The channel model optimization algorithm plays a critical role in novel communication method research. Wireless sensor connections using capacitive coupling communication inside a metal cabinet such as spacecraft are an emerging communication technology. However, channel modeling along with optimization methods have not been systematically investigated. In this paper, a modified artificial immune algorithm (MAIA) was developed to optimize a few tens of model parameters for the capacitive coupling communication channel within a metal cabinet. The mathematical model of the communication channel was derived from the equivalent circuit model by analyzing the capacitive coupling electric field distribution. Unknown parameters in the model were optimally estimated by adopting MAIA with the objective of minimizing the root mean square error (RMSE) between the model computed data and simulation or experimental data. The proposed scheme enhanced the convergence performance by incorporating the artificial bee colony (ABC) algorithm, modifying the strategies of immune operations and introducing a similarity detection step. Validation results showed that the frequency response of the optimized model matched well with the simulation and experimental data, verifying the feasibility and robustness of the proposed MAIA. Compared with three other state-of-the-art ABC algorithms and three enhanced intelligent algorithms, it was demonstrated that the proposed algorithm performed better with respect to convergence speed and accuracy. The study provided a multiparameter channel model estimation solution for capacitive coupling communication within a metal cabinet research.
Capacitive coupling intra-body communication (CC-IBC) has become one of the candidates for healthcare sensor networks due to its positive prevailing features of energy efficiency, transmission rate and security. Under the CC-IBC scheme, some of the electric field emitted from signal (SIG) electrode of the transmitter will couple directly to the ground (GND) electrode, acting equivalently as an internal impedance of the signal source and inducing considerable energy losses. However, none of the previous works have fully studied the problem. In this paper, the underlying theory of such energy loss is investigated and quantitatively evaluated using conventional parameters. Accordingly, a method of electric active shielding is proposed to reduce the displacement current across the SIG-GND electrodes, leading to less power loss. In addition, the variation of such loss in regard to frequency range and positions on human body was also considered. The theory was validated by finite element method simulation and experimental measurement. The prototype result shows that the receiving power has been improved by approximate 5.5 dBm while the total power consumption is maximally 9 mW less using the proposed technique, providing an energy efficient option in physical layer for wearable and implantable healthcare sensor networks.
In order to solve the problem of extracting signals from impulse radio ultra-wideband (IR-UWB) receivers in a low signal-to-noise ratio (SNR) environment, this paper uses circuit transient analysis to establish a mathematical model of an ultra-wideband wireless communication receiver based on a balanced sampling and integration circuit (BSIC). The effect of receiver circuit component parameters on the output signal is simulated and tested, and the optimization of circuit component parameters for UWB wireless communication receivers is achieved. The sampling capacitance in the receiver circuit ranges from 1pF to 6pF, depending on the 200ps pulse width. The output signal amplitude increases as the sampling capacitance increases. The range of integral capacitance is from 2.5pF to 25pF, which is based on a 100ns interval between two pulses. The output signal amplitude decreases as the integration capacitance increases and the signal waveform becomes better as the integration capacitance increases. The effect of SNR from 0 to 30dB on the receiver output is simulated, and the results show that the Bit Error Ratio (BER) of the receiver is less than 10−3 when the SNR is greater than 15dB. The simulation and test results show that the model developed in this paper is useful as a guide for optimizing the receiver parameters at low SNR.