Distance Dependent Radiation Patterns in Vehcile-to-Vehicle Communications

SP Technical Research Institute of Sweden Distance Dependent Radiation Patterns in Vehcile-to-Vehicle Communications Kristian Karlsson, Jan Carlsson, Torbjörn Andersson, Magnus Olbäck, Lennart Strandberg, Mattias Hellgren SP Report 20145:07

Distance Dependent Radiation Patterns in Vehicle-to-Vehicle Communications Kristian Karlsson 1, Jan Carlsson 1 Torbjörn Andersson 2 Magnus Olbäck 3 Lennart Strandberg 4 Mattias Hellgren 5 1 Electronics Department, SP Technical Research Institute of Sweden, Borås, Sweden. 2 Volvo Cars, Gothenburg, Sweden. 3 Volvo Trucks, Gothenburg, Sweden. 4 ACTIA Nordic, Linköping, Sweden. 5 Smarteq Wireless, Kista, Sweden.

3 Abstract Distance Dependent Radiation Patterns in Vehicle-to- Vehicle Communications We present a method for repeatable measurements of antennas integrated on vehicles in vehicle-to-vehicle communications. The method can handle vehicles and heavy vehicles and results in distance dependent radiation pattern measurements, or near-field coverage measurements. The method is exemplified with measurement results. Key words: Channel models, radiation pattern, vehicles, V2V, measurement SP Sveriges Tekniska Forskningsinstitut SP Technical Research Institute of Sweden SP Report 2015:07 ISBN 978-91-88001-36-8 ISSN 0284-5172 Borås 2013-12-15

4 Contents Abstract 3 Contents 4 Preface 5 1 Vehicles containing antennas 5 2 Two-ray path loss model 6 3 Methodology 7 4 Results 7 5 Conclusions 11 6 Acknowledgment 11 7 References 11

5 Preface Vehicle-to-Vehicle (V2V) communications based on the standard ITS-G5 have received increasing attention past years. This standard uses a high carrier frequency (5.9 GHz) and at this frequency multi-path propagating is sometimes poor [1] and propagation though objects weak [2] resulting in lost packages. Moreover, antennas in these systems are usually mounted on vehicles, and affected by other components on the vehicles. This means that the antenna design and placement have significant impact on the outcome of the transmission. Previous studies show that radiation nulls caused by shadowing from the supporting vehicle can be avoided [3], but is even more challenging for e.g. heavy vehicles with trailers shadowing the signal. The wireless channel between the vehicles has been investigated and field tests show [1, 4, 5] that in rural and highway environments ground reflections have significant influence on the received signal. These channel measurement are compared to the two-ray path loss model with good agreement, which result in areas with low signal levels. Besides shadowing and channel properties, parts of (or auxiliary equipment on) the vehicle constantly affecting the antenna performance will cause radiation nulls [6] and far-field distance to increase [7, pp. 38] as the antenna inherently becomes larger. To summarize: good antenna design and accurate antenna placement are important and should be tested with repeatable methods. Therefore in this paper we suggest a repeatable radiation pattern method which can handle: Effect of ground plane reflection (two-ray path loss channels). Antennas affected by the supporting vehicle, with radiating nulls and long far-field distance as result. Complex test structures (cars, cars with trailers, heavy vehicles). Suggested measurement site is an outdoor range (typically asphalt plane) providing space for heavy vehicles and long measurement distances. Such ground plane is also a very likely ground plane when the ITS-G5 systems are operating. 1 Vehicles containing antennas In the introduction it was argued that a measurement site for V2V communications according to ITS-G5 standard needs to handle long measurement distances. Alternatively near-field measurements with known phase can be used, but near-field ranges for heavy vehicles are rare and expensive. Long distances are important as the distance to far field from the test object is sometimes large. According to [7, pp.38] distance to far-field (r) occurs gradually and 2 should be larger than r 2D where D is largest dimension of the antenna and is the wavelength. At 5.9 GHz the wavelength is approximately 5 cm and the formula simplifies 2 to r 40D. The dimension D of the antenna includes not only the antenna element itself, but also parts of the vehicle constantly affecting the antenna. This increases the far-field distance significantly, e.g. a reflection in the hood or rear view mirror one meter away from the antenna element in one direction results in a far-field distance of approximately 40 m. At measurement sites such as near-field measurement sites distance to far-field is not an issue. Besides the problem with handling large vehicles in these sites, much of the communication will take place within near-field range meaning that near-field data cannot simply be converted to far-field, which is what is normally done at such measurement ranges [8]. The near-field coverage must be considered.

6 2 Two-ray path loss model Investigations of the radio channel at 5.9 GHz shows that on highway and rural roads measured channel at 5.9 GHz [1, 4, 5] is similar to the two-ray path loss model. This indicates a poor multipath environment as reflections from other objects in the environment are dominated by the ground reflection. Areas dominated by destructive interference and/or situations with non-line-of-sight (LOS) scenarios can cause lost packages distributed randomly or in consecutive bursts [2]. At urban roads multi-path environment is more present, reducing the effect of ground reflection. To further examine the radio channel at rural roads a range measurement was performed on a straight rural road. In Fig. 1 it can be seen that measured data (which consists of an average of two range measurements) conform very well to simulated Tworay path loss model data. For the Two-ray path loss model a reflection coefficient (R) of 0.7 was used. The value of R is simply found by fitting simulated data with measured data: e e Prec, db GTx, db GRx, db 20log10 R 4 r r jkr1 jkr2 1 2 Where G Tx, db and G Rx, db are assumed equal for both direct and reflected path and r 1 and r are distances described in (2). 2 1 2 1 2 2 1 2 2 r h h d 2 2 r h h d Here d is distance between vehicles, h 1 height of transmitting antenna and h 2 height of receiving antenna. Fig. 1 tells us that radiation pattern measured at 30 m radius will differ from e.g. radiation pattern measured at 35 m. Fig. 1. Range measurement between Volvo S60 with ETTE_ANT antenna and measurement vehicle.

7 3 Methodology Discussions in Chapter I-III prompt new type of radiation pattern measurements performed at both large and several radiuses. We suggest a EUT standing still and a measurement vehicle with measurement antenna moving around the EUT, logging received power together with GPS position. The test setup is described in Fig. 2: a signal generator located inside the cabin of the vehicle under test is connected to the antenna under test via an RF cable. Vehicle in combination with antenna is our EUT. On the receive side a measurement antenna should be connected to a receiver (e.g. spectrum analyzer). The receive antenna should preferably have a ground plane to avoid effects of the vehicle supporting the antenna, and also a smooth radiation pattern reducing effect of the non-perfect direction of the measurement antenna towards EUT during movement of the measurement vehicle. We use a dipole over a ground plane (gain ~6 dbi). The receiver should be connected to a computer which log received power and GPS position. A measurement can now be performed as follows: EUT is placed at the origin of a coordinate system on the test site. GPS antenna is placed close to the measurement antenna on the measurement vehicle to reduce positioning errors. The measurement vehicle drives slowly around the EUT in e.g. a spiral shaped track. GPS position is sampled and received power is logged at each position sample. Fig. 2. Test setup. A signal generator located in the test vehicle (left) tranmits a CW signal to the measurement vehicle equipped with spectrum analyser, GPS receiver and computer with measurement software. 4 Results Two EUTs were tested: a Volvo S60 and a Volvo FH truck, both with an ETTE_ANT antenna module, which is a shark-fin like antenna module developed with in the ETTE project and consists of several antenna elements including a monopole element for 5.9 GHz. On the S60 the module was mounted on the rear part of the roof, centred between sides of the car; and on the FH truck the antenna module was mounted on the cabin roof at the right side in the front. Tests were performed on a large asphalt area where the EUT was placed at the origin of a polar coordinate system, see Fig. 3-4. The measurement vehicle drove slowly (~7 km/h) around the EUT in circles marked on the ground. During these rotations the left pair of wheels on the measurement vehicle was following the painted circles, which is natural as the measurement antenna was located on the left side. Measurements started at a radius of 50 m and after one revolution the radius was decreased to 45 m, and so on, down to approximately 10 m, see Fig. 5 for a typical GPS trace of a measurement. Repeatability was verified by measuring the S60 with antenna module twice, see Fig. 6. As can be seen in this figure the near-field coverage is very similar for both cases.

8 To emphasize the need for measurement at several radiuses, radiation pattern of S60 with antenna module was measured at two different radiuses (40 m and 45 m). As can be seen in Fig. 6 there can be large differences in the results depending on measurement distance. Finally we present measurement result from the suggested method, see Fig. 8-10. During measurement of the FH truck the measurement antenna was mounted at two different heights representing communication to car (measurement antenna height 1.50 m) and truck (3.64 m), respectively. Fig. 9-10 shows large areas with low signal levels, these are areas where signal quality could be improved. Presented result is interpolated by using linear interpolation between samples (interpolated data always goes through the data points dictated by the sample). Fig. 3. Aerial photo of skidpad. Picture from http://kartor.eniro.se/. Fig. 4. Test object Volvo S60 in the middle of Skidpad and measurement vehicle driving around it following circle lines on different radiuses. Upper right shows antenna mount at 150 cm height on the measurement vehicle.

9 Fig. 5. Typical GPS trace during the measurements. Each cross (+) is a logged GPS position. Fig. 6. Volvo S60 with ETTE_ANT antenna measurement repeated two times. Fig. 7. Radiation pattern of Volvo S60 with ETTE_ANT antenna measured at two different distances.

10 Fig. 8. Volvo S60 with ETTE_ANT antenna measured at GPS positions as descibed in fig. 5. Red line marks circle with 40 m raduis and blue line marks circle with 45 m radius. Colorbar unit: dbm. Fig. 9. FH Truck with ETTE_ANT antenna. Measurement height 1.50 m. Fig. 10. FH Truck with ETTE_ANT antenna. Measurement height 3.64 m.

11 5 Conclusions Due two ground reflection and possibly also reflections in the vehicle itself, radiation pattern measurements should be performed at different radiuses. In this paper a method for logging data to overcome this problem has been explained and measurement result is presented. The method also handles heavy vehicles. As mentioned in the summary, coverage is strongly influenced by the vehicle on which the antenna is mounted and is affected by large metallic objects such as the container on a truck. Measurement results show that behind a truck with container a significant part of the signal is lost. Previous studies comparing theoretical two-ray path loss models with ITS-G5 radio channel on rural roads and highway are strengthened with new measurement results. 6 Acknowledgment This work has been supported in part by The Swedish Governmental Agency for Innovation Systems (VINNOVA) within the ETTE project. 7 References [1] J. Kunisch, J. Pamp, "Wideband Car-to-Car Radio Channel Measurements and Model at 5.9 GHz," in 68th Vehicular Technology Conference, (VTC 2008-Fall), IEEE, Calgary, Canada, Sept. 2008. [2] K. Karlsson, C. Bergenhem, E. Hedin, "Field Measurements of IEEE 802.11p Communication in NLOS Environments for a Platooning Application," Vehicular Technology Conference (VTC 2012-Fall), IEEE, Québec, Canada, Sept. 2012. [3] K. Yamamoto, K. Ohno, M. Itami, "Improving performance of DS/SS-IVC under shadowing environment using LMS adaptive circular array antenna," Intelligent Transportation Systems (ITSC), 14th International IEEE Conference on, pp. 2074-2079, Washington, DC, USA, Oct. 2011. [4] J. Karedal, N. Czink, A. Paier, F. Tufvesson, A.F. Molisch, "Path Loss Modeling for Vehicle-to-Vehicle Communications," Vehicular Technology, IEEE Transactions on, vol.60, no.1, pp.323-328, Jan. 2011. [5] C. Sommer, S. Joerer, F. Dressler, "On the applicability of Two-Ray path loss models for vehicular network simulation," in Vehicular Networking Conference (VNC), IEEE, pp.64-69, 14-16 Nov. 2012. [6] A. Sambell, P. Lowes, E. Korolkiewicz, "Removal of surface-wave induced radiation nulls for patch antennas integrated with vehicle windscreens," Antennas and Propagation, IEEE Transactions on, vol. 45, no. 1, pp. 176, 177, Jan 1997.

12 [7] P-S. Kildal, Foundations of antennas a unified approach. Studentlitteratur, Sweden, 2000. [8] J.E. Hansen, Spherical near-field antenna measurements. Peter Peregrinus, United kingdom, 1988.

SP Sveriges Tekniska Forskningsinstitut SP Technical Research Institute of Sweden Our work is concentrated on innovation and the development of value-adding technology. Using Sweden's most extensive and advanced resources for technical evaluation, measurement technology, research and development, we make an important contribution to the competitiveness and sustainable development of industry. Research is carried out in close conjunction with universities and institutes of technology, to the benefit of a customer base of about 10000 organisations, ranging from start-up companies developing new technologies or new ideas to international groups. SP Technical Research Institute of Sweden Box 857, SE-501 15 BORÅS, SWEDEN Telephone: +46 10 516 50 00, Telefax: +46 33 13 55 02 E-mail: info@sp.se, Internet: www.sp.se www.sp.se SP Report 20145:07 ISBN 978-91-88001-36-8 ISSN 0284-5172 More information about publications published by SP: www.sp.se/publ