Xianyue Wu

Antenna Design and Channel Measurements for On-Body Communications at 60 GHz

On-body communication is of increasing interest for a number of applications, such as medical-sensor networks, emergency-service workers, and personal communications. This paper reviews 60 GHz on-body communication and its benefits and challenges. Two novel low profile high gain, end-fire wearable antennas are then described. Measurements with an experimental phantom and real human body are presented. Results show antennas achieve good performance close to a phantom. Shadowing effects and polarisation issues for on-body communications at 60 GHz are discussed.

Novel Yagi-Uda antennas for on-body communication at 60GHz

On-body communication is of increasing interest for a number of applications, such as medical-sensor networks, emergency-service workers, and personal communications. This paper reviews 60 GHz on-body communication and its benefits and challenges. Two novel low profile high gain, end-fire wearable antennas are then described. They are promising candidates for on-body channels at 60 GHz. The first one is a conventional printed Yagi array, which gives up to 17 dB gain and 6 GHz bandwidth. The other is a substrate integrated waveguide Yagi antenna which also gives 15 dB gain. Because these two antennas have different polarization, they can be used to investigate 60 GHz on-body channels in future work.

Body-area propagation at 60 GHz

This paper reviews recent results on radiowave propagation measurements, models and electromagnetic analysis methods for 60 GHz body-area networking (BAN) applications. The viability of mm-wave BANs is discussed and established. The advantages of adopting a mm-wave band for body area networking over lower frequency microwave bands are discussed and quantified. Finally, a number of outstanding research challenges are identified.

Interuser Interference in Adjacent Wireless Body Area Networks

The interuser interference between wireless body area networks worn by two moving persons in an indoor environment at 60 and 2.45 GHz is experimentally investigated. Both omnidirectional antennas (monopoles) and directional antennas (horns) were used in the measurements. The interference power-level variation and carrier-to-interference ratio (CIR) were measured and characterized. Median interference power-level reduction of nearly 20 dB was achieved in all measured channels by adopting 60-GHz radio transmissions compared to 2.45 GHz, both with omnidirectional on-body antennas. A further 20 dB of interference-level reduction was achieved at 60 GHz by adopting directional antennas confining the radiated wave along the body surface. Level crossing rates for interference power variation using omnidirectional antennas range from 2.7 to 7.2 s-1 at 2.45 GHz and 32-64 s-1 at 60 GHz for static to progressively more dynamic links, whereas the corresponding range using 60-GHz directional antennas is reduced to 39-54 s-1. The median improvement in the instantaneous CIR for the chest-head channel between 60 and 2.45 GHz was approximately 30 dB. The measured interference power level and CIR, in dB, were found to satisfactorily fit the normal distribution according to the normalized root-mean-square error-based fit metric.

De-Polarization of On-Body Channels and Polarization Diversity at 60 GHz

Measurements of on-body dynamic propagation channels have been performed at 60 GHz using dual-polarized scalar horn antennas. Comparison of the statistics of the measured signals showed that the choice of polarization (vertical or horizontal) does not affect the path losses significantly, and the relative polarization of the two antennas depends on their placement on the body. In volatile links, such as those from the wrist to other parts of the body, random movements equalize the differences between different polarization configurations. These depolarization effects can be exploited to improve link performance through the use of receive diversity with maximal ratio combining. More advanced multiple-input multiple-output diversity methods are found to produce only marginally better performance compared to receive maximal ratio combining.

Preliminary estimate for observability of 60 GHz wireless body area networks

Observability of 60 GHz wireless body area networks is being investigated, in the context of the maximum outdoor detection range for such a network. An off-body channel within a scattering environment has been characterized. By decomposing the detection channel into a free-space channel and a body-local environment channel, an observability estimation model is proposed based on the measured data.

Investigation of inter-user interference of wireless body area networks at 60 GHz

Inter-user interference between two wireless body area networks at 60 GHz has been investigated. A series of data sets have been gathered in a laboratory environment for random realistic activities and body orientations, and have been subjected to statistical analysis. The characterization of interfering signal strength variations for several links and statistical parameters such as level crossing rate and average peak duration are presented.

Effect of wearable antenna polarization and directivity on on-body channel path gain at 60 GHz

Measurements of on-body dynamic propagation channels have been performed at 60 GHz using dual-polarized scalar horns. The measured cumulative distribution functions (CDF) were compared to those reported previously with monopole antennas. The performances of horn and monopole antennas were compared in various on-body channels and it was found that omnidirectional low-gain antennas produce stronger path gain in some channels. It was concluded that polarization diversity is unlikely to provide a significant gain but the channel capacity may be improved by using several polarization channels in a multiple-input multiple output (MIMO) system.

BodySim: a multi-domain modeling and simulation framework for body sensor networks research and design

Modeling and simulation play important roles in engineering research and design. These techniques are especially helpful in the early phases where limited detail is available about the design and where design changes are less costly. In addition, high-fidelity models can be employed at the later stages to complement testing. Models are also important research tools for understanding complex phenomena.