The wireless connections based on IEEE 802.11b/g/n (WiFi) have been the majority of market of short and medium-range wireless access. The WiMAX systems based on standard of IEEE 802.16e is an emerging technology for long range wireless access. To make the systems more cost effective, the antennas can cover multi-band or multi-standard have been required. This work is to address the demand to design antenna for covering the frequency ranges of 2.3-2.8 GHz (WLAN / WiMAX / Wi-Broad) and 3.2-3.8 GHz (WiMAX). The antenna should be low in profile in order to be embedded into an access point devices. To achieve this design, we selected the Suspended Plate Antenna¡ªusing air as it substrate¡ªas the broadband solution, which radiator is usually raised to a height of approximately 0.06 times. To enhance the bandwidth and achieve dual broadband design, we propose a folded antenna a modified version of the suspended plate antennas. The antenna structures are presented here with the two layers of radiators as shown in Figure 1. All the radiators and the ground plane (100 mm X 100 mm) are oriented horizontally in the x-y plane. The upper radiator has size of 37 mmX 37 mm. It is fed by a vertical tapered feed connected to the edge of the lower radiator with a separate of 6 mm. The lower radiator measuring 15 mm X21 mm is positioned at a height of 10 mm. The lower radiator is fed by a tapered feeding portion excited by a 50 Ohm probe through the ground plane with feedgap of 1mm. Each radiator can provide a broadband operation. The upper one is expected to operate at the lower band of 2.3-2.8GHz and the lower one the higher band of 3.2- 3.8 GHz. Therefore, this antenna forming the basic multiple broadband resonance structure is expected to operate at the dual broadband. The tapered broad band feed and the supports between radiators are aligned vertically in the x-z plane. Figure 2 compares the simulated and measured return losses. The good agreement has been observed. It is clear that the well-matched impedance bandwidth for the return losses less than -15dB covers both 2.2-2.8 GHz and 3.2-3.8 GHz bands for a dual broadband operation. Figure 3 demonstrates the measured co-pol radiation patterns in x-z (H) and y-z (E) planes at 2.3, 2.7, 3.3, and 3.6 GHz, respectively. It can seen that the antenna has a stable radiation performance in the H planes with stable beamwidth and maximum radiation direction whereas the radiation patterns become asymmetric and squint of 7-10 degree in the E planes at higher frequencies due to the asymmetrical structure. Figure 4 gives the measured gain of higher than 7.5dBi which are with less than 3 dB of variations over the frequency ranges of interest. However, the ratio of cross-pol to co-pol levels increases from -15 dB at 2.3 GHz to -8 dBi at 3.8GHz when the operating frequency increases, especially in the H-planes. Therefore, a dual broadband folded antenna which can be embedded into a wireless access point device has been proposed and implemented. The folded suspended plate antenna has provided two broadband operations with high gain and simple structure for low cost consideration.

The new generations of mobile and wireless personal devices challenge ever more the ability of antenna designers. A lot of research goes on in this area, but is often performed without sufficient systems background and connection to the market. The newest trends in terminal design have to be rapidly detected, as they will have a major influence on the requirements for antenna technology and design.

The market of communications terminals shows great flexibility, with ever growing demands for miniaturised, multi-application devices. Antenna design must be adapted to the need for increased efficiency and bandwidth. Also, new techniques and architectures can be exploited, with the introduction i.e. of new materials, diversity schemes or MIMO systems. At the same time, new applications as Multiradio or UWB systems are being developed, which will demand yet more effort from antenna engineers, to satisfy new requirements.

This paper gives an overview of the activities carried out within the ACE 2 (Antenna Centre of Excellence) Network of Excellence, regarding antenna technologies for small terminals and their applications. The aim is to carry on with the work initiated in the first ACE, where new had already been identified. In a context in which terminal technology evolves rapidly, it is essential to make a continuous survey of the latest developments in this area, in order to update and optimise the technology and design methodologies for mobile and wireless terminals.

The work is also focussed on antennas for advanced applications, such as MIMO, DVB-T/H, UWB systems, RFID or bio-medical applications, both in-body (implants) and on-body. Special attention will is given to the problem of the integration of antennas (single and multimode elements or multiple antennas) into portable devices.

When an electrically small antenna is placed on a small finite ground plane, as in mobile telephones, the ground plane resonances strongly affect the impedance and radiation characteristics of the antenna. The operation bandwidth of a small antenna can be maximized by utilizing well-radiating resonant modes of the ground plane as part of the antenna system as efficiently as possible. Characteristic mode analysis can be used to compute the resonant frequencies and resonant modes of ground planes based on the eigenvalues and eigenvectors of the coefficient matrix of the method of moment formulation. In this paper, we show how characteristic mode analysis can predict the optimal locations of antennas on ground planes. As patch-type antennas couple to the ground plane resonances through the electric field, it is natural that the strongest coupling and thus the maximum impedance bandwidth is obtained when the antenna is placed at the maximum of the electric field of a ground plane mode. This is verified by moving a very small patch-like antenna at selected locations over a rectangular 100 mm by 40 mm ground plane and computing the obtainable bandwidth at each point. The figures below show the contours of the obtainable bandwidth for the patch-type antenna and the contours of the electric field of the ground plane mode 5 mm above the ground plane. Indeed, the maximum bandwidth is obtained when the antenna is placed near the corners of the ground plane, corresponding to the maximum of the electric field of the resonant ground plane mode. The above procedure can be used to compute the antenna placement also for more complex ground plane shapes and configurations.

The growth in the number of wireless systems over the past decade has been coupled with the ever-decreasing size of the RF systems. A modern mobile terminal is not only physically small but also required to operate well across a large number of wireless standards. The evolution of an important number of communication applications which require mobility such as RFID and sensing systems enforces the use of devices with an energy efficiency that allows reducing costs and size. All these new applications imply new requirements on the device that connects the terminal to free-space; namely the antenna. No longer can large, single-band, externally mounted antennas be considered. The future, therefore, lies in small, internally mounted, antennas that are able to work well across a large bandwidth.

Neither the radiation mechanism nor the design of small antennas in modern applications such as Ultra Wide Band, MIMO systems (Figure 1) and sensor networks, can be understood with traditional radiation theories for small antennas. New antenna requirements entail new modeling tools and new design technologies.

This paper presents the activities carried out in the ACE Network of Excellence regarding small antenna technologies. The aim of our work is to optimize the existing design methodologies, and make use of the possibilities offered by new methodologies applied to small antennas.

Current theories are revised in order to get a deeper understanding of the behaviour of small antennas, when trying to investigate the fundamental limits with respect to efficiency and quality factor Q (Figure 2). Such analysis derives in the definition of a common set of basic parameters for small antenna specification and the identification of the existing design methodologies for small antennas. Current design techniques focused on the miniaturization of the antennas are detected and described. In addition, advanced evaluation and simulation tools are considered and validated in the analysis.

Since the compactness of the RF transmitter and receiver is increasing, the effects of integrating the antennas with the active element are studied. The newest trends in compact small antenna design have to be rapidly detected.

Figure 1: Handset antenna with 2 PIFA elements

Figure 2: Fundamental Limit and Realistic Limit of quality factor Q

Antenna designers' primary goals are to meet or exceed certain design specifications in the most efficient manner possible. Goals might include specifications for center frequency, bandwidth, radiation pattern, multi-band operation, and minimum fabrication tolerance sensitivity, among others. However, with real world product design, limitations are often imposed on the size, form factor, materials, and topology, which force the antenna designer to be creative with the design implementation. Required creativity can imply greater design complexity and potentially more difficult design optimization. Several examples of such implementations include folded, stacked, multilayer, and meandered antennas, to name a few.

The challenge of such non-standard implementations is in establishing intuitive relationships between the geometric variations and their contributions to the desired output parameters. An effective method for establishing these relationships and the empirical effects on the output parameters, is to sweep the values of design variables in simulation and to observe the related trends in antenna performance. Computer models are often used to set up variations of a design and to crunch numbers for each design until the results are obtained.

In this paper, an advanced technique for tuning and optimization of a planar antenna for GSM applications with reduced cycle time is demonstrated. The technique includes the use of parametric studies of antenna performance using distributed analysis. A setup process which involves full design parameterization is described. The method of distributing the analysis across multiple computers is also described and the time savings quantified. Finally the results of the distributed parametric study is used as a design surface for an intelligent optimization which is also carried out with distributed computing. The results of this method are a greatly reduced design time, and a more thorough understanding of the effects of variable changes on antenna performance.

1.Introduction

Recently, the MIMO technologies for the mobile terminal are investigated to achieve high-speed data communication. A smaller and more efficient antenna system is preferable for the terminal. Though built-in type antennas are good for convenience using, an influence of the human body becomes larger. Up to now, we have proposed the folded loop antenna [1] (FLA) with the balanced feed and the parallel plane antenna [2] (PPA) with unbalanced feed. In this paper, 3 types of 2x2 MIMO antennas are investigated to show the effect of the array antenna for the terminal under multi-path wave environment. The channel capacity is employed to estimate antenna performance.

2.Evaluation method and model

In this paper, the integrated simulation technique is employed, which combines the FDTD [3] calculation for Directivity and Ray Launching method [4]. The antenna performance under an actual propagation environment as receive signal level, eigenvalue and channel capacity can be analyzed from this method. A frequency is 2.6GHz, spread angle of Ray is 0.6 deg., reflection is under 5 times and diffraction is 1 times, respectively. In order to estimate in a multi-path environment, an actual indoor room shown in Fig.1 is simulated. The transmission point (TX) is fixed and the receiving point (RX) is changed along Y axis shown in Fig.1. The 2x2 MIMO antenna for the terminal is placed in RX.
Fig.2 shows 3 types of array antennas under estimate; FLA-A as a space diversity model of which the balance feed FLA elements feed in balance are placed vertically, FLA-C as a polarization diversity model of which the L shape FLA are placed diagonally on a case, PPA as an unbalanced feed model made by two parallel elements and case.
The received characteristic of MIMO antenna under the multi-path wave environment is compared between the different antenna structures.

3.Conclusions

The channel capacities of MIMO and SISO are shown in Figs.3 and 4. The channel capacity is evaluated when cumulative probability value of movement section 6m is 50%.
The correlation coefficient between elements in each composition was 0.4 or less, and the change in the average receiving power was a principal cause in the change of the channel capacity. In comparison between the balance and unbalance system antenna, though PPA has the best channel capacity in a vertical position (Fig.3), the FLA-C becomes best in an inclination position as 45 deg. shown in Fig.4. This is because the vertical polarization wave which is dominant in this environment (XPR=8dB) can be received well when FLA-C is inclined. The polarization and directivity can�ft control well by the PPA because it is excited using the whole chassis. On the other hand, FLA can cause an arbitrary received pattern by changing the distance and arrangement on the chassis.
As a result of all models, MIMO causes double channel capacity compared to SISO. Thus, an increase in the transmission capacity by MIMO can be expected even in small terminal.

4.Reference

[1] S.HAYASHIDA:"Wideband folded loop antenna for handsets",IEEE AP-S Proc., Jun. 2002.
[2] N.NISHIKIDO:"An Antenna System with Two Parallel Elements for a Handy Phone",IEICE Society Conf., Sep.2003.
[3] http://www.cst.com/
[4] M.Moya:"Analysis and Simulation of Smart Antennas on Ray Launching Modeling Techniques",IEEE SAMSP Workshop.2000.

As either the size of an antenna or the operating frequency decreases, there is a well known lower bound on the achievable antenna quality factor (Q), often referred to as the Chu limit. This lower bound on Q is given by



where ξ is the antenna's radiation efficiency, k is the free space wave number, and a is the radius of a sphere circumscribing the maximum dimension of the antenna. Since there is an inverse relationship between Q and matched VSWR bandwidth, a lower bound on Q implies that there must be a corresponding upper bound on bandwidth. Recent work has demonstrated that electrically small antennas can be designed such that their Q's can approach but not exceed the lower bound on Q or the Chu limit. It has been recently suggested that surrounding an electrically small antenna with a negative index Metamaterial, where either or both of the material's relative permittivity or permeability are negative, can lead to an antenna Q that does exceed the Chu limit.

In this paper, we consider the performance of electrically small dipole and loop antennas surrounded by both positive and negative index Metamaterials. We consider examples as a function of both the loss and dispersion characteristics of the Metamaterial. We present results with Metamaterials that are both lossless and lossy as well as both dispersive and non-dispersive. Results are presented using Lorentz and Drude dispersion models. We show that the inclusion of loss and dispersion significantly affects the radiation efficiency and Q of the antenna. When loss and dispersion are considered, we demonstrate that the antennas exhibit marginal radiation efficiencies and that the antenna Q does not exceed the Chu limit.

Introduction

In the last decade, the mobile industry experienced a dramatic development. The first stage was from analog standard to digital standard development, for example, analog standard such as AMPS (Advanced Mobile Phone System), NMT (Nordic Mobile Telephone) and ETACS; digital standard such as GSM, D-AMPS CDMA and WCDMA. The second stage was from single band to multi-band due to the strong capacity requirements. The increase of number of wireless standard was also coupled with the integration and down size of the mobile terminals. A significant progress of the multi-band antenna and the integrated antenna technology were developed in the last decade. Those technologies support to make the phone to be small in size, mechanical robust, lower cost and higher efficiency. In this paper, I will review some important antenna innovations in the mobile industry, which have been widely used in the practical applications.

Conclusions

In the last decade, the innovation of mobile terminal antenna has experienced a great progress. A lot of important patents were filed in this area; Ericsson has owned more than 200 patents in this area and show a world leading position. The innovation work results that the mobile terminal has multi-band, multi-system, mechanical robust, integration and miniaturizing features. Many hundred millions customers gain the benefit from those innovations.

Reduction of the power absorbed by the body or head of a person, in the vicinity of a radiating source, has been a topic of major interest for more than ten years. In the case of mobile terminals, this interest has been mainly motivated by two objectives: first, to design handsets which are preferably largely compliant with exposure limits; second, to increase the radiation efficiency of the phones in presence of the user. The aim of this paper is to analyze the limitations in the minimization of the Total Dissipated Power (TDP) in a mobile phone user, due to the constraints of minimal free-space bandwidth and efficiency of the terminal. In fact, [1] and [2] have shown that the minimization of the TDP in a lossy scatterer, exposed to the field of a radiating source contained in a given coordinate volume V, requires to excite some high order modes outside of V. However, it is well known, e.g. [3], [4], that an antenna system with minimal radiation quality factor Q should only excite the lowest order modes. A limitation in simultaneously minimizing the TDP in the user of a mobile phone and the free-space radiation Q consequently appears.

In order illustrate this idea, the equivalent junction model [1], [2] is applied to the case of a 2-D cylindrical homogeneous phantom or body model, exposed to TM cylindrical sources, contained in a fixed cylinder. Among others this model allows to characterize the sources minimizing the TDP in the phantom. For instance, Fig. 1a shows the E-field magnitude of such sources, operating at f=900 MHz and exciting at most modes of order L=1-3, out of a 2 cm radius cylinder, centered 2 cm away from the surface of a 10 cm radius phantom.

Fig. 1: (a) E-field magnitude of 2-D TM sources of maximal order L, minimizing the TDP in a 10 cm radius homogeneous cylinder (relative permittivity: 41.5; conductivity: 0.97 S/m). f=900 MHz, and P_rad=1W/m (free-space radiated power by unit length). (b) Lower bounds of the radiation Q and TDP P_d, normalized to the same quantities for L=0, as functions of the maximal mode order L.

Lower bounds of the Q factors of these sources can be obtained by considering only the energy stored out of V [4]. As shown on Fig. 1b, whereas the TDP decreases rapidly with the order of the source, the lower bound of Q increases nearly as rapidly. On the other hand, this paper also shows how constraints on the minimal free-space bandwidth and efficiency impact on the optimal sources with respect to TDP minimization. The presented results suggest that handset antenna design methods, focused on the optimization of free-space quantities, significantly reduce the possibility to optimize the radio-performances in presence of the user.

[1] B. Derat, J.-Ch. Bolomey, "Analytical lower and upper bounds of power absorption in near-field regions deduced from a modal-based equivalent junction model," Progr. In Electrom. Res. (PIER), vol. 58, pp. 21-49, 2006.
[2] B. Derat, On the improvement of antenna design methods for 2G and 3G mobile phones, Ph. D. thesis, March 2006.
[3] L. J. Chu, "Physical limitations of omnidirectional antennas," J. Appl. Physics, vol. 19, pp. 1163-1175, Dec. 1948.
[4] R. E. Collin, S. Rotschild, "Evaluation of Antenna Q," IEEE Trans. Antennas Propagat., pp. 23-27, Jan. 1964.

Experiences and results from the benchmarking of small terminal antenna measurement facilities that has been done in the framework of ACE - Antenna Centre of Excellence will be presented. The emphasis was put on measurements in the most popular communication frequency bands. Several test cases for passive antennas as well as active devices were defined and collected in a test set that was sent to ten different test facilities around Europe in a round robin test.

The measured parameters include radiation efficiency, total radiated power, diversity gain and receiver sensitivity. Test cases also include different locations of the test devices relative to a lossy cylinder that has dielectric characteristics resembling that of a human head. In order to guarantee comparable results only one test kit was sent around. In addition control measurements to check that nothing had happened to the test devices were performed at three occasions, at the start of the benchmarking, in the middle and at the end.

Various measurement methods were used by the participating organisations, e.g. 3D radiation pattern integration in fully anechoic chambers, spherical near field methods, random positioner system and reverberation chambers.

Results from measurements of radiation efficiency for the passive antennas included in the benchmarking show that the deviation from mean value is less than 1.5 dB except for the 5.2 GHz slot antenna. For this case the maximum deviation from mean value is 2.67 dB but if one particular participant is disregarded the maximum deviation from mean value is 1.5 dB for all cases. Only two participants measured diversity gain.

The maximum difference in measured total radiated power from GSM phones was 3.18 dB. No significant difference between the GSM bands, 900, 1800 and 1900 MHz, could be observed. Only two participants measured total isotropic sensitivity.