Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

(RFD0505) Advances in Handset Antenna Design

Download as pdf or txt
Download as pdf or txt
You are on page 1of 4

Microwave/Millimeter Wave Technologies

Advances in handset antenna design


This article shows how dielectrics can be used to improve the performance of electrically small antennas, as well as describes techniques to integrate the antenna, also known as an aerial, with the radio to create an antenna module.
By Simon Kingsley
aerials has The subject of dielectric advanced to not yet been developed and the radiates rather than any associated metal in the feed or the ground plane. An example is shown in Figure 1. DRAs have some excellent properties such as being small, efcient and with good resistance to proximity detuning[4]. However, they tend to have limited bandwidth, typically around 5%, whereas with modern communications systems nearer 12% to 15% is needed. The bandwidth of DRAs can be improved dramatically if the ground plane is taken away from underneath them[5] but this diminishes the advantages and is not popular with handset manufacturers who like to make use of every square millimeter of printed circuit board (PCB) available. Another way to use dielectrics in antennas is to put them into the feed mechanism of a conventional antenna such as the planar inverted F antenna (PIFA). This has the advantage of removing some inductance and introducing capacitance into the feed network. It turns out to be much easier to design and match an antenna for wide bandwidth operation with this type of feed structure. Dielectric resonator antennas are 3-D devices, so if the frequency of operation is halved, and the wavelength doubled, then the antenna becomes eight times bigger and heavier. This means they are usually unsuitable for use below 1 GHz. One way around this is to use the dielectric to excite a parasitic conducting antennathis keeps many of the advantages of dielectrics, but allows a low frequency resonance to be introduced into the system without a big increase in size.

same degree as that of other types of microwave aerial. These were the opening words of the preface to a book entitled Dielectric Aerials by D. G. Kiely[1]. True enough, but the surprising thing about this perhaps rather obvious statement is that it was made in 1952. Fifty years later, his next sentence also remains largely true, No complete theory of operation exists and methods of theoretical treatment vary in initial assumptions and in obtained results. Despite this neglect, dielectric antennas, or antennas containing dielectrics, can have some signicant advantages over conductive-only (metal) antennas. In this article, we look at how dielectrics can be used to improve the performance of electrically small antennas, and how how they can help to integrate the antenna with the radio. Modern dielectrics are excellent materials to use in radio systems. They can have relative permittivities up to about 100, be low loss and have a small temperature coefcient. Generally speaking, modern low loss dielectric materials were not developed for antennas particularly, but for the dielectric resonator lter; where requirements are stringent, so good quality materials have become readily available to the antenna designer. The effective wavelength of a radio wave is shorter in a dielectric material than in free space, so one reason for using dielectrics in antennas is that they can be made smaller than conventional metal antennas. Another reason is that the antenna is generally more resistant to proximity detuning when placed close to another object (e.g., a handset antenna near your head). The higher the dielectric constant, the better these advantages become. If the dielectric material is used in the antenna where the E-elds or RF currents are highest, then the antenna can also be more efcient than its all-metal counterpart. Dielectrics can be used in antennas in several different ways. The earliest use is the dielectrically loaded antenna; here the metal is still the radiator but the antenna is modied by the presence of the dielectric, which gives the benets mentioned above, but also tends to reduce the bandwidth. For many years this was the only type of dielectric antenna,

Figure 1. A dielectric resonator antenna designed to work at 2.4 GHz. The blue/purple material is a piece of dielectric ceramic with a relative permittivity of nearly 100.

then in 1939 along came R. D. Richtmyer. Robert Richtmyer was a mathematical physicist who worked at Stanford University and, during the war, at Los Alamos where he lead the theoretical division. He worked with Enrico Fermi in developing the Monte Carlo simulation technique and with John von Neumann on hydrodynamics. In 1939, Richtmyer published a remarkable theoretical paper[2] in which he reasoned that a suitably shaped dielectric object can act as an RF resonator and he offered a proof that such a device must radiate based on the boundary conditions at the interface between the dielectric and the surrounding air. This work was not followed up straight away, perhaps because the war intervened, but eventually dielectric resonator antennas (DRAs)[3] were developed in the 1980s. DRAs are true dielectric antennas in the sense that it is the dielectric that

Figure 2. A quadband antenna in a currently mass-produced cellular radio handset. The left hand picture shows the underneath of the antenna with the ceramic on the left hand side. The right hand picture shows the top surface and the parasitic metal component.

16

www.rfdesign.com

May 2005

Some recent multiband handset antennas have been designed in this way; an example is shown in Figure 2.

Balanced antennas

The introduction of dielectrics into antenna technology has brought many benets but has not solved one of the main problems with handset antennasthe extent to which they must be customized for every handset model produced. Antenna customization is an expensive process and can take place throughout the design phase of a phone. Virtually all modern handset antennas, including dielectric variants, are unbalanced meaning what we think of as the antenna is only half of the radiation mechanism and the ground plane is the other half. If the ground plane size is changed, or other components are moved around on the ground plane, the antenna detunes and it has to be redesigned. If antennas were to be integrated with other RF components into an RF module, the cost of customization would be even greater. Unbalanced antenna architecture has thus been a major stumbling block to the development of full radio and antenna integration within handsets. The alternative strategy is to use a balanced antenna, as it has recently been shown that a balanced handset antenna is effective for

reducing the currents on the ground plane[6]. This can be thought of as one in which the reference is not the ground plane but an inverted version of the antenna itself. A dipole is probably the simplest and best known example of a balanced antenna but like most balanced antennas it works best in free space away from other conductors. The problem with using them for mobile communications is that modern handset PCBs have a full ground plane and the antenna usually sits a tiny fraction of a wavelength above it. Dipoles just dont work in this sort of environment. A solution to this predicament has been developed recently and involves using two unbalanced antennas arranged back-to-back and fed as a balanced or complementary pair. By back-to-back we mean a pair disposed with mirror symmetry along the axis of charge acceleration, as with a dipole. Using two unbalanced antennas together has the advantage that each antenna expects to have the ground plane present, but the combination is free from ground plane effects. This phenomenon arises because, although each antenna does generate currents in the ground plane directly beneath the antenna, the current pair cancels leaving negligible currents owing on the rest of the ground plane. This effect is shown in Figure 3. One of the most popular unbalanced

Figure 3. Surface RF currents at 1900 MHz on a PCB with a radiating balanced antenna pair. The plot shows good cancellation everywhere except under the antenna. A 0-30 A/m linear scale has been used.

antennas used in a handset is the PIFA. PIFAs are electrically small, efcient and well suited to handsets. Figure 4 shows a complementary pair of PIFAs used as a dual-band CDMA antenna with an extra resonance for GPS reception at 1575 MHz. The whole antenna measures only 40 x 20 x 5 mm, which is a volume comparable with conventional single PIFAs used at the same frequencies. Reducing the dependence of the antenna pair on the ground plane has many advantages including good efciency and reduced detuning when the phone is held in a hand and placed near the head. Figure 5 shows the antenna return loss (how well the antenna matches to 50 Ohms) for the case when the PCB is held in the hand and when it is in free space. The two plots are almost identical showing how little effect the hand has on the antenna. In the low band (800 MHz or 900 MHz) the antenna needs to be unbalanced because
Circle 12 or visit freeproductinfo.net/rfd

18

www.rfdesign.com

May 2005

the wavelength is so long that the entire PCB is required to be the primary radiator. The antenna on its own is inside the ChuHarrington limit[7,8], which means it will either be an inefcient radiator or lack sufcient bandwidth. Unfortunately, this means that the requirement is for an antenna that is balanced above about 1.5 GHz but unbalanced below this. There are several ways to excite an unbalanced mode in an antenna including creating a separate patch for the low band or by forming a conductive bridge between the two PIFAs, creating a single lower-frequency resonance. A separate antenna, such as a monopole at the top of the PCB, can also be used to accommodate the low band. The CDMA antenna shown in Figure 4 is balanced at GPS and PCS frequencies, but unbalanced in the 800 MHz cellular band.

Figure 4. A complementary pair of planar inverted F antennas (PIFAs) used as a dualband CDMA + GPS antenna.

can be used that is simple to install and easily transported between different products. If no unbalanced low-band antenna is present then the module can be used anywhere on the product and at any orientation with only minimal changes to the matching components. If an unbalanced low-band antenna is present then some customization is necessary.

The future

Radio antenna modules

We mentioned earlier that unbalanced antenna architecture has been a major stumbling block to the development of greater

Once radios and antennas are locked together in the same unit they can be optimized as a pair. For example, there is no reason to stay with 50 Ohm systems if other impedances make more sense and this can be advantageous to the power amplier designers who are always searching for opportunities

We mentioned earlier that unbalanced antenna architecture has been a major stumbling block to the development of greater integration within handsets.
integration within handsets. With this obstacle now removed, integration of radios with antennas is under way. The complementary antenna pairs can easily be disposed like a bridge over the top of the RF chipset, although with the current technology the height of the antenna then increases from 5 mm to 7.5 mm. By splitting the module horizontally, a lower half can be created that encompasses all the RF semiconductors plus the balanced drive needed for the antenna, and an upper half that contains the complementary antenna pair. This arrangement simplies testing and enables different manufacturing technologies to be used for the two halves of the module. Combining radios and antennas in this way has many benets; there are further efciency gains to be made by removing the transmission line between the two and a reduction in the real estate required on the PCB. However, the most important advantage is that the OEMs and ODMs developing handsets can reduce some of the labor and costs involved in designing the RF front-ends. Less customization equals a faster time to market and this is important in todays cellular radio industry. A complete radio antenna module
Circle 14 or visit freeproductinfo.net/rfd

to increase efciency. Similarly, the radio diplexer can be adapted to give an unbalanced output in the low band and a separate balanced output for the high bands and this helps the antenna designer. The two technologies discussedthe use of dielectrics and balanced antenna designsboth lead to improvements in efciency and resistance to detuning. The next obvious step is to nd the optimum way of combining the two so as to maximize the antenna's immunity to its surroundings. This should lead to even greater efciencies when the phone is used in the talk position. If a mobile phone is held vertically, most antennas have a radiation pattern that is predominantly a horizontal doughnut shape. This means that a signicant amount of power ends up in the users head, resulting in a moderate specic absorption rate (SAR) value. In contrast, balanced antennas tend to radiate predominantly away from the PCB and the users head and low SAR performance can therefore be expected. If a push-pull power amplier is used (i.e., not only balanced but also referenced to ground) odd harmonics are suppressed; if a balanced conguration is maintained between the power amplier and

20

www.rfdesign.com

May 2005

Figure 5. Left hand plot:return loss for a typical unbalanced antenna when it is in free space and when the PCB is held in the hand. Antenna detuning is obvious. Right hand plot: a balanced antenna showing virtually no detuning when the antenna is held in the hand.

the antenna there is no possibility of their regrowth. Radio antenna modules could be developed to maximize these advantages as well, rather than concentrating solely on efciency and detuning. In the high bands, there is also scope for some diversity between modules allowing the potential for multipleinput, multiple-output (MIMO) operation in handsets and possibly some form of switched diversity within a single module.

opportunity for the two parties to work together toward RF solutions for handsets that are more efcient and much easier to install into a new product. RFD

References

Conclusion

Historically, antenna designers and the manufacturers of RF semiconductors for the mobile phone industry have had little interaction with each other, both parties sticking to the 50 Ohm unbalanced-feed format. The development of complementary antennas and radio antenna modules now gives the

1. D. G. Kiely, Dielectric Aerials, Methuen & Co, March 1952. 2. R.D. Richtmyer, Dielectric Resonators, J. Applied Physics, 10, pp. 391-398, 1939. 3. S.A. Long, M.W. McAllister and L.C. Shen, The Resonant Cylindrical Dielectric Cavity Antenna, IEEE Transactions On Antennas and Propagation, AP-31, pp. 406412, 1983. 4. R.K. Mongia and P. Bhartia, Dielectric Resonator AntennasA Review and General Design Relations for Resonant Frequency and Bandwidth, International Journal of

Microwave and Millimeter-Wave Computer-Aided Engineering, 4, (3), pp. 230-247, 1994. 5. Z. Wang, C. C. Chiau, X. Chen, B. S. Collins, S. P. Kingsley and S. Puckey, Study and Optimisation of a Broadband Dielectric Antenna, presented at the IEEE International Workshop on Antenna Technology: Small Antennas and Novel Metamaterials, Singapore, March 7-9, 2005. 6. D. Sasaki, et. al. A Planar Folded Dipole Antenna for Handset, presented at the IEEE International Workshop on Antenna Technology, Singapore, March 2005. 7. L. J. Chu, Physical Limitations of Omni-Directional Antennas, Journal of Applied Physics, Vol. 19, pp. 1163-1175, 1948. 8. R.C. Hansen, Fundamental Limitations In Antennas, Proceedings of the IEEE, Vol. 69, No. 2, pp. 170-182, 1981.

ABOUT THE AUTHOR


Simon Kingsley is chief scientist at Antenova Ltd. in Cambridge, UK. He also is the co-inventor of the high dielectric antenna (HDA) technology. He has academic qualications in physics, radio astronomy, space physics, and a lifetime of research into radar and antennas including four years at GEC Marconi Research Center developing over-the-horizon radar systems. Kingsley is a visiting professor at Shefeld University, a Fellow of the Institution of Electrical Engineers (IEE), and a Fellow of the Royal Astronomical Society. In 1997, he was an IEE Faraday lecturer. He is also a member of the IEEE.
Circle 15 or visit freeproductinfo.net/rfd

22

www.rfdesign.com

May 2005

You might also like