sensors Article
An Energy Efficient Technique Using Electric Active Shielding for Capacitive Coupling Intra-Body Communication Chao Ma 1 , Zhonghua Huang 1 , Zhiqi Wang 1 , Linxuan Zhou 2 and Yinlin Li 1, * 1 2
*
School of Mechatronical Engineering, Beijing Institute of Technology, Beijing 100081, China;
[email protected] (C.M.);
[email protected] (Z.H.);
[email protected] (Z.W.) The Bradley Department of Electrical and Computer Engineering, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, USA;
[email protected] Correspondence:
[email protected]; Tel.: +86-010-6891-1081
Received: 24 July 2017; Accepted: 5 September 2017; Published: 8 September 2017
Abstract: 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. Keywords: intra-body sensor networks; capacitive coupling; electric active shielding; energy efficiency; finite element method (FEM)
1. Introduction Intra-body Communication (IBC) provides an appealing wireless connection of wearable and implantable biomedical sensors for ubiquitous monitoring of the patient’s physiological parameters. Due to its energy efficiency and security features [1,2], IBC was selected as a physical layer candidate in the IEEE 802.15.6 (WBAN) standard. IBC, which uses body tissue as the signal transmission medium, is generally divided into two categories: galvanic coupling and capacitive coupling [3]. Galvanic coupling uses electrodes in direct contact with the skin of the subject [4,5], whereas the signal electrode of the capacitive coupling can be either attached to the skin directly or separated by clothes or insulators from the skin [6]. In addition, capacitive coupling has a higher gain and a relatively higher frequency range of operation, between 1 and 100 MHz, which enables higher data transmission rates [7–10] and less power consumption [1] than the galvanic coupling method. As a result, capacitive coupling has been one of the most active research topics in the IBC field, mainly focusing on channel modeling [10–16], measurement [17,18], physical principle analysis and prototype development [19,20] to date. The conventional capacitive coupling IBC, depicted in Figure 1, was put forwarded by Zimmerman in 1995 [21]. For the electric field model in Figure 1a, the signal (SIG) electrodes of transmitter and receiver Sensors 2017, 17, 2056; doi:10.3390/s17092056
www.mdpi.com/journal/sensors
Sensors 2017, 17, 2056
2 of 14
focusing on channel modeling [10–16], measurement [17,18], physical principle analysis and prototype development [19,20] to date. Sensors 2017, 17, 2056 2 of 14 The conventional capacitive coupling IBC, depicted in Figure 1, was put forwarded by Zimmerman in 1995 [21]. For the electric field model in Figure 1a, the signal (SIG) electrodes of are the attached to (GND) the body while theare ground (GND)inelectrodes floating aretransmitter attached toand the receiver body while ground electrodes left floating the air. Inare thisleft case, human in the air. In this case, human body is modeled as the node of a perfect conductor, and the electric body is modeled as the node of a perfect conductor, and the electric couplings among the electrodes of the couplings among the electrodes of the transceiver, body and ground plane are modeled as capacitors. transceiver, body and ground plane are modeled as capacitors. In Figure 1b, an equivalent circuit model In Figure 1b, an equivalent circuit model is presented accordingly. As shown in the figure, the is presented accordingly. As shown in the figure, the capacitances of the various electrode pairs play capacitances of the various electrode pairs play a critical role to determine the path transmission gain, a critical role to determine the path transmission gain, which significantly affects both size and power which significantly affects both size and power consumption of the IBC system. Therefore, the consumption of the IBC system. Therefore, the configuration of the electrodes is clearly considered to be configuration of the electrodes is clearly considered to be the most important consideration in the IBC thesystems most important consideration in the IBC systems [1]. [1] .
(a)
(b)
Figure 1. (a) The conventional electric field model; (b) The circuit model of capacitive coupling IBC. Figure 1. (a) The conventional electric field model; (b) The circuit model of capacitive coupling IBC.
Several studies have been conducted on the electrode configuration of capacitive coupling IBC. In Several studies have conducted on the electrode configuration capacitive coupling [13], the impedance of been capacitive IBC transmitter’s electrodes for bothof signal-ground and IBC. In [13], the was impedance ofThe capacitive IBC transmitter’s electrodes both were signal-ground and electrode–skin provided. transmission gains of the capacitive IBCfor channel investigated electrode–skin was provided. The transmission gains of the capacitive IBC channel were investigated in [15] with the frequency range from 100 kHz to 100 MHz, respecting to horizontal and vertical in electrode [15] withsetups the frequency from 100material. kHz to 100 and vertical as well asrange three electrode The MHz, energyrespecting loss due toto thehorizontal return path parasitic capacitance between GNDThe electrodes groundpath plane was electrode setups as wellthe astransmitter/receiver’s three electrode material. energyand lossthe dueexternal to the return parasitic studied in between [22]. However, there has been a lack ofGND research on the effect theexternal parasiticground capacitance capacitance the transmitter/receiver’s electrodes and of the plane between SIG and GND electrodes of the transmitter, designated as equivalent capacitor A in Figure 1. was studied in [22]. However, there has been a lack of research on the effect of the parasitic The parasitic capacitor A, connected parallel signal source, could aspotentially the capacitance between SIG and GND electrodes of to thethe transmitter, designated equivalentleak capacitor displacement current directly from the SIG to GND electrode, which can lead to an internal power A in Figure 1. The parasitic capacitor A, connected parallel to the signal source, could potentially consumption increasecurrent and a directly transmission gainSIG loss. the parasitic capacitance the leak the displacement from the to While GND electrode, which can leadbetween to an internal SIG-GND electrodes of the transmitter has been identified explicitly in [12,13,21,23], their influences power consumption increase and a transmission gain loss. While the parasitic capacitance between the on the power consumption and transmission gain as well as corresponding solutions have not been SIG-GND electrodes of the transmitter has been identified explicitly in [12,13,21,23], their influences fully studied so far. on the power consumption and transmission gain as well as corresponding solutions have not been In this study, our first aim is to analyze the power consumption caused by the parasitic fully studied so far. capacitance between transmitter electrodes of the capacitive coupling IBC. Secondly, we intend to In this study, our first aim is to analyze the power consumption caused by the parasitic propose an electrode configuration using the scheme of capacitive reflector [24] to reduce such power capacitance between transmitter electrodes of the capacitive coupling IBC. Secondly, we intend to loss. Considering the fact that the vertical electrodes (parallel-plate structure) [1,12,13,22–25] of the propose an electrode configuration usinginthe scheme of transmission capacitive reflector reduce such power transmitter is the optimal configuration regard to the gain [1][24] andto system size, as well loss. Considering the fact that the vertical electrodes (parallel-plate structure) [1,12,13,22–25] as that the vertical electrode has minimal plates distance and hence a larger amount of energy lossofinthe transmitter is the optimal to theelectrode transmission gainof [1]the and system size, as well the system, therefore ourconfiguration study focusesin onregard the vertical structure transmitter. as that This the vertical has plates distance and amountcapacitance of energy loss paper iselectrode organized as minimal follows: Section 2 presents thehence effect aoflarger the parasitic and in thethe system, our study the the vertical electrode structure of the transmitter. theorytherefore of the electric active focuses shieldingon from perspective of circuit model. In Section 3, the finiteThis paper is organized as follows: Section 2 presents effect ofand thesimulated parasitictocapacitance element method（FEM）model of a capacitive coupling IBC is the established evaluate and theory of the electric activefrom shielding fromofthe perspective of circuit model. InSection Section thethe effect of the shield electrode the angle electric field strength distribution. 4 3, the finite-element method (FEM) model of a capacitive coupling IBC is established and simulated to evaluate the effect of the shield electrode from the angle of electric field strength distribution. Section 4 presents an experimental validation of the shield electrode using a prototype transmitter powered by battery. Section 5 presents the final conclusions of this paper.
Sensors 2017, 17, 2056
3 of 14
presents experimental validation of the shield electrode using a prototype transmitter powered Sensors 2017, 17,an 2056 3 of 14 by battery. Section 5 presents the final conclusions of this paper. 2. Effect of Electrodes Parasitic Capacitance andand Electric Active Shielding 2. Effect of Electrodes Parasitic Capacitance Electric Active Shielding TheThe vertical electrode plate pairpair in parallel structure is conceptually equivalent to atocapacitor. vertical electrode plate in parallel structure is conceptually equivalent a capacitor. Ideally, the parasitic capacitance of the vertical electrode in free space is proportional to the to area its of Ideally, the parasitic capacitance of the vertical electrode in free space is proportional theofarea plates inversely proportional to the plate separation. When an alternating signal source is applied its and plates and inversely proportional to the plate separation. When an alternating signal source is to the plates, a displacement current flow through them, as illustrated in Figure 2a. The equivalent applied to the plates, a displacement current flow through them, as illustrated in Figure 2a. The circuit model iscircuit presented in Figure 2b, where the parallel-plate is modeled as capacitor CA . When the equivalent model is presented in Figure 2b, where the parallel-plate is modeled as capacitor parallel-plate is placed on human body as in Figure 2c, the current supplied by the signal CA. When the parallel-plate is placed ondepicted human body as depicted in Figure 2c, the current supplied source is divided into two branches: some of the current flows across the parallel-plate directly, and the by the signal source is divided into two branches: some of the current flows across the parallel-plate restdirectly, flows through all the way passing the forward path of human body and then the return path and the rest flows through all the way passing the forward path of human body and then from ground to theground negative port the signal source. Accordingly, a simplified circuit model the return plane path from plane toof the negative port of the signal source. Accordingly, a simplified is built in Figure 2d, where C and C are the equivalent capacitors of forward path and return path,and F circuit model is built in Figure 2d,Rwhere CF and CR are the equivalent capacitors of forward path respectively. Obviously, the capacitor CA the divides a portion of the current away from the branch with the return path, respectively. Obviously, capacitor CA divides a portion of the current away from CF , branch resulting in a power loss in the same way as the internal impedance of a signal source. Therefore, with CF, resulting in a power loss in the same way as the internal impedance of a signal source. it isTherefore, desirable toitfind a solutiontotofind minimize the internal displacement currentdisplacement (IDC) flowingcurrent across the is desirable a solution to minimize the internal (IDC) capacitor CAacross for thethe purpose of reducing consumption. flowing capacitor CA for thepower purpose of reducing power consumption.
(a)
(b)
(c)
(d)
Figure Themodel modelofofthe the parasitic parasitic capacitances electrodes of IBC system. (a) The Figure 2. 2. The capacitanceson onthe thetransmitter transmitter electrodes of IBC system. parasitic capacitance between SIG-GND electrodes; (b) Simplified circuit model of (a); (c) (a) The parasitic capacitance between SIG-GND electrodes; (b) Simplified circuit model of (a);The parasitic capacitances among electrodes, human body andand ground plan; (d) (d) Simplified circuit model (c) The parasitic capacitances among electrodes, human body ground plan; Simplified circuit of (c). model of (c).
Inspired by the concept of electric active shielding (or capacitive reflector) proposed by National Inspired by the concept of electric active shielding (or capacitive reflector) proposed by National Aeronautics and Space Administration’s (NASA) Goddard Space Flight Center [24], a capacitive Aeronautics and Space Administration’s (NASA) Goddard Space Flight Center [24], a capacitive shield shield electrode is introduced into the electrode configuration of the IBC transmitter in order to electrode is introduced into the electrode configuration of the IBC transmitter in order to reduce the IDC, as depicted in Figure 3. The objective of NASA’s capacitive reflector is to develop a proximity
Sensors 2017, 17, 2056 Sensors 2017, 17, 2056
4 of 14 4 of 14
reduce the IDC, as depicted in Figure 3. The objective of NASA’s capacitive reflector is to develop a sensing that would allowinaFigure robot 3. to The intruding objects capacitive without blind spots updevelop to one foot. reduceskin the IDC, as depicted objective of NASA’s reflector is to a to proximity sensing skin that would allow asense robot to sense intruding objects without blind spots up It uses a capacitive sensing element backed by a reflector element, which is driven by the same voltage proximity sensing skin that would allow a robot to sense without blind spots up one foot. It uses a capacitive sensing element backed by intruding a reflectorobjects element, which is driven bytothe as the sensing electrode. Acting as a shield to reflect all field lines away from the grounded robot arm, one foot. It uses a capacitive sensing element backed by a reflector element, which is driven by the same voltage as the sensing electrode. Acting as a shield to reflect all field lines away from the same voltage as the sensing electrode. Acting as a shield to reflect all for fieldthe lines away of from the the range of the sensor can be extended. In our study, the shield is used purpose reducing grounded robot arm, the range of the sensor can be extended. In our study, the shield is used for the grounded robottherefore arm, the the range of theloss sensor can IBC be extended. In our study, the shield is used for the the IDC instead; power for the system can reduced. purpose of reducing the IDC instead; therefore the power loss forbethe IBC system can be reduced. purpose of reducing the IDC instead; therefore the power loss for the IBC system can be reduced. As illustrated in Figure 3a, the shield shield electrode electrode for the GND GND As illustrated in Figure 3a, the for IBC IBC transmitter transmitter has has the the same same size size as as the As illustrated in Figure 3a, the shield electrodeelectrodes. for IBC transmitter has the same size amplifier as the GND electrode, and it is placed between the SIG-GND Moreover, an isolation (IA) electrode, and it is placed between the SIG-GND electrodes. Moreover, an isolation amplifier (IA) is electrode, and it is placed between the SIG-GND electrodes. Moreover, an isolation amplifier (IA) is is deployed to connect the SIG and shield electrode, keeping the voltage of the two electrodes in deployed to connect the SIG and shield electrode, keeping the voltage of the two electrodes in the deployed to connectand the phase. SIG andThe shield electrode,between keeping the of the two electrodes electrodes in thethe the same amplitude capacitance the voltage GND and shield and same amplitude and phase. The capacitance between the GND GND andshield shield electrodes and the input same amplitude and phase. The capacitance between the and electrodes and the input input impedance of IA are referred as C and Z , respectively. The model of the shielded transmitter S IA impedance , respectively. The model of the shielded transmitter impedanceofofIAIAare arereferred referred as as CCSS and and Z ZIA IA, respectively. The model of the shielded transmitter electrodes along with with its simplified circuit arebuilt builtininFigure Figure3b. 3b.AsAs can be seen, theS C ZIA S and electrodes along its simplified circuit are can seen, the and IA are electrodes along with its simplified circuit are built in Figure 3b. As can bebeseen, the CSCand ZIAZare are connected in series, and hence the current of this branch decreases when the Z is introduced. IA connected inin series, branch decreases decreaseswhen whenthe theZIA ZIAis is introduced. Since connected series,and andhence hencethe thecurrent current of of this this branch introduced. Since Since the input impedance of an IA is up to 10 GΩ, the IDC flowing from the SIG to GND electrode thethe input impedance IDC flowing flowingfrom fromthe theSIG SIGtotoGND GND electrode becomes input impedanceofofan anIA IAisisup upto to10 10GΩ, GΩ, the the IDC electrode becomes becomes negligible. negligible. negligible.
(a) (a)
(b)
(b)
Figure 3. The principle of the shield electrode applied to the IBC transmitter electrodes. (a) The Figure 3. The The principle of shield the shield electrode applied to the IBC transmitter electrodes. Figure 3. principle the to the IBCshield transmitter electrodes. (a)itsThe capacitive-coupled IBC of with the shieldelectrode electrode;applied (b) Model of the electrodes along with (a) The capacitive-coupled IBC with the shield electrode; (b) Model of the shield electrodes along capacitive-coupled IBC with the shield electrode; (b) Model of the shield electrodes along with its simplified circuit. with its simplified circuit.
simplified circuit.
In order to exemplify the effect of the shield electrode quantitatively, the ratio of receiving In order totoexemplify effect of the shield electrode the ratio receiving current order exemplify the effect of the shield electrode quantitatively, theof ratio of receiving current to the source, orthe the current transmission gain, isquantitatively, investigated based on the conventional to the source, or the current transmission gain, is investigated based on the conventional model current to the source, or the current transmission gain, is investigated based on the conventional model by Zimmerman. The circuit of Figure 1b can be further described as Figure 4, in which the by Zimmerman. circuit The of can be further described aselectrode Figure 4,isin which the4,capacitor is model by Zimmerman. Figure 1b can be shield further described as Figure in whichCAthe capacitor CAThe S and1b Zof IA in serial when the investigated. is replaced byFigure Ccircuit replaced and ZIA in when theserial shield electrode is investigated. capacitorby CACisS replaced ZIA in when the shield electrode is investigated. byserial CS and
Figure 4. Simplified conventional IBC lumped circuit.
Figure Simplified conventional IBC lumped circuit. Since the signal source and4. capacitor CA are connected in parallel, Figure 4.the Simplified conventional IBC lumped circuit.the current source can be used for analysis. According to the Kirchhoff’s current law, the relation of the currents in Figure 4 is Since the as signal source and the capacitor CA are connected in parallel, the current source can be represented follows:
used for analysis. According to the Kirchhoff’s current law, the relation of the currents in Figure 4 is represented as follows:
Sensors 2017, 17, 2056
5 of 14
Since the signal source and the capacitor CA are connected in parallel, the current source can be used for analysis. According to the Kirchhoff’s current law, the relation of the currents in Figure 4 is represented as follows: i4 /i3 = ZE /( ZE + Z1 ) i /i = Z / Z + Z 3 2 2) B ( B (1) i2 /i1 = Zint /( Zint + Z3 ) i1 /i0 = Ri /( Ri + Z A //Z3 ) where ZB , ZC , ZD , ZE , ZF , ZH, ZG are the impedance of capacitors CB , CC , CD , CE , CF , CH, CG , respectively. Z1 , Z2 and Z3 can be written as: Z1 = ZF + ZG + ZH //Ro Z2 = ZD + ZE //Z1 Z = Z + Z //Z 3 2 B C
(2)
and Zint is the impedance of the parasitic capacitance between SIG-GND electrodes, given by: ( Zint =
1 jωC A ,
ZI A +
without shield electrode 1 jωCS ,
with shield electrode
(3)
In reality, the distance between the shield electrode and the GND electrode can be regarded as the one between SIG and GND, thus C A ≈ CS (4) Accordingly, based on the set of equations in (1), the transfer function of the transmission current gain can be written as: G=
ZE Z B R i i4 = i0 ( ZE + Z1 )( ZB + Z2 )(1 + Z3 /Zint )( Ri + Z A //Z3 )
(5)
As can be seen from Equation (5), the current gain is proportional to the value of Zint , indicating that the gain is improved by introducing the shield electrode. To evaluate the effect of the shield electrode on the power consumption, the software Multisim is used. Multisim (National Instruments, Austin, TX, USA) is an electronic schematic capture and simulation program that employs the original Berkeley SPICE software. Simulating the circuit with SPICE is the industry-standard way to verify circuit operation at the transistor level before committing to fabrication. It provides a reliable approach to achieve the power consumption of the circuit, with the consideration of the power consumed by the isolation amplifier. The conventional IBC circuit model is constructed in Multisim workbench, as illustrated in Figure 5. For analysis purpose, the signal source is transformed equivalently from current source into voltage source with an internal resistor connected in series. The chip AD8005 from Analog Device is used as the isolation amplifier, which has a bandwidth of 270 MHz and a maximal power supply consumption of 2 mW. Since the equivalent circuit of the human model and the receiver in Multisim are the same as in Figure 4, the rest of the circuit is represented as a gray rectangle in Figure 5. The total power consumption of the circuit is defined as: ( ∗ ), Re(Urms Irms without shield electrode P= (6) ∗ Re(Urms Irms ) + V Idc , with shield electrode where Urms is the complex effective voltage of the signal source and I∗ rms is the complex conjugate of the signal source effective current, V is the DC power supply to the isolation amplifier and Idc is its supply current.
Sensors 2017, 2056 Sensors 2017, 17, 17, 2056 Sensors 2017, 17, 2056
6 6ofof1414 6 of 14
Figure 5. The conventional IBC lumped circuit built in Multisim software. Figure5.5.The The conventional conventional IBC lumped Figure lumpedcircuit circuitbuilt builtininMultisim Multisimsoftware. software.
The power consumption in Equation (6) is simulated with sweeping frequencies and The power consumption in Equation (6) is simulated with sweeping frequencies and The powerof consumption in Equation simulated with increases sweepingfrom frequencies capacitances capacitances transmitter electrode. As(6) theissignal frequency 100 kHzand to 50 MHz, the capacitances of transmitter electrode. As the signal frequency increases from 100 kHz to 50 MHz, the consumption is obtained as in Figure 6a. The results show total consumption of power transmitter electrode. As the signal frequency increases fromthat 100the kHz to power 50 MHz, the power power consumption is obtained as in Figure 6a. The results show that the total power consumption is significantly reduced as when frequency is larger than 300 that KHzthe by total usingpower the shield electrode.is consumption is obtained in Figure 6a. The results show consumption is significantly reduced when frequency is larger than 300 KHz by using the shield electrode. Moreover,reduced the capacitor A is varied pF to nF at MHz, in Moreover, order to significantly whenCfrequency isfrom larger30than 30010KHz byfrequency using the point shield10 electrode. Moreover, the capacitor CA is varied from 30 pF to 10 nF at frequency point 10 MHz, in order to emulate theCAvarious electrode configuration conditions, i.e. electrode sizes, distances, etc. The the capacitor is varied from 30 pF to 10 nF at frequency point 10 MHz, in order to emulate emulate the various electrode configuration conditions, i.e. electrode sizes, distances, etc. Thethe calculated result in Figure 6b exhibits thati.e. theelectrode power consumption with etc. shield electrode is much various electrode conditions, sizes, distances, The calculated result calculated result configuration in Figure 6b exhibits that the power consumption with shield electrode is much smaller than that without the shield electrode. Therefore, the effect of parasitic capacitor C A on the insmaller Figure than 6b exhibits that the power consumption with shield electrode is much smaller than that that without the shield electrode. Therefore, the effect of parasitic capacitor CA on the power consumption is non-negligible and the shield electrode could be one of the viable solutions to without shield electrode. Therefore,and thethe effect of parasitic consumption A on power the consumption is non-negligible shield electrodecapacitor could be C one of the the power viable solutions to reduce such losses. is reduce non-negligible and the shield electrode could be one of the viable solutions to reduce such losses. such losses.
(a) (a)
(b) (b)
Figure 6. Calculated result ofthe thetotal totalpower powerconsumption consumption comparison comparison between with and without shield Figure betweenwith with and without shield Figure6.6.Calculated Calculatedresult result of of the total power consumption comparison between and without shield electrode. (a) Power consumption with respect to frequency ranging from 100 kHz to 50 MHz; (b) Power electrode. (a) Power consumption with respect to frequency ranging from 100 kHz to 50 MHz; (b) Power electrode. (a) Power consumption with respect to frequency ranging from 100 kHz to 50 MHz; (b) Power ranging from 30 pF to 10 nF10at frequency consumption with respect to the capacitor CA from consumption respect to the CA ranging 30 pF to 10 from 30nFpFat frequency to 10 nF atMHz. frequency consumptionwith with respect to capacitor the capacitor CA ranging 10 MHz. 10 MHz.
It is that, on the the conventional conventionallumped lumped It noteworthy is noteworthy that,although althoughthe theabove aboveresult resultisis derived derived based based on It is noteworthy that, although the above result is derived based on the conventional lumped circuit model with thethe specific parameters of of human body from Zimmerman, the conclusion can circuit model with specific parameters human body from Zimmerman, the conclusion canalso alsobe circuit model withdistributed the specific circuit parameters of human body fromdue Zimmerman, theof conclusion can also applicable to other models of human body, to the nature parallel connection be applicable to other distributed circuit models of human body, due to the nature of parallel applicable to other circuit models of human body,plates. due to the nature of parallel tobe the signal source of thedistributed parasitic by electrode connection to the signal source of capacitance the parasitic caused capacitance caused by electrode plates. connection to the signal source of the parasitic capacitance caused by electrode plates. 3. Validation Using Finite Element 3. Validation Using Finite ElementMethod Method 3. Validation Using Finite Element Method It has been proven that model the thecapacitive capacitivechannel channel It has been proven thatfinite finiteelement elementmethod method(FEM) (FEM) can can properly properly model It has been proven that finite element method (FEM) can properly model the capacitive channel incorporating thethe electrodes of transmitter [11,12]. In order to further study the effect shieldofelectrode incorporating electrodes of transmitter [11,12]. In order to further study theofeffect shield incorporating the electrodes of transmitter [11,12]. In order to further study the effect of shield on electrode the transmission in a more gain realistic ANSYS Maxwell employed accordingly. ANSYS on the gain transmission in manner, a more realistic manner, isANSYS Maxwell is employed electrode on the transmission gain in a more realistic manner, ANSYS Maxwell is employed Maxwell is a FEM simulation with the capability to solve the magnetic orthe electric field accordingly. ANSYS Maxwell software is a FEM simulation software with the capability to solve magnetic accordingly. ANSYS Maxwell is a FEM simulation software with the capability to solve the magnetic or electric in field distribution a finite region and theprescribed volume with prescribed boundaryIn conditions. distribution a finite regionin and the volume with boundary conditions. the study, or electric field distribution in a finite region and the volume with prescribed boundary conditions. the differences of the electric field distribution around human body between whether having the shield electrode, are compared using the Maxwell FEM software.
Sensors 2017, 17, 2056
7 of 14
Sensors 2017, 17, 2056
7 of 14
In the study, the differences of the electric field distribution around human body between whether having the shield electrode, are compared using the Maxwell FEM software.
3.1. FEM Simulation Model and Setup 3.1. FEM Simulation Model and Setup The FEM model and simulation environment including external ground plane, air, human body The FEM model and simulation environment including external ground plane, air, human body and transmitter electrodes, are established in Figure 7. The ground plane is modeled as a cylinder with and transmitter electrodes, are established in Figure 7. The ground plane is modeled as a cylinder a diameter of 640 cm and a height of 300 cm. The human body is separated from the external ground by with a diameter of 640 cm and a height of 300 cm. The human body is separated from the external a 2 cm height rubber, emulating the effect of shoes. The surrounding air, also modeled as the cylinder, ground by a 2 cm height rubber, emulating the effect of shoes. The surrounding air, also modeled as is designed to be the same size as the ground plane. The transmitter electrode in vertical configuration the cylinder, is designed to be the same size as the ground plane. The transmitter electrode in vertical is placed on the left wrist of human body, and the signal electrode is set to be 0.5 cm above the body, configuration is placed on the left wrist of human body, and the signal electrode is set to be 0.5 cm in orderthe to emulate non-contact electrode setup. Other specifications and dimensions are and denoted above body, in the order to emulate the non-contact electrode setup. Other specifications in Figure 7b. dimensions are denoted in Figure 7b. The human body bodymodel modelisiscomposed composedofof head, neck, chest, arms is modeled The human head, neck, chest, arms andand legs.legs. It is Itmodeled as a as a concentric cylinder or a sphere with multiple layers of skin, fat, muscles and bones, as illustrated concentric cylinder or a sphere with multiple layers of skin, fat, muscles and bones, as illustrated in in Figure Thethicknesses thicknessesofofthe thetissue tissue layers different body parts defined in Table 1, their and their Figure 7b. The layers in in different body parts are are defined in Table 1, and electrical parametersare arekept keptthe thesame sameasasinin [22]. electrical parameters [22].
(a)
(b)
Figure 7. FEM model and simulation environment. (a) FEM model of the IBC channel with external Figure 7. FEM model and simulation environment. (a) FEM model of the IBC channel with external ground. (b) Human body model and the transverse constituent sections. ground. (b) Human body model and the transverse constituent sections. Table 1. Thicknesses of the tissue layers (mm). Table 1. Thicknesses of the tissue layers (mm). Arm Leg Torso Head Skin 1.26 Arm 1.26Leg 1.26 1.26 Torso Head Fat 8.74 8.74 8.74 2 Skin 1.26 1.26 1.26 1.26 Muscle 28 34 30 2 Fat 8.74 8.74 8.74 2 Bone 22 26 20 10 Muscle 28 34 30 2 Bone 22 26 20 10
Neck 1.26 Neck 8.74 1.26 42 8.74 23 42 23
The vertical configuration of a conventional capacitive IBC electrode consists of a SIG and a GND electrode, as shown in Figure 8a. Slightly different from the conventional vertical electrode The vertical capacitive IBC electrode a SIG configuration, an configuration extra electrodeofora conventional shield electrode is introduced between consists the SIG of and GNDand a GND electrode, as shown in Figure 8a. shield Slightly different from theare conventional electrode electrode, as illustrated in Figure 8b. The and GND electrode in the same vertical dimension of configuration, an extra electrode or shield electrode is introduced between the SIG and GND electrode, 3 cm × 3 cm, spaced by 0.2 cm. In order to reduce the parasitic capacitance from the edge of the SIG as in Figure 8b. of The and GND electrode are in the dimension of 3 cm to illustrated GND electrode, the size theshield SIG electrode is set to be smaller thansame the shield electrode, with×a3 cm, dimension cm ×In 3 cm. The areparasitic also made of copper from with athe thickness 0.1 SIG cm. The spaced by of 0.23cm. order toelectrodes reduce the capacitance edge ofofthe to GND electrostatic is selected as is theset solver the 3Dthan FEMthe software, an electric electrode, thesolution size of type the SIG electrode to beinsmaller shield and electrode, withpotential a dimension difference 5 VrmsThe is applied as the excitation. of 3 cm × of 3 cm. electrodes arevoltage also made of copper with a thickness of 0.1 cm. The electrostatic solution type is selected as the solver in the 3D FEM software, and an electric potential difference of 5 Vrms is applied as the voltage excitation.
Sensors 2017, 17, 2056 Sensors 2017, 17, 2056
8 of 14 8 of 14
Sensors 2017, 17, 2056
8 of 14
(a)
(b)
Figure Figure 8. 8. Model Model of of transmitter transmitter electrodes electrodes in in vertical vertical configuration. configuration. (a) (a) Conventional Conventional vertical vertical electrode electrode (a) electrode. (b) The electrode configuration with (b)shield electrode. configuration without shield configuration without shield electrode. (b) The electrode configuration with shield electrode. Figure 8. Model of transmitter electrodes in vertical configuration. (a) Conventional vertical electrode
For the configuration electrode, the voltage excitation is applied configuration withoutwith shield shield electrode. (b) The electrode configuration with shield electrode.to both SIG and For the configuration with shield electrode, the voltage excitation is applied to both SIG and shield shield electrode in order to keep the same electric potential. Moreover, in order to emulate the electrode in order to keep thewith same electric potential. Moreover, in order to emulate the dispersive the configuration shield electrode, voltage applied to both and dispersiveForproperty of human body tissues in the regard to excitation different isfrequencies, theSIG frequencyproperty of electrode human body tissues in regard to different frequencies, the frequency-dependent dielectric shield in order to keep the same electric potential. Moreover, in order to emulate the 500 dependent dielectric constants, conductivities, and loss tangents at frequency points of 100, 200, constants, conductivities, and loss tangents at frequency points of 100,frequencies, 200, 500 kHz, 1,frequency2, 5, 10, 20, 30 dispersive property of human body tissues in regard to different the kHz, 1, 2, 5, 10, 20, 30 and 50 MHz are determined according to that in [22], and configured to be the dependent dielectric constants, conductivities, loss and tangents at frequency of 100, 200, 500tissue and 50 MHz are determined according to that and in [22], configured to bepoints the human body human body tissue layers for each frequency point. 2, 5,frequency 10, 20, 30 and 50 MHz are determined according to that in [22], and configured to be the layerskHz, for 1, each point. human body tissue layers for each frequency point.
3.2. Simulation Results and Discussion 3.2. Simulation Results and Discussion 3.2.representative Simulation Results and Discussion A simulated result of electric field strength distribution around human body at A representative simulated result of electric field strength distribution around human body at 10 MHz isAillustrated in Figure 9. result of electric field strength distribution around human body at representative simulated 10 MHz is illustrated in Figure 9. 10 MHz is illustrated in Figure 9.
(a)(a)
(c)
(b)(b)
(d)
Figure 9. Representative (d) body with and (c) electric field strength distribution in proximity to the human
Figure 9. Representative electric field strength distribution in proximity to the human body with and without the shield electrode, indicating that the electric field strength is enhanced when the shield without9.the shield electrode, indicating that the electric field strength is when the shield Figure Representative electric fieldview strength distribution in proximity toenhanced the view human bodyshield with and electrode is introduced. (a) Coronal without shield electrode. (b) Transverse without electrode is introduced. (a) Coronal view without shield electrode. (b) Transverse view without shield without the shield electrode, that the electric field strength enhanced when the shield electrode. (c) Coronal view indicating with shield electrode. (d) Transverse view withisshield electrode. electrode.is(c) Coronal view with shield (d) Transverse with shieldview electrode. electrode introduced. (a) Coronal viewelectrode. without shield electrode.view (b) Transverse without shield
electrode. (c) Coronal view with shield electrode. (d) Transverse view with shield electrode.
Sensors 2017, 17, 2056
9 of 14
Sensors 2017, 17, 2056
9 of 14
Figure without the the shield shield electrode, electrode, Figure 9a 9a shows shows the the coronal coronal view view of of the the field field strength strength distribution distribution without and Figure 9b exhibits its corresponding transverse view. Similarly, Figure 9c,d present the and Figure 9b exhibits its corresponding transverse view. Similarly, Figure 9c,d present the electric electric field with the the shield shield electrode electrode in in the the coronal coronal and and transverse transverse view view respectively. respectively. field strength strength distribution distribution with By the two By comparing comparing the two rows, rows, it it can can be be found found that that the the area area with with bright bright color color around around human human body body in in Figure 9c,d tend to be bigger than the one in Figure 9a,b, implying that the electric field strength Figure 9c,d tend to be bigger than the one in Figure 9a,b, implying that the electric field strength around around human human body body is is improved improved when when the the shield shield electrode electrode is is introduced introducedto tothe thetransmitter. transmitter. To further quantitatively investigate the effect of the shield electrode on the electric field field strength strength To further quantitatively investigate the effect of the shield electrode on the electric distribution different frequencies, (head), P2 P2 (arm), (arm), P3 P3 (chest) (chest) distribution with with respect respect to to different frequencies, four four typical typical positions positions P1 P1 (head), and P4 (leg) around human body, as denoted in Figure 7b, are chosen for comparison. The simulated and P4 (leg) around human body, as denoted in Figure 7b, are chosen for comparison. The simulated results in Figure 10, in10, which the normalized electric strength arevalues used forare comparison results are areshown shown in Figure in which the normalized electric values strength used for in condition of with and without shield electrode, and in frequency range from 100 kHz to comparison in condition of with and without shield electrode, and in frequency range from 50 100MHz. kHz The corresponding increased percentage of electric field strength value using shield electrode also to 50 MHz. The corresponding increased percentage of electric field strength value using is shield presented on the top of each figure. Figure 10a–d show the simulated electric field strength measured electrode is also presented on the top of each figure. Figures 10a–d show the simulated electric field at positions of P1, P2, P3 and P4,ofrespectively. It can be seen that all fourbe positions 70% strength measured at positions P1, P2, P3 and P4, respectively. It can seen thatachieve all fourabout positions increase of the electric field strength by using shield electrode. achieve about 70% increase of the electric field strength by using shield electrode.
(a)
(b)
(c)
(d)
Figure 10. 10.The The comparison of simulated field distribution strength distribution at 4 Figure comparison of simulated electricelectric field strength measured atmeasured 4 representative representative positions around the human body, with and without shield electrode. (a) Electric field positions around the human body, with and without shield electrode. (a) Electric field strength strength distribution at position P1 (head). (b) Electric field strength distribution at position P2 (arm). distribution at position P1 (head). (b) Electric field strength distribution at position P2 (arm). (c) Electric Electric field field strength strength distribution distribution at at position position P3 P3 (chest). (chest). (d) (d) Electric Electric field field strength strength distribution distribution at at (c) position P4 (leg). position P4 (leg).
When a receiver electrode is placed on the human body, an equivalent capacitor is established between the human body and the receiver electrode. Considering the realistic context using alternating signal as the excitation source on the transmitter electrodes, a displacement current flow
Sensors 2017, 17, 2056
10 of 14
When a receiver electrode is placed on the human body, an equivalent capacitor is established between the body and the receiver electrode. Considering the realistic context using alternating Sensors 2017, 17,human 2056 10 of 14 signal as the excitation source on the transmitter electrodes, a displacement current flow across the across the receiver electrodes. to the of the electromagnetic field,ofthe relation of receiver electrodes. According According to the rationale of rationale the electromagnetic field, the relation displacement displacement Id and electricinside field strength inside a capacitor can be written as: current Id andcurrent electric field strength a capacitor can be written as: dE AdE IdI = d ε 0 0A dt dt
(7) (7)
where is the area of the receiver electrode the electric thepermittivity, permittivity,A A where ε00 isisthe is the area of the receiver SIGSIG electrode andand E is Etheiselectric field.field. As a result, the displacement current is proportional to the fieldfield strength. Therefore, based on the As a result, the displacement current is proportional to electric the electric strength. Therefore, based on increase of the electric field strength adjacent transmission the increase of the electric field strength adjacenttotohuman humanbody, body,ititisisconcluded concluded that that the transmission gain of of the the IBC, IBC, in in terms terms of displacement current, is improved by using the proposed method. gain 4. Experimental Experimental Validation Validation 4. In order order to to further further verify verify the the assumption, assumption, an an experiment experiment is is performed performed using using aa transmitter transmitter device device In with and and without without the the shield shield electrode. electrode. with 4.1. Transmitter Transmitter Device Device 4.1. The transmitter transmitter is is aa battery-powered battery-powered module module composed cm ××33cm cmcopper copperGND GNDelectrode, electrode, The composed of of aa 33 cm cm ××33cm cm copper copper shield shield electrode, electrode, aa 33 cm cm ××33cm cmcopper copper SIG SIG electrode electrode and and an an excitation excitation signal signal aa 33 cm generator board, as shown in Figure 11. When conducting the experiment without the shield electrode, generator board, as shown in Figure 11. When conducting the experiment without the shield the middlethe board in Figure is removed. electrode, middle board11a in Figure 11a is removed.
(a)
(b)
(c)
(d)
Figure Figure 11. 11. Transmitter Transmitter device device and and receiver receiver electrode electrode used used in in the the experiment. experiment. (a) (a) The The transmitter transmitter with with the shield electrode. (b) The receiver electrode. (c) The signal source board. (d) Functional the shield electrode. (b) The receiver electrode. (c) The signal source board. (d) Functional block block diagram diagram of of the the transmitter transmitter signal signal source source board. board.
The signal generator board placed below the shield electrode is shown in Figure 11c. It consists of an AD9854 direct digital synthesizer (DDS), a STM32 micro-controller unit (MCU), an amplifying and filtering module based on AD8045, a shield driver module based on AD8045, and a button cell battery. The DDS is controlled by the MCU to provide a sine wave signal with programmable frequency ranging from 100 kHz to 50 MHz. The output sine wave is amplified and filtered, and then
Sensors 2017, 17, 2056
11 of 14
The signal generator board placed below the shield electrode is shown in Figure 11c. It consists of an AD9854 direct digital synthesizer (DDS), a STM32 micro-controller unit (MCU), an amplifying and filtering module based on AD8045, a shield driver module based on AD8045, and a button cell battery. Sensors 2017, 17, 2056 11 of 14 The DDS is controlled by the MCU to provide a sine wave signal with programmable frequency ranging from 100 kHz to 50 MHz. The output sine wave is amplified and filtered, and then connected connected to the SIG electrode and the shielding module simultaneously. The signal voltage applied to the SIG electrode and the shielding module simultaneously. The signal voltage applied between the between the SIG-GND electrodes is 5 V. The detailed functional block diagram of the signal generator SIG-GND electrodes is 5 V. The detailed functional block diagram of the signal generator is shown is shown in Figure 10d. The receiver electrode in vertical configuration, as shown in Figure 11b, is in Figure 10d. The receiver electrode in vertical configuration, as shown in Figure 11b, is selected to selected to pick up the signal transmitting from the source. It is made up of a SIG and a GND electrode pick up the signal transmitting from the source. It is made up of a SIG and a GND electrode with a with a size of 3 cm × 3 cm, spaced by 0.5 cm. In addition, a SMA port and a cable are used to connect size of 3 cm × 3 cm, spaced by 0.5 cm. In addition, a SMA port and a cable are used to connect to a to a measurement device. measurement device. 4.2. Measurement Setup 4.2. Measurement Setup The experimental equipment includes two types of battery-powered transmitters (with and Thethe experimental equipment includes two of battery-powered transmitters and without shield electrode): the receiver, and types a spectrum analyzer (Agilent N9030A,(with Agilent without the shield electrode): the USA). receiver, spectrum analyzer Agilent Technologies Inc, Santa Clara, CA, In and ordera to isolate the GND (Agilent electrode N9030A, and instrument Technologies Inc, Santa Clara, CA, USA). In order to isolate the GND electrode and instrument ground, an uninterruptible Power System (UPS, APC RS1000, APC By Schneider Electric,ground, West an uninterruptible Power System (UPS,up APC APC By Schneider Westheight Kingston, RI, Kingston, RI, USA) is adopted to power theRS1000, spectrum analyzer. A maleElectric, subject with of 175 USA) adopted up the for spectrum analyzer. A male subject with heightonofthe 175wrist, cm and weight cm andisweight of to 75power kg is chosen the experiment. The transmitter is placed and four of 75 kg is chosen forand the P4) experiment. Theparts transmitter is are placed on the wrist, the andreceiver four positions (P1, positions (P1, P2, P3 on different of body chosen to place electrode, P2, P3are andthen P4)connected on different of body analyzer, are chosen place in theFigure receiver which are then which to parts the spectrum asto shown 12a.electrode, In order to evaluate the connected to the spectrum analyzer, as shown in Figure 12a. In order to evaluate the total power total power consumption, an ammeter (model WR5145-PR-5V, BBYE Inc., Guangzhou, China) is consumption, an ammeter (model Inc., Guangzhou, China)inis Figure inserted12b. between inserted between the positive portWR5145-PR-5V, of battery andBBYE the transmitter, as denoted The the positive of battery andrange the transmitter, denoted in Figure 12b. The current ammeter hasport a current input from 0 to as 200 mA, with resolution of ammeter 0.01 mA. has Thea power input range from 0 to 200 mA, with resolution ofof0.01 The power consumption is voltage calculated by consumption is calculated by the multiplication themA. measured current and battery read the multiplication from a multimeter. of the measured current and battery voltage read from a multimeter.
(a)
(b)
Figure Figure12. 12.Experiment Experimentsetup. setup.(a) (a)Setup Setuptotomeasure measurethe thereceived receivedpower; power;(b) (b)Setup Setuptotomeasure measurethe theDC DC current supply of the transmitter. current supply of the transmitter.
4.3. Results and Discussion 4.3. Results and Discussion The measured results are illustrated in Figure 13, in which Figures 13a–d present the comparison The measured results are illustrated in Figure 13, in which Figure 13a–d present the comparison of the receiving power at four representative positions of P1 (head), P2 (arm), P3 (chest) and P4 (leg), of the receiving power at four representative positions of P1 (head), P2 (arm), P3 (chest) and P4 (leg), respectively. The results are measured at the typical frequency points including 100, 200, 500 kHz, 1, respectively. The results are measured at the typical frequency points including 100, 200, 500 kHz, 1, 2, 2, 5, 10, 20, 30 and 50 MHz, for both cases of with and without shield electrode on the transmitter. 5, 10, 20, 30 and 50 MHz, for both cases of with and without shield electrode on the transmitter. It can be found that the receiving power measured using the transmitter with shield electrode is approximately 5.5 dBm larger than the one without the shield electrode, in majority portion of the frequency range. With such observations, the measured results at the 4 positions are consistent with the simulation results, suggesting that the shield electrode between the SIG-GND electrodes of the transmitter could be an effective solution to improve the capacitive IBC transmission gain.
Sensors 2017, 17, 2056
12 of 14
Sensors 2017, 17, 2056
12 of 14
Sensors 2017, 17, 2056
12 of 14
(a)
(a)
(c)
(b)
(b)
(d)
Figure13. 13.Received Receivedpower powercomparison comparisonbetween betweenthe thetwo twotransmitter transmitterconfigurations configurationswith withand andwithout without Figure the shield electrode in frequency range from 100 kHz to 50 MHz, at 4 representative positions around the shield electrode in frequency range from 100 kHz to 50 MHz, at 4 representative positions around human body. (a) P1 (head), (b) P2 (arm), (c) P3 (chest), (d) P4 (leg). human body. (a) P1 (head), (b) P2 (arm), (c) P3 (chest), (d) P4 (leg).
(d) (c) The average power saving at above representative situations is obtained by subtracting the It can be found that the receiving power measured using the transmitter with shield electrode 13. Received power comparison the two transmitter with and without powerFigure consumption with shield electrodebetween from that without shieldconfigurations electrode. The measured power is approximately 5.5 dBm larger than the one without the shield electrode, in majority portion of the thein shield electrode ranging in frequency range kHz to 50are MHz, at 4 representative positions around savings frequencies from 100 from kHz 100 to 50 MHz present in Figure 14, showing that the frequency range. With such observations, the measured results at the 4 positions are consistent with body.can (a) P1 (head),the (b) total P2 (arm), (c) P3consumption (chest), (d) P4 especially (leg). shieldhuman electrode reduce power in higher frequency region. the simulation results, suggesting that the shield electrode between the SIG-GND electrodes of the Meanwhile, the transmission gain is improved as well by introducing the proposed active shielding transmitter could bepower an effective to improve the capacitive IBCistransmission The average savingsolution at above representative situations obtained bygain. subtracting the method. The average power saving at above representative situations is obtained by subtracting the power power consumption with shield electrode from that without shield electrode. The measured power consumption with shield electrode from that without shield electrode. The measured power savings in frequencies ranging from 100 kHz to 50 MHz are present in Figure 14, showingsavings that the inshield frequencies ranging from 100 50 MHz are present especially in Figure 14, that the shield electrode can reduce thekHz totaltopower consumption in showing higher frequency region. electrode can reduce the total power consumption especially in higher frequency region. Meanwhile, Meanwhile, the transmission gain is improved as well by introducing the proposed active shielding the transmission gain is improved as well by introducing the proposed active shielding method. method.
Figure 14. The measured average power saving in frequency range from 100 kHz to 50 MHz when using the shield electrode.
5. Conclusions In this paper, a method of electric active shielding for transmitter electrode is proposed, Figure 14. The measured average power saving in frequency range from 100 kHz to 50 MHz when attempting to improve the power of the coupling Thetoproposed method is Figure measured averageefficiency power saving in capacitive frequency range from IBC. 100 kHz 50 MHz when using 14. the The shield electrode. firstly investigated by means of circuit analysis, using the conventional model by Zimmerman. The using the shield electrode. power consumption caused by the internal displacement current between the SIG-GND electrodes 5. Conclusions of the transmitter is evaluated and the fundamental theory of the shield electrode is explained In the thiscircuit paper,model. a method of electric shielding for perspective transmitterofelectrode is proposed, through Secondly, furtheractive validation from the electric field strength attempting to improve the power efficiency of the capacitive coupling IBC. The proposed method is firstly investigated by means of circuit analysis, using the conventional model by Zimmerman. The power consumption caused by the internal displacement current between the SIG-GND electrodes
Sensors 2017, 17, 2056
13 of 14
5. Conclusions In this paper, a method of electric active shielding for transmitter electrode is proposed, attempting to improve the power efficiency of the capacitive coupling IBC. The proposed method is firstly investigated by means of circuit analysis, using the conventional model by Zimmerman. The power consumption caused by the internal displacement current between the SIG-GND electrodes of the transmitter is evaluated and the fundamental theory of the shield electrode is explained through the circuit model. Secondly, further validation from the perspective of electric field strength distribution is conducted using ANSYS Maxwell FEM simulation software. Finally, the effect of shield electrode on the transmission gain and total power consumption is verified by experiment using battery-power transmitter together with spectrum analyzer. The simulations are implemented and compared in frequency range from 100 kHz to 50 MHz, with experiments on four representative positions around human body and in two conditions of transmitter with and without shield electrode. As a result, the simulation shows that the electric field strength in proximity to human body is improved by approximately 70% using the shield electrode. Similarly, the received power in experiment is increased by 5.5 dBm while total power consumption is reduced by 9 mW when using the shield electrode than the case without it. Both the simulation and experiment result indicate that applying shield electrode between the SIG-GND electrodes of the transmitter is a viable solution to improve the energy efficiency for capacitive coupling IBC. The proposed method can provide an energy efficient option in physical layer for wearable and implantable healthcare sensor networks. Author Contributions: Chao Ma designed and performed the simulations and experiments, and also contributed to the data collection, analysis and writing of the corresponding paragraphs. Zhonghua Huang provided many useful comments and constructive discussions. Zhiqi Wang worked on the measurements. Linxuan Zhou proofread the manuscript and improved the style of the writing and grammar. Yinlin Li conceived of the idea of the whole study, and contributed to the drafting and implementing of the manuscript. All authors read and approved the final manuscript. Conflicts of Interest: The authors declare no conflict of interest.
References 1. 2. 3. 4.
5.
6.
7.
8. 9.
Bae, J.; Yoo, H.J. The Effects of Electrode Configuration on Body Channel Communication Based on Analysis of Vertical and Horizontal Electric Dipoles. IEEE Trans. Microwave Theory Tech. 2015, 63, 1409–1420. [CrossRef] Cho, N.; Yan, L.; Bae, J.; Yoo, H.-J. A 60 kb/s-10 Mb/s Adaptive Frequency Hopping Transceiver for Interference-Resilient Body Channel Communication. IEEE J. Solid-State Circuits 2009, 44, 708–717. [CrossRef] Callejon, M.A.; Naranjo-Hernandez, D.; Reina-Tosina, J.; Roa, L.M. A Comprehensive Study Into Intrabody Communication Measurements. IEEE Trans. Instrum. Meas. 2013, 62, 2446–2455. [CrossRef] Song, Y.; Zhang, K.; Hao, Q.; Hu, L.X.; Wang, J.W.; Shang, F.Z. A Finite-Element Simulation of Galvanic Coupling Intra-Body Communication Based on the Whole Human Body. Sensors 2012, 12, 13567–13582. [CrossRef] [PubMed] Gao, Y.M.; Wu, Z.M.; Pun, S.H.; Mak, P.U.; Vai, M.I.; Du, M. A Novel Field-Circuit FEM Modeling and Channel Gain Estimation for Galvanic Coupling Real IBC Measurements. Sensors 2016, 16, 471. [CrossRef] [PubMed] Nguyen, X.T.; Muramatsu, D.; Sasaki, K. Measurement of Human Body Communication Transmission Characteristics at 20 MHz. In Proceedings of the 2015 9th International Symposium on Medical Information and Communication Technology (Ismict), Kamakura, Japan, 24–26 March 2015. Callejon, M.A.; Reina-Tosina, J.; Naranjo-Hernandez, D.; Roa, L.M. Galvanic Coupling Transmission in Intrabody Communication: A Finite Element Approach. IEEE Trans. Biomed. Eng. 2014, 61, 775–783. [CrossRef] [PubMed] Xu, R.; Zhu, H.; Yuan, J. Electric-field Intrabody Communication Channel Modeling with Finite-element Method. IEEE Trans. Biomed. Eng. 2011, 58, 705–712. [PubMed] Cho, N.; Yoo, J.; Song, S.-J.; Lee, J.; Jeon, S.; Yoo, H.-J. The Human Body Characteristics as a Signal Transmission Medium for Intrabody Communication. IEEE Trans. Microwave Theory Tech. 2007, 55, 1080–1086. [CrossRef]
Sensors 2017, 17, 2056
10. 11. 12. 13. 14. 15.
16.
17.
18.
19.
20. 21. 22.
23.
24. 25.
14 of 14
Nie, Z.D.; Ma, J.J.; Li, Z.C.; Chen, H.; Wang, L. Dynamic Propagation Channel Characterization and Modeling for Human Body Communication. Sensors 2012, 12, 17569–17587. [CrossRef] [PubMed] Zhu, X.Q.; Guo, Y.X.; Wu, W. Investigation and Modeling of Capacitive Human Body Communication. IEEE Trans. Biomed. Circuits Syst. 2017, 11, 474–482. [CrossRef] [PubMed] Park, J.; Garudadri, H.; Mercier, P.P. Channel Modeling of Miniaturized Battery-Powered Capacitive Human Body Communication Systems. IEEE Trans. Biomed. Eng. 2017, 64, 452–462. [CrossRef] [PubMed] Pereira, M.D.; Alvarez-Botero, G.A.; Rangel de Sousa, F. Characterization and Modeling of the Capacitive HBC Channel. IEEE Trans. Instrum. Meas. 2015, 64, 2626–2635. [CrossRef] Lucev, Ž.; Krois, I.; Cifrek, M. A Capacitive Intrabody Communication Channel from 100 kHz to 100 MHz. IEEE Trans. Instrum. Meas. 2012, 61, 3280–3289. [CrossRef] Zhang, K.; Hao, Q.; Song, Y.; Wang, J.W.; Huang, R.B.; Liu, Y. Modeling and Characterization of the Implant Intra-Body Communication Based on Capacitive Coupling Using a Transfer Function Method. Sensors 2014, 14, 1740–1756. [CrossRef] [PubMed] Chen, X.M.; Mak, P.U.; Pun, S.H.; Gao, Y.M.; Lam, C.T.; Vai, M.I.; Du, M. Study of Channel Characteristics for Galvanic-Type Intra-Body Communication Based on a Transfer Function from a Quasi-Static Field Model. Sensors 2012, 12, 16433–16450. [CrossRef] [PubMed] Wegmueller, M.; Felber, N.; Fichtner, W.; Lehner, A.; Hess, O.; Froehlich, J.; Kuster, N.; Reutemann, R.; Oberle, M. Measurement system for the characterization of the human body as a communication channel at low frequency. In Proceedings of the 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference, Shanghai, China, 1–4 September 2005; pp. 3502–3505. Okamoto, E.; Kato, Y.; Kikuchi, S.; Mitamura, Y. Measurement of Electrode-tissue Interface Impedance for Improvement of a Transcutaneous Data Transmission Using Human Body as Transmission Medium. Bio-med. Mater. Eng. 2014, 24, 1735–1742. Harikumar, P.; Kazim, M.I.; Wikner, J.J. An Analog Receiver Front-End for Capacitive Body-Coupled Communication. In Proceedings of the NORCHIP 2012, Cpenhagen, Denmark, 12–13 November 2012; pp. 1–4. Li, J.; Liu, Y.; Nie, Z.; Qin, W.; Pang, Z.; Wang, L. An Approach to Biometric Verification Based on Human Body Communication in Wearable Devices. Sensors 2017, 17, 125. [CrossRef] [PubMed] Zimmerman, T.G. Personal area networks: Near-field Intrabody Communication. IBM Syst. J. 1996, 35, 609–617. [CrossRef] Xu, R.; Ng, W.C.; Zhu, H.; Shan, H.; Yuan, J. Equation Environment Coupling and Interference on the Electric-field Intrabody Communication Channel. IEEE Trans. Biomed. Eng. 2012, 59, 2051–2059. [CrossRef] [PubMed] Haga, N.; Motojima, K.; Shinagawa, M.; Kado, Y. Equivalent-Circuit Expression of Environmental Noise Electric Fields in Intrabody Communication Channels. IEEE Trans. Electromagn. Compat. 2016, 58, 294–306. [CrossRef] Bae, J.; Cho, H.; Song, K.; Lee, H.; Yoo, H.-J. The Signal Transmission Mechanism on the Surface of Human Body for Body Channel Communication. IEEE Trans. Microwave Theory Tech. 2012, 60, 582–593. [CrossRef] Xie, Y.; Li, Y.L.; Huang, Z.H.; Cui, Z.Z. Preliminary Study of Rain Effect on Channel Gain of Body Area Communication. In Proceedings of the 2009 Joint Conferences on Pervasive Computing (JCPC), Tamsui, Taipei, Taiwan, 3–5 December 2009; pp. 657–660. © 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
