 |
D. Laffey, Tyndall National Institute, Cork, Ireland
J. Buckley, Tyndall National Institute, Cork, Ireland
J. Barton, Tyndall National Institute, Cork, Ireland
S.C. O'Mathuna, Tyndall National Institute, Cork, Ireland
Design/methodology/approach – The experimental test methodology, fixtures, conditions and results used to characterize the performance of these antennas (both state of the art commercially available and those developed in-house) are presented.
Findings – The underwater test results show a near omni-directional pattern about the three principal axes, thus showing that a spherical field has been achieved for localisation purposes (with a certain loss of resolution).
Originality/value – The paper focuses on the development of a new antenna scheme to enable under water communications between robotic agents.
1 Introduction
The continued miniaturisation of wireless sensor network requires the use of small, high-performance antennas that are often one of the key components in determining the overall performance of a wireless sensor network. Tyndall's miniaturised wireless sensor mote is shown in Figure 1(a), utilizing a commercially available planar integrated antenna. The main advantage of this antenna is its small footprint, allowing for integration into the RF transceiver layer shown in Figure 1(b), where the antenna is shown on the right-hand side. This antenna, however, has a number of disadvantages in its current implementation including limited bandwidth (BW), efficiency and gain, thereby limiting the overall performance of the sensor node in different deployment scenarios.
The particular implementation discussed in this paper required the development of a new antenna scheme for the motes to enable under water communications between the robotic agents developed as part of the SOCIAL (SOCIALIST project fact sheet, http://cordis.europa.eu/fetch?ACTION=D&CALLER=PROJ_IST&QM_EP_RCN_A=67037) project. These robotic agents, known as collaborative autonomous agents (CAAs), were developed on the Hydron (Østergaard et al., 2005) platform utilising the Tyndall Mote (O'Flynn et al., 2005) as the communications platform for the system. The CAA, shown in Figure 1(c), measures 10 cm in diameter. The RF communication system in the Hydron was designed for two major functions: primarily the system should be able to send sensory information to a base station, but it should also generate a field that is roughly spherical in shape to allow for CAA tracking using a received signal strength indicator (RSSI) at the receiver antenna.
The particular problem with the immersion of the RF system in water is that water is an extremely lossy material, with an attenuation of 382 dB/m at 3 GHz (Jacobi et al., 1979). This is due to the fact that water molecules are quite polar, with the poles being rotated by the alternating electromagnetic field, which in turn heats the water. This principle is used in microwave ovens and thus converts the energy that was to be used for transmission into heat.
2 Antenna system development for underwater application
The goal of the antenna systems design was to:
- Provide an antenna with good performance at 2.45 GHz and low size.
- Generate a spherical (or near spherical) radiation pattern about the Hydron structure.
- Implement a system to overcome the attenuation of water and the proximity of the antennas to the electronics, and the lead ballast in the CAA.
2.1 Number of antennas
The RF field generated by the antenna needs to be spherical so that the RSSI measurement can be used for localisation (range finding) purposes. As shown above in Figure 2, it can be seen that a number of “real” antenna need to be deployed so as to emulate an “ideal” antenna which would give the required spherical field. The number of antennas required is a factor of the desired shape of the field, the hydro-dynamics of the CAA (rotation being a problem), and the necessity to overcome the detuning factors of the lead ballast, the electronics and the water. In theory, two ideal antennas with the pattern shown in Figure 2(a) would provide communication in all directions, but, due to rotation and because of the fact that the orientation of the CAAs may not be aligned, it was decided that the more antennas the better. Four was chosen as a practical compromise due to the space constraints inside the Hydron (Figure 3).
2.2 Antenna design considerations
When designing an antenna there are a number of desirable antenna characteristics that are striven for. These include: a low S11 at the resonant frequency of 2.45 GHz, and a high gain and BW and a large groundplane for improved efficiency.
The desirable design characteristic of a large groundplane had to be compromised due to a lack of space in the CAA. The antenna used in the Hydron can be seen below in Figure 4. It has an S11 of roughly −13 to −15 dB in air at 2.45 GHz. It has an MMCX connector, a 50 Ω transmission line, and the groundplane is 25 × 25 mm. The total size of the board is 25 × 32.5 mm. The antenna component is a Lynx ANT-2.45-CHP commercial chip antenna. Antenna system simulations were undertaken using HFSS so as to optimise the efficiency of the antenna system and to minimise the degradation of the RF signal due to the aquatic medium the CAA is deployed in. An antenna switching system under the control of the microcontroller in the Tyndall mote was implemented so as to be able to switch between the different antennas as required.
2.3 Tuning the antennae for optimal transmission
The position of the antennas was important due to the fact that they were detuned by differing amounts in different positions. It was required therefore that this detuning effect be kept to a minimum. So as to help achieve this a number of readings of the S11 parameter of the antenna was read when it was kept at various distances from the shell of the Hydron as it was immersed in water. Figure 5 shows the results and the test fixture used. A distance of 8 mm was used in the actual CAA as it was felt that this was the optimum compromise between the S11 characteristic of the system and the form factor required by the CAA itself.
3 Results for the RF field shape in air and underwater
In order to test whether or not the RF system was producing a spherical field it was decided that the best method available to us was to use the RSSI function available on the transceiver chip. This was used to establish a relative field strength pattern about the Hydron at a fixed distance over a number of points about a given axis. An initial field-shape test was carried out in air at 75 cm by rotating the CAA about its y-axis at increments of 10°. The 100 RSSI readings were taken at each angle (25 for each of the antennas), and these were used to form a relative field shape diagram (Figure 6). This established a reference field-shape as well as tested the experimental approach in a more accessible environment. The field in air was found to be far more omni-directional than that of the water tests, discussed below.
3.1 Water tests
To perform under-water tests it was required that a gantry was developed to hold the Hydron at a specific depth and angle. This was developed entirely using Perspex, to minimize the effect of the apparatus on the results and was used to perform the same test as described above around the three principal axis of the Hydron. The results represent the shape of the RF field in each plane. Owing to the degradation effects of the water medium of the CAA, and the effect of the proximity of conductors to the detuned antenna, the RF fields can be seen to be non spherical as in Figure 7. A scaling factor can be utilised on the measured results to make the field “appear” more omni-directional to a positioning algorithm. Utilising this scaling factor, however, decreases the resolution of the positioning algorithm by a proportion equal to the scaling factor and also decreases the resolution of the positioning system by the same amount. Figure 7 also shows the field shape with a scaling factor applied to the data set.
These results show a near omni-directional pattern about the three principal axes. This shows that a spherical field has been achieved for localisation purposes (with a certain loss of resolution). Certainly the field generated about the Hydron is usable not only for communication, but also (with suitable averaging functions and analysis of the field) it is usable for the function of tracking the Hydron.
4 Conclusions
The results can be improved by investing more time and effort into making sure that the antennas are placed in the optimal positions. This work would be extremely time-consuming, however, involving many minute movements of each antenna, exact recording of placement and orientation as well as the chore of having to re-run each and every axis reading for each movement. Another method of improving the results would be to redesign the tangible agent, modelling its electronics and ballast exactly, and incorporating the antennas into the superstructure. The easiest way to improve the results would be to use a much better receiver antenna outside the tank, as this is not limited by size or shape.
 |
 |
 |
 |
 |
 |
 |
| References |
|
|
|
|
|
|
|
|
|
|
|
Brendan O'Flynn can be contacted at: brendan.oflynn@tyndall.ie |