Point-to-point radio link variation at E-band and its effect on antenna design Al-Rawi, A.N.H.; Dubok, A.; Herben, M.H.A.J.; Smolders, A.B.

Transcription

1 Point-to-point radio link variation at E-band and it effect on antenna deign Al-Rawi, A.N.H.; Dubok, A.; Herben, M.H.A.J.; Smolder, A.B. Publihed in: PIERS 215 Prague Publihed: 1/1/215 Document Verion Publiher PDF, alo known a Verion of Record (include final page, iue and volume number) Pleae check the document verion of thi publication: A ubmitted manucript i the author' verion of the article upon ubmiion and before peer-review. There can be important difference between the ubmitted verion and the official publihed verion of record. People intereted in the reearch are advied to contact the author for the final verion of the publication, or viit the DOI to the publiher' webite. The final author verion and the galley proof are verion of the publication after peer review. The final publihed verion feature the final layout of the paper including the volume, iue and page number. Link to publication Citation for publihed verion (APA): Al-Rawi, A., Dubok, A., Herben, M. H. A. J., & Smolder, A. B. (215). Point-to-point radio link variation at E- band and it effect on antenna deign. In PIERS 215 Prague: Progre In Electromagnetic Reearch Sympoium (pp ). Cambridge: The Electromagnetic Academy. General right Copyright and moral right for the publication made acceible in the public portal are retained by the author and/or other copyright owner and it i a condition of acceing publication that uer recognie and abide by the legal requirement aociated with thee right. Uer may download and print one copy of any publication from the public portal for the purpoe of private tudy or reearch. You may not further ditribute the material or ue it for any profit-making activity or commercial gain You may freely ditribute the URL identifying the publication in the public portal? Take down policy If you believe that thi document breache copyright pleae contact u providing detail, and we will remove acce to the work immediately and invetigate your claim. Download date: 13. Sep. 218

2 Progre In Electromagnetic Reearch Sympoium Proceeding 913 Point-to-point Radio Link Variation at E-band and It Effect on Antenna Deign A. N. H. Al-Rawi, A. Dubok, M. H. A. J. Herben, and A. B. Smolder Electromagnetic Group, Department of Electrical Engineering Eindhoven Univerity of Technology, The Netherland Abtract Radio propagation will trongly influence the deign of the antenna and front-end component of E-band point-to-point communication ytem. Baed on the ITU rain model, the rain attenuation i etimated in a tatitical ene and it i concluded that for backhaul link of 1 1 km, antenna with a gain of 49.5 dbi are required. Moreover, depolarization can be a limiting factor for backhaul ytem that are employing orthogonal polarization in order to improve capacity. Antenna mat movement become a relevant problem due to the narrow beamwidth of the high gain antenna, which i around.7. We propoe to implement a focal plane array a feed for the parabolic reflector antenna. Thi i to tackle the mat movement by electronic beam teering and to increae EIRP by increaing the number of active antenna element, and to ait the mechanical alignment during intallation. 1. INTRODUCTION The network architecture of LTE and LTE-advanced promie the mobile uer a wired experience in their wirele network. Thi put high demand on the capacity (bit/ec) of the backhaul of thee cellular ytem. Fiber optic can due to it virtually unlimited capacity (1 of Gbp) fulfill thee demand. However laying optical fiber in an urban and rural area i very expenive, and therefore a wideband point-to-point radio link can be a good alternative. Millimeter-wave wirele ytem operating in E-band (71 76, 81 86, GHz), offer a bandwidth of 13 GHz which i ufficient to produce a bit rate of 1 Gbp with imple modulation cheme uch a QPSK. Thi make thee ytem a good option for the backhaul. There i an increaing demand on extending the E-band backhaul range. Thi i to reduce the contruction and maintenance cot. It relie on increaing the antenna gain. A a conequence, the unwanted movement of the antenna mat become very relevant at thi band becaue it caue a drop in the SNR. In addition, the antenna alignment during intallation become cumberome. 2. POINT-TO-POINT LINK VARIATION The atmopheric window at E-band how very low attenuation wherea it i ignificantly high at the neighboring 6 GHz band due to oxygen aborption [1]. Therefore, the 6 GHz band i limited to indoor and hort range application. The free-pace and rain attenuation i high at E-band and they become very ignificant on the extended range of 1 km. Thee effect have been conidered in [2]. Baed on the ITU rain model [3] and data [4, 5] we analyzed the effect of rain in a tatitical ene (ee Fig. 1). For a backhaul length of 5 km, with rain attenuation of.1% of the time (i.e., 99.99% ytem availability); the total attenuation will be [A tot = free-pace db + rain attenuation 15.2 db + atmophere attenuation 4 db = db]. A large antenna gain i required to compenate for uch large attenuation. At an antenna gain of 41 dbi, a SNR at the receiver input of 4.4 db, and with the aumption that the receiver ha a noie figure of 1 db, the SNR d at the detector input i 3.4 db. Thi i quite ufficient for QPSK with a bit error rate of 1 6, that uually require an SNR around 1 db. For the longer path length of 1 km, the total attenuation i db, and SNR d at the detector input i 16.5 db. Thi i till adequate for the demodulator to accurately perform. Exceeding the rain attenuation of.1% to.1% of the time (i.e., % ytem availability) to increae the reliability of the point-to-point wirele communication, the total attenuation for 5 and 1 km path length i 175 db and 195 db, repectively. The SNR d for the 5 km path length i 21 db wherea for the 1 km path length it i 1 db. The latter can be improved by increaing the receiver antenna gain to 49.5 dbi. Thi bring SNR d to 9.5 db, which i ufficient for QPSK with a bit error rate of 1 6. Thi alo can be further improved by reducing the receiver noie. In addition, and due to the aniotropic nature of the rain medium, a pure vertically-polarized wave will arrive at the receiver with a mall horizontally-polarized field component. Thi effect

3 914 PIERS Proceeding, Prague, Czech Republic, July 6 9, Path length= 1 km Path length= 5 km Path length= 1 km 1 Path length = 1 km Path length = 5 km Path length = 1 km % of the Time % of the time Rain Attenuation [db] XPD [db] Figure 1: Fraction of the time veru the rain attenuation. Figure 2: Fraction of the time veru cro polar dicrimination D =.3 m D =.6 m D = 1 m.7 max θ Frequency [GHz] Figure 3: The motion of the antenna mat a twit and way. Figure 4: Maximum allowed twit/way. caue co-channel interference and due to that the channel capacity of wirele communication ytem employing orthogonal polarization in order to double the capacity will be reduced. The depolarization i commonly quantified by the reduction in the cro-polar dicrimination (XPD). The minimum XPD occurring during.1% of the time i related to the rain attenuation exceeded during.1% of the time. Mot model and the tandard ITU model [3] ue thi fact to etimate the XPD. A hown in Fig. 2, a horter path length, i.e., 1 km ha larger cro-polarization dicrimination a compared to the longer one. Moreover, reducing.1% of the time to.1% will reduce the cro-polarization dicrimination. At.1% of the time the XPD can drop to 15 db at a path length of 1 km. The vertical gradient of the atmophere refractive index can caue variation on the angle-ofdeparture and -arrival. However, baed on the ITU model and data [6], thee variation are not o ignificant for a path length of 1 or 1 km. For intance, the variation for.1% of the time i.9 degree for a 1 km path length. Thi i quite mall a compared to the beamwidth of a typical high-gain E-band antenna ( ). The movement of the antenna mat i very relevant becaue high-gain antenna with mall beamwidth are ued for E-band backhaul ytem. Thi motion i bet decribed a twit and way of the antenna mat, and illutrated in Fig. 3. Thi i mainly happening when the wind blow. Due to mat movement the propagating wave will no longer leave/enter the tranmit/receive antenna at the maximum of it radiation pattern. Thi lead to an additional attenuation. The radiation pattern of the parabolic reflector antenna i ued to etimate the 3 db beamwidth. Thi etimation i repeated for three reflector diameter. The 3 db point will et the pecification on the maximum allowed twit and way. Sway and twit exceeding the correponding value in Fig. 4 will caue a ignificant degradation of the communication link. The wind force can caue twit or way up to 1 degree.

4 Progre In Electromagnetic Reearch Sympoium Proceeding E-BAND ANTENNA DESIGN The propoed configuration conit of a ymmetrical reflector antenna and a focal plane array (FPA) a the feed antenna. Thi configuration i depicted in Fig. 5. The elected diameter i 95λ (i.e.,.4 m) and deliver a maximum gain of 49.5 dbi at 71 GHz. The elected diameter i needed in order to compenate the total attenuation for the extended range of 1 km. The antenna gain i limited by the FCC regulation for E-band [7]. The configuration i modeled by uing phyical optic (PO) and a plane wave incident to the reflector antenna [8, 9]. The encircled power analyi [1, 11] i then applied in the focal plane to etimate the FPA ize a a function of F/D, can angle, and antenna efficiency. The efficiency here i the available power on the urface of the FPA dik normalized by the power captured by the reflector. The required can angle i +/ 5 degree, thi i to track the mat movement and to ait the mechanical alignment. Due to the mat motion (twit and way), the main beam i moving within a cone, therefore the topology of the FPA i choen to be a circular dik. In Fig. 6 the efficiency a function of FPA ize i plotted for practical F/D ratio, and at the maximum required can angle of 5. An efficiency criterion of 8% criterion i ued to elect an appropriate F/D ratio and the FPA ize. Thi criterion i to enure high efficiency for the extended range of 1 km. F/D =.6 yield the mallet FPA ize and thi i important to minimie feed blockage. Thi mean that the FPA ha a 3 mm radiu. The trade-off i that the main beam in the focal plane i concentrated in a mall region around the focal point (ee Fig. 7). With the aumption that the antenna element pacing i equal to λ = 2. Practically, thi limit the number of antenna to about 28 element per can angle (ee Fig. 8). Therefore the EIRP can not be improved by exciting more array element to create the beam. However, thi can be improved by axially diplacing the FPA in order to broaden the field D = 95λ, Scan angle θ = 5 degree FOD =.4 FOD =.6 FOD =.8.6 Efficieny FPA Radiu (mm) Figure 5: Artit impreion of the FPA and reflector for point-to-point E-band wirele communication. Figure 6: Encircled power analyi i applied in order to etimate the FPA ize veru efficiency. (mm) Relative Power Level (db) Relative power (db) θ = θ = 1 θ = 2 θ = 3 θ = 4 θ = 5 D = 95λ, θ : Scan angle, F/D = (mm) FPA radiu (mm) Figure 7: The field ditribution image in the focal plane. Figure 8: Cut of the field ditribution along the beam can.

5 916 PIERS Proceeding, Prague, Czech Republic, July 6 9, D = 95λ, Scan angle θ = degree Efficieny dz = dz = 1λ dz = 2.3λ dz = 3λ Figure 9: Model. Axial diplacement of the focal plane FPA Radiu (mm) Figure 1: Efficiency veru FPA ize for a range of axial diplacement (dz). (F/D =.6). Efficieny D = 95λ, Scan angle θ = 1,3, and 5 degree θ = 1 θ = 3 θ = 5 θ = 1, dz=2.3λ.2 θ = 3, dz=2.3λ.1 θ = 5, dz=2.3λ FPA Radiu (mm) Figure 11: Efficiency of the axially diplaced focal plane a a function of can angle. Relative Power (db) θ = θ = 1 θ = 2 θ = 3 θ = 4 θ = 5 D = 95λ, θ : Scan angle, F/D = FPA radiu (mm) Figure 12: Cut of the field ditribution along the beam can of the axially diplaced plane. ditribution. The axial diplacement (dz) i with λ tep toward the reflector a depicted in Fig λ i choen a the optimum axial diplaced focal plane. By exceeding it, the main bean of the focal plane pattern i going to be bifurcated. The efficiency curve of it i le teep and indicate the effect of broadening. Moreover, the 8% efficiency criterion till give a mall enough ize of the FPA (ee Fig. 1). In Fig. 12 the efficiency curve of the can beam cenario, how a beam broadening up to 3. Increaing the angle of incidence the effect of broadening i decreaed becaue the canned beam i getting narrower. We believe thi i happening becaue of ome focuing effect by the reflector. The beam broadening a hown in Fig. 12 indicate that the number of antenna element can be increaed ignificantly. About 78 element will be involved up to 3 and to a le extent at 4 and 5. Hence, EIRP can be improved and becaue of the overlap of the beam, array element reue i poible. 4. CONCLUSION High gain antenna at both the tranmit and receiver ite of the backhaul communication link are neceary to deliver acceptable level of SNR. Thi i due to the rain attenuation and the large freepace attenuation. A a conequence of the high gain antenna requirement, the radiation pattern ha a very narrow beamwidth. Due to thi, the mat movement i very relevant. Thi problem ha to be tackled, otherwie, additional attenuation will occur, reulting an outage. With the propoed antenna configuration of the FPA thee iue can be overcome. The elected value of the reflector antenna diameter, F/D ratio, and FPA ize are a good tarting point for an optimized antenna deign.

6 Progre In Electromagnetic Reearch Sympoium Proceeding 917 ACKNOWLEDGMENT Thi work i upported by the Netherland Foundation of Technical Science (STW). REFERENCES 1. Rec. ITU-R P.676-6, Attenuation by atmopheric gae, Dyadyuk, V., J. D. Bunton, and Y. J. Guo, Study on high rate long range wirele communication in the and GHz band, Proceeding of the 39th European Microwave Conference, Rec. ITU-R P.53-15, Propagation data and prediction method required for the deign of terretrial line-of-ight ytem, Rec. ITU-R P.838-3, Specific attenuation model for rain for ue in prediction method, Rec. ITU-R P.837-6, Characteritic of precipitation for propagation modelling, Rec. ITU-R P.453-1, The radio refractive index: It formula and refractivity data, Federal Communication Commiion, Allocation and ervice rule for the GHz, GHz, and GHz band, Minnett, H. C., B. Mac, and A. Thoma, Field in the image pace of ymmetrical focuing refelctro, Proceeding of the IEE, Vol. 115, , Ng Mou Keha, N. and L. Shafai, Characterization of dene focal plane array feed for parabolic reflector in achieving cloely overlapping or widely eparated multiple beam, Radio Sci., Vol. 44, Ivahina, M. V. and C. G. M. van t Klooter, Focal field analye for front-fed and offet reflector antenna, 23 IEEE Antenna and Propagation Society Internationa Sympoium, Vol. 2, , Hayman, D. B., T. S. Bird, K. P. Eelle, and P. Hall, Encircled power tudy of focal field for etimating focal plane array ize, 25 IEEE Antenna and Propagation Society Internationa Sympoium, Vol. 3A, , 215.