Experimental implementation of an Ultra-Wide Band MIMO radar

Transcription

1 Experimental implementation of an Ultra-Wie Ban MIMO raar Ettien Lazare Kpré, Thomas Fromenteze, Cyril Decroze, Davi Carsenat To cite this version: Ettien Lazare Kpré, Thomas Fromenteze, Cyril Decroze, Davi Carsenat. Experimental implementation of an Ultra-Wie Ban MIMO raar. Eumw-Eura 215, Sep 215, Paris, France <hal > HAL I: hal Submitte on 2 Oct 215 HAL is a multi-isciplinary open access archive for the eposit an issemination of scientific research ocuments, whether they are publishe or not. The ocuments may come from teaching an research institutions in France or abroa, or from public or private research centers. L archive ouverte pluriisciplinaire HAL, est estinée au épôt et à la iffusion e ocuments scientifiques e niveau recherche, publiés ou non, émanant es établissements enseignement et e recherche français ou étrangers, es laboratoires publics ou privés.

2 Experimental implementation of an Ultra-Wie Ban MIMO raar Ettien L. Kpré, Thomas Fromenteze, Cyril Decroze, Davi Carsenat, Xlim Research Institute, Limoges University, France ettien-lazare.kpre@unilim.fr Abstract In this paper, a raar system implementation is aresse to etect a point target, that is a Multiple-Input Multiple-Output (MIMO) Raar. It is an emergent concept in raar fiel an has reache a substantial consierations. The avantages of MIMO raars is presente in comparison of an raars. The simulation an measurement results show that the MIMO Raars can reach a better angular resolution while keeping a small number of antennas. I. INTRODUCTION Microwave imaging has been an important fiel of research an evelopment, thanks to its performances in the terms of military an civilian applications. MIMO raars have receive great interest over recent years [2]-[3] ue to their benefits compare to conventional raars. The essence of this concept is to probe the channel with M orthogonal signals an recor the backscattere signals with N receivers which are spatially inepenents to the transmitters. Thus the receive signal from each transmitter can be separate from others by correlation in pairs. The scattere signal for each couple transmitter/receiver provies two main benefits : spatial iversity gain [1] an increase egree-of-freeoms (DOFs)[4]. Each couple provies an information about the probe channel as shown in Fig.1. H 11 H H 1n H 21 H H 2n..... H m1 H m2... H mn Figure 1: Illustration of a Raar MIMO scattering matrix. H i j enotes the scattere signal from the i th transmitter to the j th receiver. Accoring to this general efinition, many traitional raars can be consiere as a special case of MIMO raar. The synthetic Aperture raar which consists on moving antennas to create a synthetic aperture will measure informations corresponing to the iagonal of the channel matrix H, whereas the Single-Input Multiple-Output raar is similar to the measurement of one row of the matrix H. The remainer of the paper is organize as follows: MIMO scattering matrix formulation is first consiere in the Section II focusing on the benefits of MIMO raars. Simulation an experimental results are then presente in Section III & IV to valiate the concept. II. MIMO SCATTERING MATRIX FORMULATION The term MIMO means any raar system that probes a channel by transmitting inepenent waveforms an receive with some specific signal processing. For a one-shot measurement of the matrix elements, waveforms shoul the orthogonal in orer to separate each couple of transmitter/receiver. Thus, waveform esign is significant issue associate with MIMO Raar [5], but it is not aresse in this paper. For every examples iscusse here, signal separation has been exploite to enable the scattering matrix elements H i j. s 1ptq s iptq s mptq x m R s1pt τ1jq h1j sipt τijq hij smpt τmnq hmn jr Orthogonal waveforms T itmt y n n R Target Figure 2: Illustration of a Raar MIMO principle. h i j represents the MIMO matrix components ue to i th transmitter an the j th receiver. x m an y n are respectively the transmitters an the receivers positions. The receive signal from nt transmitters can be expresse in frequency omain as : nt 1 S R = i= [H].S i + N G (1) where N G is the aitive Gaussian noise, S the Fourier transform of transmitte signals, H the scattering matrix that contains the probe channel informations an expresse as : (H) m,n exp( jk.(x m + y n )..sin( )) (2) where refers to the signal Direction of Arrival (DoA) while consiering the target at far-fiel istance from the raar system an k the wavenumber. This channel matrix contains informations referring to the target parameters. After estimation of the channel matrix, a classic beamforming in time omain algorithm is performe to etect the target position.

3 III. SIMULATION RESULTS Let s consier a monostatic MIMO array of M transmitters an N receivers (M = N = 4) with an inter element spacing of =.7 λ c. A point target with an isotropic RCS is place at a istance r > 2 D 2 /λ c (far-fiel conition) with λ c the central wavelength in the ban 3 6Ghz. Fig.3 shows the simulation setup. Fig.5 shows a beamforming comparison when Fig.3 array is use in conventional raars an in MIMO raar. Target with isotropic RCS s 1 (t) s 2 (t) =.7 λ c s 3 (t) D s 4 (t) uncoherent signals Beam magnitue (B) 1 2 MIMO Figure 3: Computing setup. A target with an isotropic RCS place at the far-fiel of a monostatic raar array of 4 elements. When this array is use in a conventional phase raar system, each antenna transmit a full scale waveform forming a strong transmit beam in the esire irection. Beamforming is performe only by the receivers array. Thus, the egree of freeom is N. However MIMO beamforming is performe by both transmitters an receivers, increasing the egree of freeom up to M N. While conventional raars employ only spatial iversity, MIMO raars employ both spatial an waveform iversity to improve spatial resolution, targets parameters ientifiability an iscrimination. It has been shown in [3] that MIMO raar provies the same resolution with its equivalent phase raar while keeping a small number of antennas. This phenomenon is explaine by the spatial convolution of the transmitters an receivers arrays that creates an enlarge virtual array aperture. In this example the virtual array consists of N+M-1 elements with an inter-element spacing of =.7λ c. Some elements are reunant as shown in Fig.4. As consequence, the esign of the MIMO raar array is an important issue associate with MIMO Raars. array { } = { } array (Degree) Figure 5:, an MIMO imaging results. The computing results show that the provies a better angular resolution (1 ) compare to the raar (18 ) but the sielobes level are very high. In contraiction, the raar presents lower sielobes but the angular resolution is worse. Thus the MIMO raar, presents the same resolution as the an the sielobes levels a lower. In the case of a bistatic array, the MIMO raar resolution can be improve with an appropriate array esign. Fig 6 shows the beamforming results with ifferent inter-element spacing while keeping the same number of antennas. Consequently, a sparse array can be use in MIMO raars to create a fille virtual array which provies a better angular resolution. Beam Magnitue(B) 1 2 Beam patterns T=R T=2xR T=3xR T=4xR { } 5 (Degree) Virtual array Figure 4: MIMO Virtual array. Figure 6: Angular resolution comparison with ifferent inter-element spacing.

4 IV. MEASUREMENT RESULTS The presente MIMO theory has been valiate using a monostatic linear array. Fig.7 an Fig.8 show the measurement setup use for the experimentation. The array consists of 4 transmitters an 4 receivers Vivali antennas arrange along azimuth with an inter-element spacing of =.7 λ c. A foam plate is use to support the antennas. The scattering matrix is measure with a network analyzer through a 4 16 microwave switch in 3 6Ghz ban. Two metallic cyliners with isotropic Raar Cross Sections in azimuth are place in front of the raar system. sielobes with an accuracy angle resolution an ones presents a worse resolution with lower sielobes level. The influence of the antennas raiation pattern is consiere in the beamforming, which explains the ifference between the theoretical an the experimental curves. array Anechoic Chamber array Beam magnitue(b) 1 2 MIMO Microwave Switch Azimut angle(degree) Figure 9: Imaging results of two metallic cyliners. Pc+Matlab Ethernet Switch ZVL Figure 7: Experiment measurement synoptic. In such setup the coupling signal between antennas an the elay ue to cables must be calibrate in orer to extract only the echo reflecte by the targets. The scattering matrix use for the beamforming is then given by the equation 3: (H) m,n = (H measure H coupling ) H cable (3) Where : H cable : Scattering matrix ue to the cables. H coupling : Scattering matrix measure without any target. H measure : Scattering matrix measure in presence of targets. In aition, the resolution can be improve keeping the same number of antenna an those, thanks to the optimization of antennas array [3]. As an example, an other transceiver array has been set up with the same number of elements. In this case, the transmit array has been split in two subarrays space of N.7 λ c. A same scenario is teste with two ifferent arrays as shown in Fig.1. Imaging results show the performances of the secon array. The angle resolution is improve compare to the first one so that the angle estimation of the target is more accurate. The secon element spacing is chosen in orer to avoi information reunancy in the scattering matrix. (a) (b) Figure 8: Measurement setup of 4 4 / Vivali antennas. Two metallic cyliners are place in the anechoic chamber in front of the transceivers array. (c) () A igital beamforming is applie to focus the beam in the irection of the targets. MIMO imaging case supports the presente theory. It presents the best compromise resolution/sielobes. Whereas imaging present a high level of Figure 1: Imaging results of a pair of scissors from two ifferent MIMO setups. (a) Photo of the regular spacing array. (b) Imaging result of the scissors using regular spacing array. (c) Photo of the irregular spacing array. () Imaging result of the scissor using the irregularly spacing array.

5 V. CONCLUSION In this paper, an experimental MIMO system has been presente. A first antenna array was use to highlight the benefits of MIMO raars in comparison of conventional raars. The experiment results showe that MIMO raar increases the egree of freeom an improves angle accuracy. Furthermore, MIMO raars can reach a same resolution as traitional raars with a smaller number of antennas, reucing rastically the number of elements neee. Some other measurements were set up to exhibit the iscrimination capability of MIMO raar but are not presente in this paper. Future works will focus on the generation of orthogonal waveforms to probe the channel simultaneously an fast nearfiel raar imaging algorithms evelopment for high image quality. REFERENCES [1] P. P. Vaiyanathan an P. Pal, "MIMO raar, raar, an IFIR raar: a comparison", in Signals, Systems an Computers, 29 Conference Recor of the Forty-Thir Asilomar Conference on, 29, pp [2] D. W. Bliss an K. W. Forsythe, "Multiple-input multiple-output (MIMO) raar an imaging: egrees of freeom an resolution", in Signals, Systems an Computers, 24. Conference Recor of the Thirty-Seventh Asilomar Conference on, 23, vol. 1, pp [3] Li Jian an Stoica Petre, "MIMO raar signal processing". J. Wiley & Sons 29. [4] W.-Q. Wang an H. Shao, Phase-MIMO raar with frequency iversity for increase system flexibility, in Signal Processing, Communication an Computing (ICSPCC), 212 IEEE International Conference on, 212, pp [5] S. Ying, Z. He, H. Liu, L. Jun, an S. Gao, Binary orthogonal coe esign for MIMO raar systems, in Intelligent Signal Processing an Communication Systems (ISPACS), 21 International Symposium on, 21, pp. 1 4.