Showing papers by "Per-Simon Kildal published in 2008"

Further investigations of fundamental directivity limitations of small antennas with and without ground planes

Abstract: In this paper, we have presented a heuristic formula for the maximum achievable directivity of single-port antennas of any size. The formula is well rooted in electromagnetic theory in the low and high frequency ends. It has been tested by comparison with reported theoretical directivities of some practically realized antennas in the intermediate frequency range where the diameter of the smallest enclosing sphere is between 0.1lambda and 4lambda, and it seems to work as a practical limitation.

Three Fast Ways of Measuring Receiver Sensitivity in a Reverberation Chamber

Abstract: The traditional way of evaluating receiver sensitivity is to measure the Total Isotropic Sensitivity (TIS) in an anechoic environment. Very few wireless terminals with small antennas are used in an anechoic environment. This paper will describe three alternative methods to measure the receiver sensitivity in a reverberation chamber, each of which can be done in less than 10 minutes which is much faster than measurements in anechoic chamber. The fastest method is the average fading sensitivity AFS that in contrast to TIS also be used to measure receiver sensitivity of multi antenna terminals for antenna systems with diversity and MIMO capability. The paper includes the first published TIS and AFS results for WCDMA terminals obtained in a reverberation chamber. The agreement to TIS measurements in anechoic chambers is shown to be good.

Relations between coherence bandwidth and average mode bandwidth in reverberation chamber for wireless device measurement

Abstract: The reverberation chamber is basically a metal cavity with many excited modes which are stirred to create a statistical isotropic field environment with Rayleigh distributed magnitudes of the field [1]. The reverberation chamber used here is 1.8 m х 1.4 m х 1.2 m in size. The chamber makes use of frequency stirring, platform stirring, and polarization stirring to improve measurement accuracy [1], [2]. It has been used to measure antenna radiation efficiency and diversity gains [2]. For these measurements the intrinsic average mode bandwidth of the chamber is of interest as it determines the averaged power transmission level of the chamber [3]. It may also be sued to control the number of excited modes in the chamber, and thereby also the number of independent samples in order to generate enough independent samples to improve the measurement accuracy [4], [5]. Reverberation chamber can also be used to measure total radiated power and total isotropic sensitivity of active mobile phones [6], [7]. For such measurements the delay spread and coherence bandwidth are of relevance. Coherence bandwidth is defined as the frequency range over which the channel is correlated. When a signal at a certain frequency is transmitted through a reverberation chamber, several cavity modes will be excited. Each of these resonating modes can be characterized by its

Characterization of multi-port Eleven antenna for use in MIMO system

Abstract: The purpose of the present paper is to investigate the MIMO performance of the L-band multi-port Eleven antenna. Both a single- and dual-polarized lab model are investigated, both having 4 ports per polarization. The 4 ports per polarization are normally combined in a 180 deg hybrid to produce sum and difference patterns, but we investigate them here as individual ports in an antenna for a MIMO system. The embedded element efficiency, diversity gain and maximum available capacity of both models have been measured in a Bluetest reverberation chamber. The measured results show that the Eleven antenna is a flexible antenna that can be used with around 11 dB gain and arbitrary single or dual polarization in a Line-Of-Sight system and with 4 or 8 independent ports in a MIMO system, and it can easily be switched between these two modes by appropriate combination of the antenna ports.

Blockage reduction of rhombic cylinders using meta-surfaces

Abstract: Blockage caused by struts with rhombic cross sectional shape and relatively large width can be reduced by coating them with artificial surfaces. Two different types of coating solutions which reduce blockage simultaneously for TE and TM cases were proposed. The first solution is based on hard surfaces made with narrow metallic strips and it has been proven how parameters such as the strip period and the rhombus length are critical for TE performance. The second solution (using a mushroom-type meta surface)has very good performance for TE case around 8 GHz with a very low blockage width over a significant bandwidth. This represents a new blockage reduction technique produced by a TM surface wave (relative to the surface normal) propagating around the strut. The performance for TM case is also good around 8 GHz, but within a very narrow band.

Comparison of blockage widths of ideally hard cylinders of different cross-sectional shapes

Abstract: We have computed the equivalent blockage width of ideally hard cylinders of different cross sections. We have shown that the oblong cross section is best when the physical width is larger than about 0.2square, whereas for narrower cross-sections the transverse thin rectangular cross-section is best. The latter can be strengthened by a star-shaped cross section, without significant change in the blockage width. The latter was found as a result of numerical optimization. The optimized oblong shape is explained by a smooth transition of the waves past the cylinder, which is facilitated by the GO characteristics of the hard surface [8]. The thin rectangular cross section is explained as a quasi-static solution. The cross-section is so narrow in terms of wavelengths that transverse currents cannot be induced. The star-shaped cross-section works like the transverse rectangular cross-section, because the orthogonal rectangular part making up the two other arms of the star are invisible to the wave because of their small thickness. It must be emphasized that we have treated ideally hard cylinders. This makes the results valid for metal cylinders and TE-case. The performance for TM-case depends on the realization of the PMC surface, but the results can be seen as typical performance at the center frequency. The bandwidth will normally be small, but can be up to 20-30% when the artificial magnetic conductor is realized by dielectric coatings.

Study on a parabolic reflector fed by L-band multi-port eleven antenna

Abstract: The paper studies the properties of the far field of a primary-fed parabolic reflector fed by a four-port L-band Eleven antenna. The sum beam as well as difference beams in two orthogonal planes after the reflector, which are necessary for the mono-pulse tracking, have been numerically evaluated. In addition, the relationship between the spacing between two parallel folded dipoles and the matching as well as radiation properties has been investigated.

Parametric characterization of parabolic reflector antennas fed by focal plane arrays

Abstract: The study has resulted in the following conclusions. The initial benefit of including Airy sidelobes in the FPA sampling is large, i.e. the efficiency climbs steadily as the number of sidelobes increases from zero. Hence it is inadequate to just sample the main lobe or even up to only one sidelobe. For optimal performance, it is seen from Fig. 2 that around 2 to 3 sidelobes are needed - often closer to three. The Airy excitation gives lower efficiency for deeper reflectors, as it represents the theoretically correct efficiency only for the shallow case. Therefore, there is a need for studying alternative excitation functions, or optimizing it. The coupling efficiency decreases strongly when the element spacing is smaller than 0.2 wavelengths. Therefore, very small element spacings should be avoided.

Degradation of MIMO-capacity due to correlation and efficiency of practical antennas

Abstract: Handheld terminals such as mobile phones incorporate more and more multimedia functionality. This is a challenge for the terminal designers, since higher download speed is demanded. Therefore, a lot of research effort is currently spent on approaches to enhance download speed. One approach is to use multiple antennas on both the transmitting and receiving sides. If there are M antenna ports on the transmit side and N on the receive side, there will be MxN signal channels through the multipath environment, thereby allowing for higher data speed than if only one single channel is used. This is commonly referred as a MIMO-system. Multipath environments can be emulated in a reverberation chamber and hence the performance of a MIMO system can be measured there [1]. It is not obvious how to design good antennas for MIMO systems, since classical array theory can not be used. The maximum available capacity of the MIMO antenna system in an isotropic Rayleigh environment can in principle be calculated and used as a cost function in the optimization process of the antenna. However, most antenna engineers do not have any feeling for how to optimize the MIMO capacity and can not give a good starting point for the optimization process. Hence, it will be useful to find how classical antenna performance parameters like radiation efficiency and mutual coupling affect the MIMO-capacity, as shown initially in [2] by characterization both in reverberation chamber and by simulation. The correlation between the antenna ports in the multipath environment also affect the performance but was shown to be a minor effect compared to the radiation efficiency. The reverberation chamber has also been used to characterize the MIMO antennas in [3] and [4]. The purpose of the present paper is to compare the performance of three different practical linear MIMO arrays, each with 4 radiating elements, but having different element spacing. Previously, a circular array of monopoles was studied for different element spacing in [2].