A Telemetry, Tracking, and Command Antennas System for Small-Satellite Applications

Abstract

Circularly polarized (CP) antennas are used in space applications for telemetry tracking and command (TT&C). In this paper, we present an antenna for a TT&C system that can employ only four antennas to achieve a wide quasi-isotropic coverage. The proposed single-feed circular polarized annular patch antenna is low profile and compact. It exhibits a wide impedance bandwidth of 19.7%, wide-angle circular polarization with an on-axis maximum gain higher than 6 dBi and an axial ratio below 4 dB in the frequency band 2.0–2.3 GHz. The developed low-cost antenna can be used for small-satellite data downlink where only two right hand circular polarized (RHCP) antennas and two left hand circular polarized (LHCP) antennas would be required since each antenna has an adequate CP gain on a wide angle. A prototype antenna has been fabricated and measured. The experimental validation shows a good agreement of the measured impedance bandwidth, axial ratio, and far-field pattern with the ones predicted by numerical full-wave study.

1. Introduction

The number of space missions involving small satellites in their different sizes is exponentially increasing. The reasons for this growth are mainly related to the low cost and the flexibility that a small-satellite-oriented approach can offer. Indeed, several kinds of missions, such as communication [1], remote sensing [2] and Earth observation [3], are now possible despite the small dimension of the platform. In general, multiple nanosatellites, designed for different applications, can be placed in the same orbit or different orbits by the same launcher. The TT&C (Telemetry Tracking and Command) link of a satellite is of prime importance, right from the time of its launch to the end of the mission. These subsystems provide an uplink for command signal, a downlink for monitoring signal, and various health parameters through telemetry and tracking information for monitoring its position in orbit. Right after the launch, a satellite is affected by unstable attitude conditions, therefore it must be stabilized before the payload became operational. TT&C antennas are a key component in this phase to provide the attitude signal and the position to the ground station and receive the data needed for the de-tumbling operations. The S-band in the frequency interval 2.0–2.3 GHz is used for TT&C [4]. The specification for a TT&C antenna requires full spherical coverage, since the orientation of the satellite is uncertain, and circular polarization due to the Faraday effect in the atmosphere.

It is well known that circular polarization (CP) can be obtained from linear orthogonal polarizations with an appropriate feed network to radiate them in-phase quadrature [5,6,7,8,9]. The above method has many advantages such as wide bandwidth and dual sense of polarization when a quadrature hybrid is used in the feed network [5]. However, the feeding structure complicates antenna design and fabrication. In [7] an S-band TT&C antenna employing bow-tie dipoles excited by an appropriate feeding network has been presented. To avoid complicated feeding networks, CP can also be generated with perturbation technologies [10,11,12,13,14]. This approach results in compact single-feed antennas, but such design generally has very limited bandwidth. The reduced bandwidth, typically a few percent, does not represent a limitation for relatively narrowband short-range communication systems [12,15]. A CP capacitively coupled single-feed high-gain patch antenna for wireless power transfer has been presented in [16]. However, the 3-dB axial ratio (AR) bandwidth is only 2%.

In this paper, we present a wide-bandwidth TT&C antenna system employing multiple S-band annular ring antenna in a stacked configuration [17] with over $19.7$% of bandwidth achieved thanks to a double resonance. Circular polarization is obtained in the proposed single fed antenna by exciting two orthogonal quasi-symmetrical polarizations. A parametric study is presented and a prototype, adopting space-qualified materials, is realized. Measured scattering parameters, AR, and antenna gain agree well with simulation results.

2. Antennas System

2.1. Configuration

Two RHCP antennas and two LHCP antennas are used for the proposed TT&C system. These antennas have been fabricated and have been installed at the edge of the satellite’s payload (a 70 $\mathrm{c}$$\mathrm{m}$ diameter X-band dish antenna) to avoid shadowing from the payload itself. The position of the TT&C antennas is shown in Figure 1.

Each antenna covers a hemispheric region; therefore, the antennas altogether provide a quasi-isotropic radiation pattern. In this way, during tumbling, all antennas are used to provide TT&C communications with the ground station.

The goal of the study was to obtain a wideband CP antenna with an axial ratio below 4 dB from 2.0 GHz to 2.3 GHz. Moreover, for the CP gain an 80-deg half-power beamwidth (HPBW) is required i.e., provided that the maximum CP gain is reached in boresight direction (at $\theta =0$) we require: $G(\theta )/G_{max}>1/2$ for $\theta \in [-40,40]$ deg.

To achieve wide 4 dB axial ratio bandwidth multiple resonances are necessary; moreover, to easily control CP over a wide frequency bandwidth, the possibility to independently tune the amplitude and phase of the two linearly polarized components of the radiated electromagnetic field at different frequencies is desirable.

2.2. Antenna Element

The layout of the proposed antenna element is shown in Figure 2.

In this design, two annular patch antennas are employed in a stacked design. In the bottom layer, the first annular patch antenna and the ground plane are printed on a Rogers TMM4 with a dielectric constant of ${ϵ}_{r1}=4.50$ (simulated ${ϵ}_{r1}=4.661$) and with a loss tangent of $tan\delta =0.002$. This metallization is represented in Figure 2 in dashed line with inner and outer radius $R{1}_{\mathrm{in}}$ and $R{1}_{\mathrm{out}}$ respectively.

Due to the presence of triangular extensions (with side $S1$) the TM${}_{11}$ mode, of the unperturbed annular patch, splits into two near degenerated orthogonal modes: mode 1 and mode 2 as shown in Figure 3. The equivalent resonant length along x of mode 1 is larger than the equivalent length along y of mode 2. The feed point is at coordinate $(f_x,f_y)=(f_{r}cos\varphi ,f_{r}sin\varphi )$ and, being $\varphi =45$ deg, it well couples both modes.

The two modes naturally oscillate in-phase quadrature for an appropriate choice of the geometric parameters in, particular, we will show that the phase difference in the radiated field can be controlled modifying the equivalent length along x and y acting for example on $S1$. A second coupled patch is printed on a Rogers RO4003 (Rogers Corporation, Chandler, AZ, USA) with a dielectric constant of ${ϵ}_r=3.607$ and with a loss tangent of $tan\delta =0.002$ and suspended above the main patch. These materials have been chosen to comply with the mechanical and thermal requirement. The inner and outer radii of the top annular patch are $R{2}_{\mathrm{in}}$ and $R{2}_{\mathrm{out}}$ respectively. An air gap is obtained between the two layers by using four cylindrical spacers made of PEEK and four screws made of Delrin POM. In this case, the cavity under the patch is filled with both Rogers RO4003 and air. The top patch geometric dimensions are sensibly larger than bottom ones, $R{1}_{\mathrm{in}}$ and $R{1}_{\mathrm{out}}$, since the equivalent dielectric constant filling the patch cavity is a value in between 1 (air) and 3.27 (Rogers RO4003). Also for the top patch cavity, the equivalent resonant length along x is larger than the one along y since a pair of triangular shaped aperture (with side $S2$), are etched (subtracted) along the y direction and also in this case for an appropriate choice of the design parameters CP can be optimized. By coupling the patch in the bottom layer with the one in the top layer a wider operational bandwidth is achieved. Moreover, thanks to the multiple resonances and the several tuning geometric parameters (top and bottom inner and outer radius as well as the extension/slit) CP can be optimized at the same time for slightly different frequencies, in the band of interest, for wideband operation.

Table 1. Antenna geometric parameters (mm). 1 refers to bottom patch, 2 refers to the top patch.

R2out	R2in	R1out	R1in	S1	S2	fr	fx	fy	h2	h1	ag
23.837	8.351	16.061	5.261	12.204	5.76	14.6	10.323	10.323	1.524	3.175	11.17

shows the dimensions of the proposed antenna. If the antenna is mirrored i.e., if the feed point is placed along the opposite diagonal at $(f_x,f_y)=(f_{r}cos\varphi ,-f_{r}sin\varphi )$ the polarization sense can be switched from LHCP to RHCP.

Values reported in Table 1 are the final set of parameters used for the manufactured antenna. The above geometric values have been fixed after a parametric study carried out on several antenna parameters by standard finite-element-method full-wave simulations considering the frequency band of interest as well as the angular mask for the radiated field. Circular polarization is obtained by controlling the cavity parameters in order to have the radiated field with $\theta$ and $\varphi$ component of equal amplitude and in-phase quadrature.

3. Parametric Study

The complex number:

$$p=\mathrm{j}\frac{E_{\theta }}{E_{\varphi }}=\mathrm{j}\frac{a_{\theta }}{a_{\varphi }}{\mathrm{e}}^{\mathrm{j}({\psi }_{\theta }-{\psi }_{\varphi })}$$

(1)

where

$$E_{\theta }=a_{\theta }{\mathrm{e}}^{\mathrm{j}{\psi }_{\theta }}\mathrm{and}{E}_{\varphi }=a_{\varphi }{\mathrm{e}}^{\mathrm{j}{\psi }_{\varphi }}.$$

(2)

is referred as the linear polarization ratio [18] and can be used to evaluate the polarization state. In particular we have CP for $p=\pm 1$ (i.e., for $a_{\theta }=a_{\varphi }$ and ${\psi }_{\theta }-{\psi }_{\varphi }=\pm \pi /2$ rad $=\pm 90$ deg). In the remaining of the paper we adopt the following shorthand notation $|E_{\theta }|=a_{\theta }$ for the modulus of the electric field $\theta$-component; $|E_{\varphi }|=a_{\varphi }$ for the modulus of the electric field $\varphi$-component; $\angle E_{\theta \phantom{\varphi }}={\psi }_{\theta }$ for the phase of the electric field $\theta$-component and $\angle E_{\varphi }={\psi }_{\varphi }$ for the phase of the electric field $\varphi$-component. When the ratio $\left|E_{\theta }\right|/\left|E_{\varphi }\right|$ is equal to one and the phase the phase difference $\angle E_{\theta }\phantom{\varphi }-\angle E_{\varphi }$ is 90 deg we are in presence of circular polarization. These two quantities are more directly linked to the geometric parameters used in the optimization and will be used below to discuss some results of the parametric study where appropriate.

3.1. Triangular Extensions and Slits

The triangular extensions $S1$ of the bottom patch and the triangular slits $S2$ of the top patch, by loading only one of the two modes of the corresponding patch cavity, modify only one of the resonant length of the two degenerate modes with respect to the other. This can be used to effectively control the phase difference of the radiated field that is related to the unequal resonant lengths.

As shown in Figure 4 this allows fine control of phase difference of the linearly polarized components of the radiated field; on the contrary the ratio $\left|E_{\theta }\right|/\left|E_{\varphi }\right|$ in less affected. From Figure 4e it is apparent that the effect of $S1$ on the phase difference is slightly larger at lower frequencies in relation to the fact that the bottom patch has a lower resonant frequency.

3.2. Air Gap

Many parameters have been considered and tuned during the optimization process. Among these, the air gap can be easily tuned also in the last stages of the fabrication (see Figure 5). The air gap directly modifies the upper cavity height. This has an important loading effect and an influence on the effective resonant lengths along x and y of the top patch cavity (and to a lesser extent also on the effective resonant lengths of the bottom one).

From Figure 6 and Figure 7, where we show the co- and cross-polar gain at the central frequency of 2.15 GHz for different spacing between the top and bottom substrates, it is apparent that the air gap has a great influence on the cross-polar level, this is due to the fact that the spacing affects the ratio $|E_{\theta }|/|E_{\varphi }|$ between the two linear polarization.

From the same figures, it is less evident that the air gap has also a secondary effect on the antenna co-polar gain that according to the theory, has an increase of $\approx 0.4$ dB when the gap is increased from 9 mm to 13 mm. The impact of the air gap on the quality of CP is more evident from axial ratio plotted for different value of the air gap as shown in Figure 8. From the parametric axial ratio plots, it is apparent that $ag$ has a great impact on the second resonance at higher a frequency.

The influence of the air gap on the CP can be better understood if we observe the ratio between the linearly polarized component $E_{\theta }$ and $E_{\varphi }$ of the radiated field. In Figure 9 we show the $E_{\theta }$ to $E_{\varphi }$ ratio for the cut $\varphi =0$ deg. By varying the air gap, the ratio between $|E_{\theta }|$ and $|E_{\varphi }|$ far-field components can be controlled and tuned to the target unit value.

In Figure 10 we show the phase difference between the two linear polarization $E_{\theta }$ and $E_{\varphi }$, for the cut plane $\varphi =0$ deg. The on-axis ($\theta =0$) phase difference, $\angle E_{\theta \phantom{\varphi }}-\angle E_{\varphi }$, is differently affected at various frequencies as the air gap is increased. The required phase difference for CP is 90 deg. Figure 10 clearly shows that the phase difference $\angle E_{\theta }\phantom{\varphi }-\angle E_{\varphi }$ has a strong dependence on the air gap. This effect is larger at higher frequencies.

In

Table 2. Linearly polarized component in the direction $\theta =0$.

		ag = 9 (mm)	ag = 11 (mm)	ag = 13 (mm)
$f=2.0$ (GHz)	${\left(|E_x|/|E_y|\right)}_{\theta =0}$	1.008	1.209	1.396
	${\left(\angle E_{x\phantom{y}}-\angle E_y\right)}_{\theta =0}$	96.96	92.38	86.02
$f=2.1$ (GHz)	${\left(|E_x|/|E_y|\right)}_{\theta =0}$	1.506	1.221	0.963
	${\left(\angle E_{x\phantom{y}}-\angle E_y\right)}_{\theta =0}$	72.78	75.99	78.46
$f=2.2$ (GHz)	${\left(|E_x|/|E_y|\right)}_{\theta =0}$	1.351	1.141	0.968
	${\left(\angle E_{x\phantom{y}}-\angle E_y\right)}_{\theta =0}$	58.56	71.94	89.97
$f=2.3$ (GHz)	${\left(|E_x|/|E_y|\right)}_{\theta =0}$	1.126	1.259	1.563
	${\left(\angle E_{x\phantom{y}}-\angle E_y\right)}_{\theta =0}$	73.49	97.87	119.39

we summarize the results of the parametric study on the air gap for the particular direction $\theta =0$.

We discussed Figure 9 and Figure 10 for $\theta =0$ and in Table 2 we summarized only the case $\theta =0$. For the application at hand it is required a good CP up to $\theta =40$ deg. The same figures can be used to control CP at several $\theta$ angles. With reference to Figure 9 we can observe that $\left|E_{\theta }\right|/\left|E_{\varphi }\right|$ has been optimized to be close to 1 on “average” over a large $\theta$ interval. As it is well known the $E_{\theta }$ components drops at grating angles as it is apparent from the same two plots. This has been compensated here by choosing $\left|E_{\theta }\right|/\left|E_{\varphi }\right|>1$ at $\theta =0$ but it can be corrected in several other manners, a well-known method is the use of a so-called “monopole fence”, positioned around the patch element [7,19].

The final set of parameters chosen for the antenna realization have been selected considering also other constraint such as the influence of the air gap on the impedance bandwidth, this study is shown in Figure 11.

4. Radome Geometry

To guarantee protection from the external environment, a radome (Figure 2 in dashed blue) made of Delrin POM with ${ϵ}_r=3.8$ and $tan\delta =0.024$ has been designed and tested. Since the presence of the dielectric radome has an impact on the antenna performance the antenna optimization and fine-tuning have been carried out considering the presence of the radome. The geometry of the proposed antenna radome is shown in Figure 12.

The thickness of the square box is 1 $\mathrm{m}$$\mathrm{m}$ and the total height is 23 $\mathrm{m}$$\mathrm{m}$. On the circular base, with thickness of $1.5$ $\mathrm{m}$$\mathrm{m}$, eight circular holes are drilled to place the antenna on the platform.

The parametric study of the previous section, the optimization and the antenna characterization has been carried out considering the presence of the radome. The impact of the radome on the radiation pattern at $f=2.1 \mathrm{G}\mathrm{Hz}$ can be observed in Figure 13.

In presence of the radome the two orthogonal modes/polarization $E_{\theta }$ and $E_{\varphi }$ (that are tuned to oscillate in-phase quadrature) have almost the same amplitude $E_{\theta }/E_{\varphi }=13.93/11.70=1.19$, due to the aforementioned optimization process. In absence of the radome, this ratio is slightly deteriorated $E_{\theta }/E_{\varphi }=15.50/9.97=1.55$ as well as the axial ratio that reaches the value of $4.45$ dB at $2.15$ $\mathrm{G}$$\mathrm{Hz}$.

5. Results

In the proposed work a total of four antennas (two RHCP and two LHCP) have been optimized to cover a frequency band from 2 $\mathrm{G}$$\mathrm{Hz}$ to $2.3$ $\mathrm{G}$$\mathrm{Hz}$. To fully validate the antennas for space application, mechanical and thermal tests have been performed but are omitted in this paper for the sake of brevity. The manufactured prototype and the radome are shown in Figure 14.

Simulation and measurement of the return loss are shown in Figure 15.

The TT&C frequency bandwidth where a $|S_{11}|<-10$ dB ranges from $f_L=1.92 \mathrm{G}\mathrm{Hz}$ to $f_H=2.34 \mathrm{G}\mathrm{Hz}$ which correspond to a bandwidth of $19.7$%. The antenna has been optimized to have an AR lower than $3.2$ dB over the TT&C bandwidth. The simulated and measured results for the AR and realized gain are presented in Figure 16. Simulations agree well with measurements.

Figure 17 shows simulated and measured radiation patterns at different frequencies.

6. Conclusions

In this paper, a suspended patch antenna has been developed and optimized. The measured radiation pattern, axial ratio, and gain are in good agreement with the simulated ones and validate the electromagnetic design. The mechanical and thermal tests confirm that all the material used substrate, screws, as well as design criteria, are appropriate to get a space-qualified design. To realize a TT&C system, four of these antennas (two RHCP and two LHCP) have been employed. In this way, a quasi-isotropic radiation pattern is obtained. It is worth pointing out that the chosen antenna design if fully compliant with the system requirements which are a $-3$ dB beamwidth wider than 40${}^{\circ }$, a maximum gain higher than 6 dBi and an axial ratio below 4 dB in the whole 2.0–2.3 GHz operating band. Moreover, the proposed antenna exhibits a wide impedance bandwidth of $19.7$%.