From the previous research inside the Antenna Group of the Public University of Navarra regarding the working principles of the human eye, one of the conclusions was that the reconstructed image in the surface of the retina is defocused, being this fact the only explanation that properly justifies the visual acuity (or angular resolution) that the human eye has.

The defocusing system is supported under the assumption of coherent coupling between the different cones (or photoreceptors), enhancing the signal/noise ratio but, in principle, creating a diffusion in the information (image) received by the cones. Luckily, because that coupling is coherent, there is no destruction of the information, so the process is reversible.

In this paper we will try to demonstrate the possibilities of recovering an image from the one we could have over the retina, as the neural networks of the human eye could be doing. The idea behind this experiment is the evaluation of the possibilities of having antenna system working with this particular philosophy, i.e., creating a diffusion of the received information using a CORPS (Coherently Radiating Periodic Structures) antenna system, obtaining similar performances in relation to angular resolution.

In principle, the idea is to create some kind of template to be applied to the defocused image recovering the perfectly focused image. The proposed method is addressed to images who had suffered a low pass filtering, and it will obtain an inverse mask of the low pass process the original image has been subjected to.

This method is based on simple frequency inversion techniques of the process’ template and its main feature is that it achieves an optimal trade-off between image restoration and inverse mask size. Therefore, if the degradation process is constant (over the entire image) and it affects several images, once obtained the inverse mask, we are able to perform fast image reconstruction by a simple spatial convolution of the degraded image with a small template. In practical cases, this processing could be implemented with a layer or two of neural networks or some electronic circuit connected permanently to the detectors combining the detected signal appropriately.

Several promising results will be presented, showing that it is really possible to recover the original image, even with enhanced resolution, by using the technique proposed here.

The possibilities to apply this technique to antenna system could open many possibilities to high resolution vision systems with very compact antennas, and it could represent a total revolution in the design of smart antennas, electronically scannable antennas, MIMO antennas, etc.

It has been widely recognized that the capacity in wireless communication systems can be greatly increased by exploiting environments with rich scattering such as urban areas or indoors. Independent spatial or polarization channels can be accessed by means of multiple antennas at both the transmitter and the receiver and the technique is thus referred to as Multiple-Input Multiple-Output (MIMO) communications. For a fixed total power and bandwidth, and with a matrix transfer function of independent complex Gaussian random variables, the MIMO wireless communication channel has an information theoretic capacity that (initially) grows linearly with the number of antenna elements.

Most often the increase in capacity is attributed to spatial channels and scalar properties of the propagating waves. Classically, it is well understood that the electromagnetic polarization of plane waves possesses two independent channels, or polarization states. However, it has been shown that the existence of multipath propagation can increase the number of polarization channels states and thus effectively increase the channel capacity of a multiple antenna system.

In this paper we address the potential gain of using MIMO techniques in order to increase the bandwidth efficiency in satellite communication systems. In particular, we consider the increase in channel capacity that is possible by exploiting satellite and polarization diversity (see Fig. 1). In addition, we investigate a number of different MIMO antenna configurations, e.g. electric/magnetic tripoles, the vector element antenna, the MIMO-Cube etc., in which the propagation environment is utilized to achieve diversity in space and polarization. We will compare these different antenna configurations in terms of the improvement in information theoretic capacity of the total transmission channel. In addition, we analyze the effect of using power control on the capacity of the different MIMO satellite diversity systems. Simulation results show that the achievable capacity from satellite diversity systems outperform that of a single satellite system (see Fig. 2) and the capacity increases (approximately) linearly with respect to the average SNR, which is in agreement with MIMO theory.

With the overwhelming evolution of wireless technology, wireless LAN and Bluetooth capabilities built into laptop computers are becoming a standard. Moreover, the growing demand of increased system capacity calls for recent achievements on Multiple Input Multiple Output (MIMO) techniques. Therefore, it is expected that the forthcoming generation of portable computers will be equipped with modern wireless MIMO interfaces.

The benefit of MIMO techniques in antenna systems can be optimized if mutual coupling is taken into account. In this paper we investigate the performance of a six Omnidirectional Wrapped Microstrip Antennas (OWMA) linear compact array for MIMO enabled laptops. This structure uses the TFT screen as a natural antenna ground plane. Thanks to the OWMA properties , each array element has a quasi omnidirectional (total gain) far field radiation pattern, which is desirable for mobile units.

An antenna array prototype (Fig. 1) has been built and tested. The experimental and theoretical results are compared and a very good agreement is achieved. It is shown that the OWMA characteristics are relatively insensitive to the influence of another nearby OWMA. In spite of very narrow antenna spacing, D=0.4 wavelengths, which corresponds to 10 mm distance between element adjacent edges, the OWMA elements maintain their operating bandwidth covering the ISM 2.4GHz band. The far field radiation patterns of each element remains almost not affected by the presence of the other elements.

The interaction between elements is evaluated in terms of mutual coupling and coupling-induced envelope correlation. It is shown that the highest values of those parameters occur for adjacent, centrally located elements and are -14 dB and 20 dB respectively. Lower values are obtained for any other pair of elements. Therefore the proposed 6-elements OWMA compact array meets the requirements of MIMO enabled terminals.

1. Introduction

We proposed a 6-port Multiple-Input Multiple-Output (MIMO) antenna for small user terminals such as PC card terminals [1]. This antenna achieves a low-antenna coupling and low-spatial correlation by utilizing antennas with different patterns and polarizations. Moreover, employing an antenna selection technique is effective in MIMO channel capacity enhancement without increasing the number of transmitters or receivers. However, this method requires many RF switches and the process of finding the optimum antenna combination is complicated. In this report, we propose an antenna selection method for the MIMO antenna using orthogonal polarization and radiation patterns. This method requires only a simple antenna switching circuit. The effectiveness of the proposed method is clarified based on a numerical analysis of the indoor propagation.

2. Proposed Antenna Selection Method
Two sets of three orthogonal dipoles as shown in Fig. 1 are employed for the antennas for the Access Point (AP) and the User Terminal (UT). Array length, d, is set to 0.43 wavelengths. In order to select 4 antennas from the 6 available antennas, the matrix switch shown in Fig. 1(b) is necessary. However, this switch requires many RF-switches and its configuration and control are quite complicated. In the proposed configuration shown in Fig. 1(a), only two Single-Pole Double-Throw (SPDT) switches are required because the vertically polarized antennas are always used and only two pairs of the horizontally-polarized antennas are alternately selected. By using this method, the calculation load for selecting the antennas can be reduced by at least 60%.

3. Numerical Analysis Conditions
Figure 2 shows the configuration of the indoor environment in the numerical analysis. The height, width, and depth of the room are 3 m, 12 m, and 15 m, respectively. The wall material is a perfect conductor and a Line-Of-Sight (LOS) condition is assumed. The position of the AP is (8 m, 10 m, 2 m), and the UT is placed randomly at z = 1 m. All 6 antennas are used in the AP, and 4 are selected from the 6 antennas in the UT. The imaging method is used for this analysis, the number of reflections is 10, and the frequency is 2.4 GHz.

4. Experimental Results
Figure 3 shows the cumulative distribution function of the Signal to Noise Ratio (SNR) at the UT. Based on this graph, the SNR of the vertical dipole is higher by 1 to 2 dB than that of the other dipoles. Figure 4 shows the channel capacity obtained using the proposed method. For comparison, the results of the capacity with the ideal antenna selection are shown. The results show that the proposed method agrees with that of the ideal antenna selection even with a simple antenna selection circuit.

5. Conclusion
We proposed a simple antenna selection method in which 4 antennas in 6 are selected and only 2 SPDT switches are used. Based on the numerical analysis, the proposed method can provide channel capacity that is as high as that provided by the ideal antenna selection.

Reference
[1] N. Honma et al., ISAP 2005, vol. 2, TB2-4, pp. 367-370, Aug. 2005.

The constant increase of wireless communications came at the cost of the need for more bandwidth, aiming more quality and diversity of services. With this increase some important points for discussion have come forward, such as the systems capability to grow enough to suit to the need. Progress brings also the need for the reconfigurability, which is essential for the interoperability between different radio systems. If the radio can be reconfigured, people do not have to change their radio devices to access different services, as mobile telephony, GPS and even HDTV.

MIMO systems have made use of both spatial diversity and in some cases polarization diversity to increase channel spectral efficiency. Also some examples of multimode antennas, i.e. radiation pattern diversity, have been proposed [1], [2]. However, in all these examples the multimode antennas were made with electrically large antennas with very uncorrelated modes, that replace a Multiple Input or Output terminal made of several small antennas for only one multimode antenna, i.e. changing spacial diversity for radiation pattern diversity.

The purpose of this work is in another direction; the idea is not to substitute the use multiple antennas for one element with multiple modes but to combine multiple compact multi-mode antennas. To this aim, first, compact multimode (two or three modes at the same frequency) patch antennas with sizes around 0.5 wavelength have been designed. These designs combine the lower order modes (01, 11 and 21 Modes) of circular and short-circuited ring patch antennas, thus allowing compactness but having a high modal correlation.

After that, a study of channel spectral efficiency using these antennas was performed. A channel model has been developed that takes into account not only the spatial correlation, but also the modal correlation. It has the advantage that allows a close formulation for modal and spatial correlation exclusively dependent on the antenna geometry and radiation patterns and in the channel angular spread. The spectral efficiency was computed comparing several multi-element-multimode antennas scenarios to alternative scenarios with single-mode antennas. The results show the advantage of this proposal for the use in size constrained terminals.

Fig. 1 shows the radiation pattern diversity of the two mode compact patch antenna whilst in Fig. 2 an example of the comparison of spectral efficiency for this two-mode receiver configuration is show (N is the number of antennas in the terminal, D is the number of modes of each antenna and M=ND). The two-mode antenna proposal performs better than an equivalent scenario with N = 3 and D = 1 and in this situation it should be highlighted that the antenna size proposed is smaller and just two radiating elements are being used compared to three. The two ideal configurations, i.e, when there is no modal correlation and when the number of single-mode antennas in the receiver equals N = 4, are shown for reference purposes. Similar results have been found for three-mode antennas proving the viability of the use of compact mutimode antennas in MIMO terminals. Polarization diversity could also be added to these antennas.

[1] C. Dietrich, K. Dietze, J. Nealy, and W. Stutzman, “Spatial, polarization, and pattern diversity for wireless handheld terminals,” IEEE Transaction on Antennas and Propagation, vol. 49, pp. 1271 – 1281, Oct. 2001.
[2] T. Svantesson, “On the potential of multimode antenna diversity,” in IEEE Vehicular Technology Conference, vol. 5, pp. 2368 – 2372, 2000.

Keywords : mutliple antennas, mobile phone, PIFA, MIMO capacity

The data flows required for mobile communications are steeply increasing to satisfy the growing need in multimedia services as live television or videophony. In the 4G, transmission rate are expected to be higher than 300 Mbps against only 2 Mbps in the 3G with HSDPA systems. MIMO technology is a worthwile solution to meet this aim. By combining antenna arrays at the transmitter and at the receiver side, it allows to increase the information transfer rate without changing the transmission power nor the frequency range allocated by taking advantage of paths multiplicity and antenna diversity [Fos98]. However, MIMO technology raises the problem to introduce multiple antennas on a small mobile terminal while maintaining diagram diversity. This problem represents an important challenge and we propose here to study a radiating structure combination on a small ground plane.

We propose here two antennas configurations composed of two Planar Inverted-F elements placed on a ground plane whose dimensions (44 x 100 mm²) are compatible with a mobile phone and working at 2 GHz. In a first configuration, the two elements are placed in perpendicular (Fig.1) to obtain polarization diversity in the UMTS band. Return losses lie between -8.97 and -19.58 dB for one element, and between -10.57 and -16.75 dB for the other. The gain is 2 dB higher for the first element. The bandwidth is 20%. In a second configuration, the radiating elements are placed in parallel to obtain diagram diversity. The distance between elements (lambda/10) was chosen to keep isolation upper than 10 dB. Return losses for the two elements are less than -17.6 dB and the bandwidth is about 31%. For these two configurations, the correlation coefficient of diagram is less than 0.011 and the isolation is better than 11 dB. So these antennas are suitable for MIMO technology.
The capacity efficiency of a mobile antenna with parallel configuration was analysed by using a ray propagation model in a typical cellular environment [Ver05]. The base station used as transmitter was composed of four vertically polarised elements separated from 0.37lambda. Increasing the number of transmitters have not a large influence due to a low space diversity and a low angle of departure of the ray paths. But increasing the number of receivers increase capacity. For a SNR of 10 dB, the capacity increases of about 47% between SISO (1x1) and MIMO (2x2).

Two diversity antennas with dimension compatible with a mobile phone have been designed. We have shown that they are siutable for MIMO use by their very low diagram correlation and by testing then with a ray model in cellular environment.

References :[Fos98] G.J. Foschini, "Layered space-time architecture for wireless communication in a fading environment when using multiple antennas", Bell Labs Tech. J., vol. 1, no. 2, pp.41-59, Autumn 1996.
[Ver05] S. Vergerio, J-P. Rossi, J-M. Chaufray, P. Sabouroux, "MIMO Capacity simulation in cellular environment with a ray model", European Test and Telemetry Conference, Toulouse (France), June 2005.

Abstract

In this paper, the behaviour of the human eye is analyzed obtaining different solutions for the main common trade-offs of antenna array systems, in particular regarding the angular resolution and the signal/noise ratio.

Normally, to improve the angular resolution, more directive beams close to each other are needed. This is difficult because of the needed overlapping of the effective radiating areas of the different independent beams.

On the other hand, to solve to problem of signal/noise ratio, a heterodyne detection is preferred instead a direct detection, since the power balance of the system ensures more power at the receptor position.

As we will see, the human eye uses very different techniques to solve these problems. In fact, both problems are solved with the same strategy, using many photorepectors, cones, to create each of the beams, and re-using many of these cones for other neighbouring beams. Thanks to this recycling technique of the cones, the effective radiating areas are practically overlapped and because are many cones detecting the same information coherently, the signal/noise ratio also improves.

The final result is well know for everyone and is that the "antenna system" of the human eye works really very well using very simple detection techniques.

Introduction

The Antenna Group of the Public University of Navarra has been studying the behaviour of the human eye trying to identify its behaviour with an antenna array. We pay special attention to the impressive performance of the human eye regarding the visual acuity, that translated to antenna language will be the angular resolution of the system.

Usually, in the conventional antenna array, every detector is responsible to extract the information from the area covered with the detector itself, the effective radiating area. The possibilities to increase the resolution of the obtained image pass by improving the individual directivity of the detectors, increasing the radiating areas. Additionally, this improvement in the directivity also helps to improve the signal/noise ratio.

However, if the size of the detectors increases the number of pixels at the focal plane would decrease, limiting the resolution of the obtained image. A trade-off is clearly established.

In the case of the human eye, the evolutional solution to this problem, results to be a quite optimum solution regarding this compromise. The cross-sections of the photoreceptors at the macula (cones) are of a diameter of 3 wavelengths. The human eye has an optical system of about 60 dioptres, i.e. a focal distance of 1/60 meters. This means that the size of the cone translated in front of the eye at a reading distance of 350 mm, is approximately 32 µm, subtending an angle of 18 minutes of arc.

This is clearly not enough to explain the performance of the human eye, since the minimal angle of resolution (MAR) is established to be 1 minute of arc, and the minimum visible threshold is approximately 1 second of arc. Therefore, we need to have more cones involved in the detection of one beam. This could explain the enormous overlapping between the different beams and the impressive signal/noise ratio achievable by the human eye.

It is quite easy to imagine the possibilities of applying this detecting philosophy in imaging systems, but also in many antenna systems where many different beams should be supported simultaneously.

For future high rate wireless systems, a very promising issue is to combine different communications standards in the same architecture. Moreover, it is well known that using multi element antennas (MEA) system improves significantly the performances of a wireless system. Traditionally, spatial diversity is only used at the receiver in a single input, multiple output (SIMO) configuration. Recently, multiple input multiple output (MIMO) methods, using antenna arrays at both transmitter and receiver sides draw more and more attention. Therefore studying performances of radio receivers combining SIMO techniques and Software Defined Radio (SDR) seems to be relevant. This summary describes a global system approach to easily develop, simulate and validate a MEA receiver structure for different transmission standards. At this time, we focus 802.11b and 802.11g standards and evaluate its performances under true operating conditions by the use of a 2x2 MIMO radio platform connected to the Advanced Design System (ADS) software from Agilent Technologies.

Global system evaluation approach for a MEA receiver

Using MEA at the receiver is a widely applied technique for reducing the effect of multipath propagation, which penalizes WLAN performances. Among all different MEA algorithms, Sample Matrix Inversion (SMI) was preferred for a first evaluation. Taking advantage of the training sequence used in 802.11b or 802.11g receivers, this algorithm estimates the optimal complex weights to apply to different received signals. At first, complete 802.11g and 802.11b transmission systems were developed with ADS to verify that this algorithm improves significantly Bit Error Rate (BER) performances in different channel conditions. Moreover, we have also to take into account fading correlation, Radio Frequency (RF) defaults coming from the antenna coupling and the impairments of the RF front ends. An efficient development way is to use a complete connected solution with MEA capabilities (figure 1). This connected solution allows to simulate and measure all parts of a wireless link, from the global system level down to the particular model of RF components or fading channels. The arbitrary waveform generator (ESG4438C) is able to generate any complex signal which is then possible to analyze after propagation with the vector spectrum analyzer (VSA89641).

Multi standard structure

In order to evaluate the potential of a SDR MEA terminal, a structure combining a 802.11b and a 802.11g receiver sharing the same SMI block was simulated. In hard channel conditions about 10 dB gain is observed for both standards with 4 antennas. Furthermore, a more complex structure sharing the same four MEA but with separated antenna processing for each standards is studied. A classical 2D-RAKE is used for 802.11b and space-time and space-frequency algorithms are compared for OFDM frames. Based on a BER criterion, the influence of all parts of link (channel characteristics, RF impairments...) could then be estimated. Our aim is at this time to add 802.11n functionalities. The 2x2 connected solution offers relevant experimental conditions to test MIMO modes of this new standard, in presence of other standards.

Introduction

This paper considers an adaptive array antenna for the interference suppression in Global Navigation Satellite Systems (GNSS). Two anti-jamming algorithms are proposed and these algorithms rely on the unique structure of the course/acquisition(C/A) code of the satellite signals. The effectiveness of those algorithms is presented by computer simulations.

Proposed Algorithms

Algorithm 1

Global Positioning System(GPS) is one of GNSS and employs direct-sequence spread-spectrum signaling. GPS uses the BPSK modulated navigation symbols. BPSK signal is a cyclostationary signal [1] and the correlation between the signal and its frequency-shifted version by twice the carrier frequency is very high. For the baseband signals, it means that conjugating the signal can make the reference signal. Therefore, MMSE(Minimum Mean Square Error) method can be applied by using the conjugate signal of array output as reference signal. The cost function is written as

This algorithm makes use of the temporal structure of the navigation symbols. Fig.1 shows the structure of the navigation signal. The BPSK symbols are spread by a C/A code with spreading gain G (G=1023 for GPS). The code sequence is repeated L times (L=20 for GPS) within each symbol. Therefore, the samples have the same values as the samples delayed by mG chips, where 1≤ m<L. The cost function is written as

Fig.1 C/A signal structure

Simulation Result

We evaluated the performance of the proposed method using the GPS C/A signals. The GPS navigation symbols are in the BPSK format and are spread by C/A codes with processing gain of 1023. The interference is a random noise jammer and has no the inherent self-coherence property like the GPS signal. The receiving antenna is 4 element linear array with a half wavelength spacing. The arrival angles of the GPS signal and jammer are 0 and 60 degrees, respectively. The input powers are 0 and 50 dB, respectively. The pseudorange are estimated by BASS method [2] using array output after the anti-jamming processing. Fig.2 shows the pseudorange error as a function of the input SNR(Signal to Noise Ratio) under the Rayleigh fading. The proposed adaptive array can improve the performance due to eliminate the interference.

Fig.2 Pseudorange error

Summary

The proposed algorithm can be run as a blind processing using the self-coherence property of GNSS signals. The effectiveness of those algorithms was also presented by computer simulations.

References

[1] W. A. Gardner, "Exploitation of Spectral Redundancy in Cyclostationary Signals," IEEE Signal Processing Magazine, April 1991.
[2] J. B. TSUI, "Fundamentals of Global Positioning System Receivers : A Software Approach," John Wiley & Sons, Inc., 2000.

In mobile communication systems the knowledge of the location of mobile users is an important issue, because this allows to improve the performance of the communication system by implementing smart antenna techniques, such as adaptive beamforming and interference cancellation. By means of an antenna array it is possible to estimate the location, or rather the direction-of-arrival (DOA) of the mobile users. In this paper, we simultaneously estimate the DOA and polarization of multiple plane waves incident on a uniform circular array (UCA). A UCA has some advantages compared to uniform linear arrays, e.g. a 360° scan angle in azimuth, as a result of the circular symmetry. If the UCA consists of identical antenna elements, which are diversely polarized, the UCA is capable of discriminating signals based on their polarization characteristics. This in contrast to UCAs consisting of uniformly polarized antennas. Since mutual coupling strongly affects the electromagnetic (EM) characteristics of UCAs we incorporate a rigorous EM description of the UCA into the formalism for estimating the polarization and the DOA of the impinging source signals. To describe mutual coupling in the UCA in a compact way, we exploit the symmetry in the UCA and expand the array manifold of the UCA into a limited number of phase modes. This number of phase modes only depends on the electromagnetic dimensions of the UCA and is independent of the severity of mutual coupling. Based on these phase modes, we apply a modified Root-MUSIC algorithm, which delivers estimates for the DOA and polarization of the impinging signals.

The exactness of the description of the electromagnetic characteristics of the UCA, including mutual coupling, and the proposed algorithm are verified by means of simulated data and measurements. By means of the Cramer Rao Bound (CRB), which defines the ultimate accuracy of any estimation procedure in the presence of noise, the efficiency of the proposed algorithm is studied. In the simulations the obtained Root-Mean Square Errors (RMSEs) of the estimates are close to the corresponding CRBs, which proves that the proposed algorithm is an efficient DOA estimation algorithm. To emphasize the generality of our method, UCAs consisting of wire antennas as well as of patch antennas are considered.

A current challenge in MIMO wireless system development is maximising capacity while working within bandwidth, power and complexity limitations. In MIMO channels, maximum capacity can be achieved when Channel State Information (CSI) is fully known at the transmitter. Predictions of capacity of MIMO systems in the case of perfect CSI have been presented, for example. However, it is more realistic to expect that only partial CSI will be available at the transmitter due to complexity and latency. This case has been investigated from an information theoretical view. In this letter, a model for a MIMO system, which includes a moving receiver, is developed. The fast fading is the dominant factor affecting on the rate of feedback scheme. A subspace method is used to determine the productive invariant dimensions at transmitter. These dimensions help the receiver to use a low rate feedback channel. It is also demonstrated that the channel capacity is also increased compared with cases of no CSI or when only the strongest dimension is exploited.

Xu, X.; Sahai, A.; Morlat, P. F.; Verdier, J.; Villemaud, G.; Gorce, J. M.
INSA de LYON, FRANCE

Introduction

In many contributions related to Wireless Local-Area-Network (WLAN), the receiver side generally uses spatial diversity. This class of systems is often named single-input multiple-output (SIMO) and the combination of SIMO architectures with the multicarrier technique orthogonal frequency division multiplexing is very promising. Nevertheless, research has mainly focussed on systems impaired by additive white Gaussian noise (AWGN) and such analyses are not entirely sufficient. The influence of non-perfect oscillators, non-perfect power or low noise amplifier and I/Q gain and phase imbalance on the performance of SIMO-OFDM system can be very critical and must be studied precisely. Moreover, the impact of RF impairments on a receiver depends on RF front-end topologies. An alternative wireless receiver architecture to the well-established superheterodyne is the Direct conversion consist , particularly for highly integrated, low power terminals. Consequently, it`s very important to analyse and to quantify the influence of RF impairments in SIMO architecture versus front-end architecture.

Evaluation of RF impairments impact

In this paper, 802.11g standard has been investigated in terms of Bit Error Rate (BER), constellation of each subcarrier and received spectrum at the receiver. Simulated and measured results have been studied in the four arms SMI (Sample Matrix Inversion) receiver as compared with a single antenna receiver. Advanced Design System (ADS- Agilent Technologies) software and a radio test-bed have been used in this work. Firstly, the influence of phase noise spectrum of local oscillators is discussed in detail. Several phase noise models have been studied, among them the classical Wiener process. One of these models, which considers the phase noise as a superposition of several components with different spectrum signatures, shows greater coherence with actual measurements. In Fig.1, Flicker frequency noise (slope of the phase noise spectrum in the Bode diagram is -30 dB/dec) is used to study the BER performance.

The effect of both, the power amplifier non-linearities and the LNA noise factor (NF), on system performance is investigated. For instance, the 1dB compression point and the third order intercept point are modeled to quantify the intermodulation distorsion characteristics of each system (fig.2 and fig.3).

A similar analysis has been made in the very important case of IQ imbalance in an AWGN channel with multipath fading. The SMI gain is also very interesting in such large angular spread conditions for IQ gain and phase imbalance. Finally, all the results presented in this paper and based on both simulations and measurements show how the impairments of the RF interface may strongly decrease the theoretical performance of receiving algorithms. The SMI processing associated with a SIMO channel can compensate for these effects, unlike a SISO receiver.

1.INTRODUCTION: It is more or less consensual that the provision of broadband capabilities in wireless communications will have to resort to multiple Tx and Rx antennas to exploit the rich scattering properties of wireless channels (MIMO). On the other hand multicarrier code-division multiple access (MC-CDMA) has been considered for the past few years by many researchers to be a very promising candidate as the access scheme for Beyond 3G mobile communications systems. This technique benefits from the features of OFDM and CDMA. One important aspect of this technique is the fact that it takes advantage of frequency diversity since signal spreading is performed purely in the frequency domain, making possible to use diversity combining based techniques for signal detection. To exploit the diversity inherent in STBC (Space-Time Block Codes) have been proposed. One of the simplest schemes is the Alamouti space-time code which does not require very complex decoders and can be integrated in the existing systems without needing much redesign. The basic Alamouti scheme which is a 2x1 (2 Tx antennas and 1 Rx) can be extended to using in parallel multiple Alamouti codes. The advantage of using a coding scheme based on multiple Alamouti is that we can lower the modulation constellation size used in the system while maintaining the same data rate and bandwidth. This in turn will allow better tolerance to errors in phase and automatic gain compensation. In this communication we consider the use of a Double Alamouti coding scheme, design the space-time equalizers and compare the performance against the conventional 2x1 Alamouti. 2.DOUBLE ALAMOUTI: Fig.1 shows the Double Alamouti scheme we propose (4x2), for the Downlink of a system using QPSK modulation, as an alternative to a standard Alamouti (2x1) scheme implemented in a system using 16-QAM modulation. The reduction in constellation size allows having better tolerance to phase and amplitude deviations caused namely by AGC or RF high power amplification. The comparison is made between these two systems since they provide the same spectral efficiency and bandwidth. 3.ILLUSTRATIVE RESULTS: To evaluate the performance of the proposed scheme, we used an implementation of a MIMO channel based on the 3GPP Spatial Channel Model, associated with the ETSI BRAN E channel model for the power delay profile (PDP). This scheme should be an interesting option when its extra complexity is not excessive for practical implementations. When compared against single Alamouti, it allows reducing the modulation order by a factor of 2 while preserving bandwidth efficiency, i.e. the throughput achieved with the combination of single Alamouti - 16QAM can now be reached with Double Alamouti - QPSK, without sacrificing spectral efficiency. The reduction of the order of the modulation scheme allows using modulations that are more tolerant to impairments (phase/ frequency errors, nonlinear distortion etc) while the increase in the number of antennas provides better performance, as shown in Fig.2. In the final communication we shall present the detailed mathematical background and further results in order to evaluate the proposed Double Alamouti in realistic scenarios.

The aim of our research is to create an increasingly realistic transmission system. An electromagnetic phenomenon seldom taken into account in the modelling of the transmission systems but which is unquestionably disruptive is the cross coupling between antennas. In order to correct this existing negative effect in our communication connections and thus to improve detection of our signals, we have integrated this parameter in our modelling. In this way, we can limit degradation on the performance of a transmission link using a 2x2 MIMO system. Indeed, it's rare that the disadvantages of the antennas are taken into account in modelling in baseband of a numerical system. This work enables us to establish the link between the signal processing and the study of antennas.

First, this article presents our numerical modelling with the effect of cross coupling on the signal processing. This cross coupling has the same representation at the emission or at the reception parts. There are the parameters noted α at the emission part and β at the reception part which model this cross coupling between antennas. Mathematically speaking, they are complex numbers used in module and phase.

This study is undertaken on the Alamouti Scheme of a 2x2 MIMO configuration. So, an additional digital processing is carried out on this system in order to take into account the effect of this cross coupling in our transmission system.

This system is initially studied in an analytical way, what makes it possible to understand the general behaviour of the connection like envisaging its performances according to the cross couplings at the emission part α and at the reception part β, but also according to the SNR_TX (Signal to Noise Ratio) at the emission. These numerical calculations thus enable us to know the behaviour of the received SNR_RX (Signal to Noise Ratio) according to these three parameters, SNR_RX=f(α,β,SNR_TX), with f a mathematical function. Digital simulations make it possible to confirm the results of these calculations and give us the behaviour of the BER (Bit Error Rate) of this system according to the same three parameters.

Finally, a comparison between analytical and numerical studies is made. The effects of cross coupling on the link budget are shown. Also, thanks to an adapted digital processing we can limit the negative effect of this cross coupling by taking into account the a priori knowledge of the couple (α,β) given by the antennas builders.

High Altitude Platforms (HAPs) are quasi-stationary aerial platforms operating in the stratosphere. This emerging technique is preserving many of the advantages of both satellite and terrestrial systems and presently started to attract more attention in Europe through the recently formed European Community COST 297 Action, in which the authors are the Swedish representatives to the Action. Using narrow bandwidth repeaters on HAP for high speed data traffic have several advantages compared to using satellites, especially when operating in a local geographical area. One of the main advantages is that the received signal from the HAP would be much stronger than a received signal of equal transmitted power from a satellite. This allows for a much lower sufficient transmitter power which would decrease the size and weight of the repeater equipment carried by the HAP. Also the HAP provides for a much easier deployment so that a high-speed connection can be made on demand for a specific geographical area. It has been widely recognized that the capacity in wireless communication systems can be greatly increased by exploiting environments with rich scattering such as urban areas or indoors. For a fixed total power and bandwidth, and with a matrix transfer function of independent complex Gaussian random variables, the MIMO wireless communication channel has an information theoretic capacity that (initially) grows linearly with the number of antenna elements. Constellations of multiple HAPs have been shown to enhance broadband fixed wireless access capacity by exploiting antenna user directionality, when using shared spectrum in co-located coverage areas, where a predominant LOS propagation is present for mm-wavebands (e.g., 47/48 GHz). In addition, High Altitude Platforms (HAPs) have been also proposed for 3G and broadband applications where multipath propagation might be significant. The main idea in this paper is to create a virtual MIMO system by exploiting the diversity provided by multiple HAPs (see Figure. 1) in order to increase the data rates in these systems. In addition, a number of different compact antenna array configurations, (e.g., the vector element antenna, the MIMO-Cube, etc.) specifically designed for MIMO applications have been proposed recently, in which the propagation environment is utilized to achieve diversity in space and polarization. Thus, in this paper we also investigate the effect of using different compact MIMO polarization antenna configurations, separation angles between HAPs and power control on the information theoretic capacity of the total transmission channel of the HAP system. In addition, since the special compact MIMO antenna array configurations depend on the array elements being positioned very closely together, the effect of mutual coupling will be also analyzed and taken into account when performing the simulation of this HAP diversity system. Simulation results show the capacity benefit of employing HAP diversity over SISO (single HAP) system (see Fig. 2). Fig 1. Multiple HAP diversity system. Fig 2. A comparison of the acheivable capacity using the MIMO Vector antenna and MIMO Cube antenna with a system employing a single HAP.

The increasing demand for mobile communication services without a corresponding increase in RF spectrum allocation motivates the need for new techniques to improve spectrum utilisation. When planning deployment of multiple competitive networks that operate in adjacent frequency bands and in the same geographical area, different techniques need to be applied in order to mitigate the interference and allow spectrum-efficient coexistence between the networks. Deploying smart antennas at base stations accomplishes both of the above mentioned goals for time- or frequency-division multiple-access (TDMA/FDMA), and for code-division multiple-access (CDMA) cellular systems.

In general, two main categories of smart antennas are distinguished: switched beams and adaptive arrays. Using smart antennas, the capacity improvement and radio coverage extensions are achieved by increasing the carrier-to-interference ratio (CIR) via three basic approaches: mainbeam steering, multipath suppression, and interference nulling.

The direct effect of smart antennas on coexistence is due to the fact that the RF energy radiated by transmitters is generally focused in specific areas of the cell and is not constant over time. Smart antennas this way directly affect the likelihood of interference in coexistence scenarios where two different radio interface technologies operate in adjacent frequency bands and in the same geographical area. If this percentage is satisfactorily small, the current coexistence rules (based on the worst case) may be relaxed.

Current radio propagation planning tools used by regulative offices are unfortunately not able to take into account the mitigation effort of smart antennas, as they do not have any satisfactory model to simulate them. This paper introduces statistical models for switched beam antennas (SB) and adaptive arrays (AA) and investigates the performance measures of cellular radio systems with and without smart antennas deployed at base stations. A cross-border scenario in which an external interfering system on one side of the border interferes with a CDMA victim system on its other side was studied.

An example result is presented below. The figure shows the non-interfered capacity (C) of the CDMA uplink as a function of the user voice activity (v). The solid line represents the case when SB with measured radiation pattern values is used and the dashed one the case when statistical SB model is taken into account. Both lines agree well and show a four-fold increase in capacity when compared to that one of the sectored cell (dash-dotted line). The line with circles represents the case when SB antenna is replaced by an AA. Almost a two-fold gain in the non-interfered capacity compared to that one of the SB case can be observed.

The basic concept of diversity is to combine the multiple signals transmitted over different fading paths in order to reduce the effect of deep fades. Antenna diversity can greatly improve the signal quality at the receiving end of a wireless communication system. Spatial diversity is commonly used at the receiver base station where the receiving antennas are separated several wavelengths giving low correlation between the antennas [1]. The use of small arrays with closely spaced antennas at the receiver and transmitter, known as a Multiple Input Multiple Output (MIMO) wireless system, is known to increase the capacity of the wireless system in a Rayleigh fading environment [2]. It is also known that the mutual coupling between closely spaced antennas results in a lower correlation than was previously predicted [3].

In this work it is shown by simulations and measurements that a decoupling network can be used giving low correlation between closely spaced antennas [4]. Two possible circuits for an antenna separation of d=0.21ë are proposed. Simulations and measurements of the diversity gain and the capacity of the antenna system with the decoupling circuits are performed. The results show that the diversity gain obtained by connecting the antenna system to the circuit results in a performance comparable to an antenna separation of d=0.5ë. This is of great interest for the integration of MIMO systems where the area for the antennas is restricted like in mobile phones.

[1] W. C. Y. Lee, "Antenna Spacing Requirement for a mobile radio basestation diversity," Bell Syst. Tech. J., Vol. 50, pp. 1859-1876, July/Aug. 1971
[2] J. H. Winters, "On the capacity of radio communication systems with diversity in a Rayleigh fading environment," IEEE J. Select. Areas Commun., vol SAC-5, pp. 871-878, June 1987
[3] R. G. Vaughan and N. L. Scott, "Closely spaced monopoles for mobile communications," Radio Sci., Vol. 28, no. 6, pp. 1259-1266, Nov.-Dec. 1993
[4] S. Dossche, S. Blanch, J. Romeu, "Optimum antenna matching to minimise signal correlation on a two-port antenna diversity system," Electr. Lett., Vol. 40, no. 19, pp. 1164 - 1165, Sept. 2004

Diversity reception technique has been one of the well-known methods to mitigate and reduce the effect of fading especially in mobile communication. This method can be applied either at base station or at the mobile reception. This basic idea also can be applied for on-body communication as well as it provides the communication of the antennas in the vicinity of human body. Reducing the path loss gain is the main objective in order to minimise the battery consumption and reduce the inference of the signals. The performances of the antennas in two specific links: trunk communication (shown in Fig 1) and hand communication are investigated by using switched beam diversity antenna as shown in Fig 2 for different body postures. The results (as shown in Fig 3) have been compared with the combination of two monopole antennas result. As a conclusion, most of the cases, the combination of monopole with CPW antenna in the main beam direction gives good result compared to the reference and other directions

Abstract
For new multimedia applications like dualphones, antennas have to be placed inside the housing of small mobile devices. The connectivity and high data rates required for multimedia applications and voice over IP communication can be achieved by MIMO and diversity requiring antenna arrays. Strategies are presented which help to find an optimum solution for the implementation and orientation of antennas. The influence of a user on MIMO and diversity performance is also considered. It will be shown that for efficient diversity and MIMO systems, antenna design and placement must take into account the complete system including the user.

Introduction
The development of small antennas for the integration into small mobile devices plays a significant role for the implementation of diversity and MIMO systems. Despite the limited space in between the antennas for small terminals, polarization diversity and pattern diversity can be used as design criteria for compact and efficient arrays. Space limitations require dual band antennas for the coverage of both wireless local area network (WLAN) spectra at 2.45GHz and at 5.2GHz.

Parts of the work are an outcome of the ACE network of excellence. For the placement, design, and selection of antennas not only the housing but also adjacent body parts of the user have to be considered.
In this paper antennas suitable for MIMO as well as for diversity applications are integrated in the housing of a small mobile device. Antennas with different polarizations and patterns are placed in the housing. Than the number of antennas is reduced to the desired number of antennas by suitable selection methods.

Integration of antennas into a dualphone
The integration of miniaturized antennas into the housing of a dualphone is challenging. Sensitivity to objects in the close vicinity of the antenna can be reduced by the use of balanced antennas. For integration two different types of antennas are employed. One design is a planar miniaturized inverted F antenna which has horizontal polarization. The second antenna type is a buckled inverted F antenna which has a slightly different pattern and polarization characteristic. A more detailed description of the placement of antennas and the antenna parameters will be given in the full paper.

Diversity and MIMO Performance
Different parameters can be used to qualify diversity and MIMO systems. For example, the received signal level correlation coefficient or coupling between the antennas can be used to investigate the antenna array's suitability. As an example the correlation coefficient for three antennas integrated in a small device is shown in figure 1.

Parts of the work are an outcome of the ACE network of excellence.

This work describes theoretical and numerical investigations of data stream parameters, such as capacity and spectral efficiency of multiple-input-multiple-output (MIMO) wireless communication systems taking into account fading phenomena caused by the multipath of radio propagation in urban environments. The propagation characteristics of the urban wireless channel are described using a unified stochastic approach. This approach allows to obtain the general Ricean parameter of the fading phenomena as a ratio of coherent and incoherent component of the total signal after multiple scattering reflection and diffraction by obstructions surrounding the terminal antennas, the base station (BS) and the moving subscriber (MS). The Ricean parameter is obtained for typical urban and suburban environments. Using the results of the Ricean parameter calculation, the capacity and spectral efficiency are numerically calculated for different urban areas and various situations with multipath, e.g., fading, phenomena. A simple formula to predict the channel capacity and spectral efficiency for MIMO wireless systems are derived. These formulas are compared with those obtained for single-input-single-output (SISO) wireless systems. Obtained results allow us to conclude that the proposed method can be very useful as a tool for a network planner in predicting the expected MIMO system performance for various urban and sub-urban environments.

The high-performance promise of multi-antenna (MIMO) systems can be significantly deteriorated by sub-channel correlation of the MIMO channel. Careful modeling of correlation and its impact on the MIMO system performance is, hence, of high importance [1-4]. Since the angular spectrum of the incoming multipath has ultimately related to the correlation and can also be measured for practical channels, its study attracted significant attention in recent years [2-6]. Most works are concentrated on a single-cluster channel with various angular probability density functions (PDF) or power spectra [2-4]. Considerably less attention is paid to multi-cluster channels. Extensive measurements of practical channels indicate, however, that multipath often arrives in several clusters in angular domain (in addition to its cluster in delay time domain), which was observed for both indoor and outdoor channels. Geometrical considerations can provide a straightforward explanation for this.

The effect of multipath clustering in angular domain and its impact on multi-antenna (MIMO) system performance (capacity, diversity gain) is studied in details. Fourier transform technique is a powerful tool that provides insight into the relationship between the angular spectrum (or angular probability density function(PDF)) of the incoming multipath and the antennas' signal correlation. Compact closed-form expressions are derived and validated in the case of small angular spread in each cluster.

Optimization problem is studied in terms of correlation and capacity/diversity gain. It is demonstrated that the capacity/diversity gain can be maximized by the antenna array orientation in such a way that individual cluster contributions to the correlations add in a destructive way. A new way to determine the angular PDF from measured data is suggested.

References

[1] J. Salz, J.H. Winters, "Effect of Fading Correlation on Adaptive Arrays in Digital Mobile Radio," IEEE Trans. Veh. Tech., V.43, N.4, pp. 1049-1057,Nov. 1994
[2] R. M. Buehrer, "The Impact of Angular Energy Distribution on Spatial Correlation," IEEE VTC, Vancouver, Canada, Sep. 2002, pp.1173-1177
[3] L. Schumacher et al, "From Antenna Spacing to Theoretical Capacities - Guidelines for Simulating MIMO Systems," PIMRC'02, Lisbon, Portugal, Sep. 2002, pp. 587-592.
[4] S. Loyka, G. Tsoulos, "Estimating MIMO System Performance Using the Correlation Matrix Approach," IEEE Commun. Lett., V. 6, N.1,pp. 19-21, Jan. 2002.
[5] G. Zhao, S. Loyka, "Impact of Multipath Clustering on the Performance of MIMO System", IEEE WCNC 2004, Atlanta, Georgia, USA, Mar. 2004.
[6] R.Vaughan, "Spaced directive antennas for mobile communications by the Fourier transform method," IEEE Trans. on Ant. & Prop., V. 48, N. 7, pp. 1025 - 1032, July 2000.

Cylindrical near field systems are appropriate measurement systems for huge L-band RADAR antennas, because the antenna can be measured on its azimuth positioner and the probe can be easily translated through a vertical linear slide. For large antennas, as this kind of L-band RADAR antennas, the mechanical aspects of the antenna measurement systems are important, and the errors in this mechanical part can affect to the far field radiation pattern. This paper presents an error estimation tool to analyze the most important errors in one cylindrical acquisition system and the effect of these errors in the calculation of the far field radiation pattern. This study has been realized to improve the cylindrical system and to evaluate the error budget of the Antenna Under Test. The simulator calculates the far field from an array of dipoles over a ground plane (Antenna Under Test model), and compares the ideal result with the electric field obtained using the cylindrical near to far field transformation algorithms. The source errors that have been analyzed are: - Measurement procedure errors: o Truncation errors: the analysis has been realized using different tower lengths. o Sample point spacing: the analysis has been realized using different sample points (2n points per revolution) - Positioning system errors: o Errors in the alignment of the vertical slide, including errors in x, y and z axes. - RF measurement system errors. o Random amplitude and phase errors (due to temperature variations) The obtained results are the variation in principal patterns of the far field, RMS errors in side lobes and maximum errors in side lobes. Random and deterministic source of errors are considered.