spectral efficiency with sector radius of 288 ; (c) cell average spectral efficiency with sector radius of 5 ; (d) cell edge UE spectral efficiency with sector radius of 5. downtilt angle deployents, and the perforance of 3D MIMO with dynaic beaforing has 1%, 14%, 15%, and 13% enhanceent copared to 2D MIMO networks, respectively. Since the radius of sector increases, the distribution of UEs within a sector changes a lot, and the elevation angles of UEs in the area decrease. Downtilt angle should be adjusted so as to cover ore UEs. It can be observed fro Figure 8 that the perforance of 12 outperfors the perforance of other 3 downtilt angles for the scenario with sector radius of 5. For cell edge UEs, the UE spectral efficiency has been increased by about 2%, 4%, 45%, and 15% respectively as presented in Figure 8(d). Although the density of UE changed, the proposed algorith can still offer considerable gain copared with 2D MIMO networks. Figure 9 shows the receive SINR CDF curves of 2D MIMO UEs, 3D MIMO without dynaic beaforing UEs, and 3D MIMO with dynaic beaforing UEs. For scenario with sector radius of 288, the downtilt angle is set for 14 as suarized above. With the proposed algorith,

7 Antennas and Propagation 7 CDF SINR (db) 2D MIMO 3D MIMO without dynaic beaforing 3D MIMO with dynaic beaforing (a) CDF SINR (db) 2D MIMO 3D MIMO without dynaic beaforing 3D MIMO with dynaic beaforing (b) Figure 9: Syste level results of receive SINR CDF for 2D MIMO, 3D MIMO without dynaic beaforing and 3D MIMO with dynaic beaforing: (a) sector radius: 288 ; (b) sector radius: 5. the SINR of the 3D MIMO with dynaic beaforing outperfors by 6 db copared to the 2D MIMO and has approxiately 2 db gain copared with the 3D MIMO without dynaic beaforing. Both cell average UE and cell edge UE can be served uch better than before. Undertheinfluenceoflargersectorradius,thechangeof downtilt configuration is not so apparent as it perfors in 288 scenario. Copared with the UEs in 288 scenario, soe UEs in 5 scenario will receive decreased signal power so that less gain can be obtained in the average received SINR of UEs in 5 scenario. Thus, there is not uch difference between the SINR curved line of 2D MIMO UE and the SINR curved line of 3D MIMO UE as depicted in Figure 9(b). Inspiteofthis, thepositiveeffectbroughtbythe proposed algorith is still obvious. 5. Conclusion In this paper, odeling of 3D channel and structure of 3D antenna have been introduced. A dynaic beaforing algorith for 3D MIMO in LTE-Advanced networks is proposed, which can be proved as an effective ethod to reduce intercell interference. Siulation results for 3D MIMO with different downtilt angle paraeters and sector radius paraeters are presented. Copared with 2D beaforing, 3D dynaic beaforing can iprove both the cell edge UE throughput and the whole syste s perforance significantly. At ost 34% gain for cell center UE and 92% gain for cell edge UE over the conventional 2D beaforing are achieved, and the perforance differences of different scenarios are analyzed in this paper. Acknowledgents This work was supported in part by the State Major Science and Technology Special Projects (Grant no. 212ZX3128-4, 213ZX311-2, and 213ZX311-3) and the Beijing Natural Science Foundation (Grant no ). References [1] M. Peng and W. Wang, Technologies and standards for TD- SCDMA evolutions to IMT-advanced, IEEE Counications Magazine,vol.47,no.12,pp.5 58,29. [2] M. Peng, W. Wang, and H.-H. Chen, TD-SCDMA evolution, IEEE Vehicular Technology Magazine, vol.5,no.2,pp.28 41, 21. [3] 3GPP TS V. 4. 1, Evolved universal terrestrial radio access (E-UTRA), Further advanceents for E-UTRA physical layer aspects. [4] L. Liu, R. Chen, S. Geirhofer, K. Sayana, Z. Shi, and Y. Zhou, Downlink MIMO in LTE-advanced: SU-MIMO vs. MU- MIMO, IEEE Counications Magazine, vol.5,no.2,pp , 212. [5] M.Peng,Z.Ding,Y.Zhou,andY.Li, Advancedself-organizing technologies over distributed wireless networks, International Distributed Sensor Networks, vol.212,articleid , 2 pages, 212. [6] O. N. C. Yilaz, S. Hääläinen, and J. Hääläinen, Analysis of antenna paraeter optiization space for 3GPP LTE, in Proceedings of the IEEE 7th Vehicular Technology Conference Fall (VTC 9), Anchorage, Alaska, USA, Septeber 29. [7] S.K.Mohaed,A.Zaki,A.Chockalinga,andB.S.Rajan, High-rate space-tie coded large-mimo systes: lowcoplexity detection and channel estiation, IEEE Journal on

8 8 Antennas and Propagation Selected Topics in Signal Processing, vol.3,no.6,pp , 29. [8] M.Peng,G.Xu,W.Wang,andH.-H.Chen, McWiLL-anew obile broadband access technology for supporting both voice and packet services, IEEE Systes Journal,vol.4,no.4,pp , 21. [9]F.Rusek,D.Persson,B.K.Lauetal., ScalingupMIMO: opportunities and challenges with very large arrays, IEEE Signal Processing Magazine,vol.3,no.1,213. [1] P. Zetterberg, Perforance of antenna tilting and beaforing in an urban acrocell, in Proceedings of the 1st International Conference on Wireless Counication, Vehicular Technology, Inforation Theory and Aerospace and Electronic Systes Technology, Wireless (VITAE 9), pp , Aalborg, Denark, May 29. [11] IST WINNER II D1.1.2 V1.2 WINNER II channel odels. [12] IST-WINNER D5.3 WINNER+ Final channel odels v1.. [13] IST WINNER D5.4 Final report on link level and syste level channel odels v1.4. [14] X. You, D. Wang, P. Zhu, and B. Sheng, Cell edge perforance of cellular obile systes, IEEE Journal on Selected Areas in Counications, vol. 29, no. 6, pp , 211. [15] T. Zhou, M. Peng, W. Wang, and H. Chen, Low-coplexity coordinated beaforing for downlink ulti-cell SDMA/OFDM syste, IEEE Transactions on Vehicular Technology,vol.62,no.1,213. [16] F. Shu, L. Lihua, C. Qiei, and Z. Ping, Non-unitary codebook based precoding schee for ulti-user MIMO with liited feedback, in Proceedings of the IEEE Wireless Counications and Networking Conference (WCNC 8), pp , April 28. [17] M.Peng,X.Zhang,W.Wang,andH.-H.Chen, Perforanceof dual-polarized MIMO for TD-HSPA evolution systes, IEEE Systes Journal,vol.5,no.3,pp ,211. [18] M.-T. Dao, V.-A. Nguyen, Y.-T. I, S.-O. Park, and G. Yoon, 3D polarized channel odeling and perforance coparison of MIMO antenna configurations with different polarizations, IEEE Transactions on Antennas and Propagation, vol.59,no.7, pp , 211. [19] G. Auer, 3D MIMO-OFDM channel estiation, IEEE Transactions on Counications,vol.6,no.4,pp ,212. [2] O. N. C. Yilaz, S. Hääläinen, and J. Hääläinen, Coparison of reote electrical and echanical antenna downtilt perforance for 3GPP LTE, in Proceedings of the IEEE 7th Vehicular Technology Conference Fall (VTC 9), Anchorage, Alaska, USA, Septeber 29.

9 Rotating Machinery Engineering The Scientific World Journal Distributed Sensor Networks Sensors Control Science and Engineering Advances in Civil Engineering Subit your anuscripts at Electrical and Coputer Engineering Robotics VLSI Design Advances in OptoElectronics Navigation and Observation Cheical Engineering Active and Passive Electronic Coponents Antennas and Propagation Aerospace Engineering Modelling & Siulation in Engineering Shock and Vibration Advances in Acoustics and Vibration