LC Oscillator As An Ultra Simple, Low Power Transmitter For Wireless Sensors

Size: px

Start display at page:

Download "LC Oscillator As An Ultra Simple, Low Power Transmitter For Wireless Sensors"

Arthur Stephens
5 years ago
Views:

1 LC Oscillator As An Ultra Simple, Low Power Transmitter For Wireless Sensors Christos V. Ilioudis Dept. of Informatics and Telecommunications Engineering University of Western Macedonia (UWM) Kozani, Greece Michail E. Kiziroglou, IEEE Member Dept. of Informatics and Telecommunications Engineering University of Western Macedonia (UWM) Kozani, Greece Abstract In this paper a new communication platform is presented for biomedical MEMS sensor network applications. The communication method is based on a very simple LC oscillator that emits a low power signal from a loop antenna. The capacitor C is charged to a voltage representing sensor data. When the LC branch is short-wired by a mechanical switch, oscillations occur, resulting in an electromagnetic pulse which can be detected in the vicinity of the system. The system parameter anlysis of the communication platform is carried out throught experiments and simulation using Matlab and Spice utilities. Experimental results demonstrate successful transmission and detection of the LC oscillations at a distance of 20 cm. As expected, the power of the detected oscillation scales with the capacitor voltage, showing that information transmision is possible by the proposed technique. Keywords-LC Oscillator; Low Power Transmitter; Wireless Sensors. I. INTRODUCTION Wireless sensor networks (WSNs) may improve and offer new possibilities in critical areas of life quality such as patient health care, safety of buildings and food hygiene standards monitoring. One of the main challenges for the deployment of wireless biometric sensors is the requirement for integration of various systems in a very small device, the biosensor, where the transmission of vital information should be performed with minimal power consumption (nj/byte). Such systems architecture is characterized by the design of separate subsystems, which are selected to perform specific functions. Especially, a biosensor operation includes detecting a physical quantity converted into electrical signal (usually dc voltage, 0-2V), strengthening and emitting this information after proper modulation. Usually, to enhance the signal a suitable power source (battery or energy harvester) is used, while the broadcast is implemented by means of a specifically designed transmitter LC. Powering sensor nodes is a major challenge for such applications, as batteries are bulky and require recharging, which is not practical in cases involving remote locations or large numbers of sensors. For this reason, a lot of research has focused on low power electronics for sensors, while novel electrical power sources such as energy harvesting devices have been developed. The energy required for biosensor operation can be collected from sunlight, heat (temperature difference), motion, electromagnetic radiation (RF), or from a combination of them depending on the sensor environment. Additionally, if power consumption is extremely low, the biosensor can operate autonomously (without battery) using only the power provided by the energy harvester. In this case, the collected energy is stored temporarily (e.g. in a supercapacitor), while the biosensor is inactive. Then the stored energy can be used during data transmittion for a short-time operation of the sensor [1], [2]. There is a growing research interest to implement wireless biosensor technology based on a combination of both low-power wireless transmission techniques and energy collection methods [3], [5]. The design of circuits to support very low-energy emissions (Ultra Low Energy Transmissions) is crucial for autonomous operation of biosensors because of their extremely limited available power. Last decade several methods have been suggested to design systems capable of wireless data transmission using ultra low energy smaller than 1nJ/byte [8]. While highly efficient low power radio systems with transmission bit rates in the range of a few kbps have already been proposed, there are applications where the required data collection rate is much lower, even below 1bps. For this reason, WSN implementations typically involve programming nodes to spend most time in sleep mode, at least for the transceiver, waking up only to quickly send/receive information in a high bit rate and go back to sleep. This approach is very effective but it is ultimately limited by sleepmode consumption, which is substantial, especially for cases where long-term operation is required. Recently, a very simple transmission platform for wireless sensors has been proposed involving direct discharge of a sensor/harvester output into a loop antenna [1]. The characteristics of the corresponding electromagnetic pulse, which can be detected locally, represent sensor information. In this work, the characteristics of such LC oscillation transmitters are studied. The frequency, amplitude and quality factor of oscillations are presented for different C values and loop antenna geometrical characteristics. The electromagnetic pulses are detected at a distance up to 1m from the oscillator and transmission of information is demonstrated. These results show that local transmission of information with energy rate of 1nJ/byte may be possible for low bit rate applications.

2 II. SYSTEM DESCRIPTION The sensor described in the following paragraphs is a fairly simplified model of an energy collector that transmits a signal via a transmitter antenna. Specifically, the circuit consists of two main parts: collector of energy (energy harvester) and emitter (transmitter). The energy harvester subsystem is responsible for collecting and supplying the sensor with required electrical energy, while transmitter provides sending data to remote receivers. Also a measuring instrument (oscilloscope) has been added to the overall circuit. This addition allows to monitor the operation of the circuit in an experimental setting. Rs Vs Vantenna 1 Sw/of f Sw/on La Vs Ch 2 Ra Co 13p Ro 1M 0 A. Energy Harvesting The operation of an energy collector at a complete standalone sensor focuses on collecting and supplying the necessary energy for the operation of the sensor. In present work, it is supposed that the total energy collected is available for sending a data signal from the sensor to a local receiver located in a relatively short distance (up to 1m). The energy collector preferred for this application is for an motion energy harvester. According to its operation principle, it is known that the amount of energy, which can be collected by such a device and eventually exported, depends on the relative motion received from its environment. Here the energy collector is simulated with a small capacitor, while the excitation is done by a voltage source in series with a resistor (see Figure 1). When the capacitor is fully charged, it is isolated from the source by means of a switch ending the phase of energy harvesting. In the simulation, it is supposed that the capacitor is charged instantly and therefore the duration of charging depended on the time constant of the circuit τ does not affect the final result. B. Signal Transmission As mentioned before, a very important part of the sensor is the data signal transmitter. Due to the limited size of the sensor device and very small amount of the available energy, the information signal is transmitted using a small loop antenna. Additionally loop antennas are preferable, because of their simplicity both in construction and operation. Figure 1 shows the equivalent circuit of the antenna consisting of a resistor and an inductor connected in series, which represent the resistance R a and inductance L a of the antenna respectively [6]. This circuit is connected in series with the energy storage element (capacitor) through a switch. Once the capacitor is charged, then it is disconnected from the voltage source setting the switch to off position. At the same time the charged capacitor is connected to the antenna discharging and emitting electromagnetic radiation. The values both of R a and L a depends on the geometric characteristics of th eantena such as the radius of the circular loop and the cross-section of the conductor. The equivalent circuit of the oscilloscope consists of a capacitor and a resistor in series. Both values of C o and L o depends on the specific oscilloscope used for the measurements. It is important to consider that the overall behavior of the circuit is affected adding the oscilloscope capacity, since its capacity of 13pF is comparable to the used storage capacity of 1pF, 10pF and 100pF. Figure 1. Frequency vs. antenna radius for different capacities C calculated from equation (1). III. SIMULATIONS In the analysis of innovative communication platform were studied the emission waveform, the effect of epidermal phenomenon in communication system, the radiation resistance, the quality factor Q and the signal strength. The theoretical measurements and graphs were made using MATLAB and SPICE programs. A. Frequency of Antena in the RLC Oscillator Circuit. As mentioned in the previous section, the signal is sent through a loop antenna fed by a capacitor. The oscillation in such simple LC circuit is harmonic with frequency given by [10]: f = 1 2 π LC (1) where the inductance of the antenna L is given by [6]: ( 8α ) L = µ oα ln 2 b Here µ ο is the permeability of free space, µ ο =4π10-7 Vs/(Am), α and b are the the radii of the loop and its cross section respectively. It can be seen from (2) that with increasing the radius of the loop antenna α, its inductance L is increasing as well. As it is shown in Figure 2, with increasing values of capacity C and radius of the loop α, the emission frequency of the circuit decreases continuously. Also the change in emission frequency is very small for small variations of the cross section of the antenna. B. Quality Factor Q for the RLC Circuit The general definition of the quality factor is based on the ratio of apparent power to the power losses in a device. In electrical circuits, the quality factor Q or Q factor is a dimensionless parameter, which indicates how under-damped an oscillator or a resonator is and it is defined by following mathematical expression Es Q = 2π (3) Ed (2)

3 where E s is the stored energy and E d is the dissipated energy per cycle. High quality factor Q indicates a lower rate of energy loss compared to the stored energy of the oscillator and therefore the oscillations last longer. Especially in a RLC circuit, the factor Q is expressed by the relation [12]: ωl Q = (4) R Here ω is the angular frequency, ω=2πf. The quality factor Q in (4) is dependent on the shape, size of the coil, used materials and operating frequency. The resistance calculation of a transmitting antenna is a bit more complex, when one considers that the total resistance of the loop antenna depends not only on the loop radius, the cross section and the transmission frequency, but also on the antena resistance variation due to the so called skin effect phenomenon. In this case the total resistance is depended on angular frequency and an approximation of R is given by ρl R π ( D δ ) δ where δ represents the skin depth defined by 2ρ δ = (6) ωµ Here ρ is the resistivity, l is the length, D is the cross-section area and µ is the permeability of the antenna conductor. Therefore it is possible to estimate the quality factor Q for the various communication platforms, i.e. for different values of L and C vs. frequency. In Figure 3 it is shown that lower capacity C deliver higher power factor Q. This is natural, since the smaller the capacitance C, the greater the oscillation frequency of the RLC circuit f, as it results from (1). Additionally considering (4), it could be seen that increasing the angular velocity of the circuit ω (ω=2πf) leads to an increase of Q as well. Although the choice of a large capacity C in an oscillator causes reduction of Q, however the discharge duration is increasing. (5) Figure 3. Quality factor Q vs. antenna radius for different capacities C as calculated from (4) using (5) and (6). IV. EXPERIMENTAL DETAILS In the simulations and experiments of communication platform (see Figure 1) the transmitter variables are selected among a range in order to study the system behaviour in more details. The main variables of the elements involved in the implementation of this communication platform are: the voltage V applied to the capacitor (power supply voltage), the capacity C, the radius of loop antenna α and the radius of cross section b. The charge voltage of 5V is selected, since it is a common low voltage found in energy harvesting systems of stand-alone wireless sensors. Capacities of 1pF, 10pF, 100pF, 1nF, 10nF and 100nF are selected, because they are characterised by limited energy storage capability as in the case of energy harvesters as well. A digital oscilloscope (model HP / Agilent DSO3202A) is used at experimental measurements with a maximum sampling frequency of 200MHz. The different type of loop antennas constructed with loop radii 0.4cm, 1.2cm and 2cm and consists of a simple single-stranded copper wire with radius of cross section 0.5mm and 0.75mm (see Figure 4). The construction process is simple first removing the outer protective casing and then bending the wire circle. Although the perfect circle formation of a given radius is quite difficult to be achieved, however the deviations were quite small. (a) (b) Figure 2. Frequency vs. antenna radius for different capacities C calculated from (1) Figure 4. Loop antennae pictures for both radii of the conductor crosssection 0.5mm (a) and 0.75mm (b).

4 V. RESULTS AND DISCUSSION The presented method was tested and verified using simulations and experiments considering both for several capacities of the energy storage element and for different geometric parameters of the loop antenna. Figure 7. Waveforms of the signal at transmitter and receiver for capacitor discharge voltage of 8V. Figure 5. Signal waveform amplitude at transmitter and its fast fourier transformation. Figure 8. Mean power received vs. distance for four discharge voltages. Figure 6. Approximation of inductance L with cross section radius 0.5mm. TABLE I. THEOTETICAL CALCULATIONS AND EXPERIMENTAL APPROXIMATIONS OF LOOP ANTENNA INDUCTANCE L. Radius of loop [cm] Radius of cross section [mm] Theoretical measurements Inductance [nh] Experimental results (±5nH) Experimental and simulation results are shown in Figure 5 considering the amplitude of transmitted signal waveforms and their quality factor Q vs. frequency. Comparing both signal waveforms, it is concluded that they have the same frequency as expected. However, the Q factor resulting from the experiments appears to be significantly smaller than the theoretical one, since the amplitude of the real signal is reduced faster. Figure 6 shows an approximation of antenna inductance L obtained experimentally for three different radii of loop. A comparison of the transmitted and received signal is shown in Figure 7 setting the capacitor discharge voltage at 8V. The mean power of the received signal is shown in Figure 8 vs. distance from the transmitter for different discharge voltages. As seen in the graph the rate of mean power reduction becomes greater as the discharge voltage increases Table I lists the theoretical and experimental values of the inductance L for the tested values of α and b (i.e. radii of antenna loop and cross-section respectively). Comparing the values of inductances, it results that the experiments give

5 values for L, which are in good agreement with the theoretically expected ones, especially at relatively large values of α. It should be noted that the differences presented in the antenna inductances between experimental and theoretical calculations there are because of the very simple construction manner of loop antenna and they are more evident at small values of α. VI. CONCLUSIONS A new developed communication platform is tested using a simple design for wireless sensor transmitter and local receiver subsystems. The aim of the presented work is to explore whether it is possible to directly send data through electromagnetic pulses of extremely low energy of 3.2µJ. Although initial experiments have been implemented for small distances (up to 20cm), the results of signal reception were successful and very encouraging. According to the results of these experiments, the realization of such a communication platform is feasible and promising as it could be subject to many improvements. The reason is that many of the factors adversely affecting the results are traceable and are mainly related to the construction method (transmit and receive circuitry) and operating environment (e.g. electromagnetic interference). In practice the construction and measument defects are mainly due to the usage of push-button switches, cables for hanging accessories and instrument (oscilloscope) limited by a sampling frequency of 200MHz. Signal transmitter and receiver circuits can be greatly improved so that the signal carrying the information can cover greater distances without electromagnetic interference. In addition, appplying filtering techniques the information of a transmitted signal could be almost perfectly recovered from measurements by rejecting electromagnetic noise components. Therefore the studied method of sending data is applicable to key areas of WSNs research. REFERENCES [1] Cairan He, Michail E. Kiziroglou, David C. Yates and Eric M. Yeatman, MEMS Energy Harvester for Wireless Biosensors, Proc. 23rd IEEE Conference in Micro-electro-mechanical Systems (MEMS 2010), pp , Hong Kong, Jan , [2] Μ. Ε. Kiziroglou, C. He and E. M. Yeatman, "Rolling rod electrostatic microgenerator," IEEE Transactions on Industrial Electronics, vol. 56, no. 4, pp , April [3] M. Gorlatova, P. Kinget, I. Kymissis, D. Rubenstein, X. Wang and G. Zussman, "Energy harvesting active networked tags (EnHANTs) for ubiquitous object networking," IEEE Trans. Wireless Commun, vol. 17, no. 6, pp , Dec [4] E. M. Yeatman and M. Ε. Kiziroglou, "Functional Materials for Energy Applications," in Chapter IV.3: Energy Harvesting, Imperial College London, [5] P. D. Mitcheson, E. M. Yeatman, G. Kondala Rao, A. S. Holmes and T. C. Green, "Energy Harvesting From Human and MachineMotion for Wireless Electronic Devices," IEEE, [6] C. C. Enz, A. El-Hoiydi, J.-. D. Decotignie and V. Peiris, "WiseNET: An Ultralow-Power Wireless Sensor Network Solution," IEEE Computer Society, pp , Aug [7] D. C. Yates, A. S. Holmes and A. J. Burdett, "Optimal transmission frequency for ultralow-power short-range radio links," IEEE Trans. on Circuits and Systems, vol. 51, pp , [8] N. Panitantum, "Ultra-Low-Energy Transmitters for Battery-Free Wireless Sensor Networks," Oregon, USA., 13 June [9] J. Ayers, K. Mayaram and T. Fiez, "An Ultralow-Power Receiver for Wireless Sensor Networks," IEEE Journal of Solid-State Circuits, vol. 45, pp , Sep [10] "IEEE Standard Definitions of Terms for Antennas," in IEEE Std , [11] H. A. Wheeler, "Formulas for the Skin Effect," Proceedings of the IRE, [12] N. I. Margaris, Analysis of Electrical Circuits,, Tziolas Editions, Thessaloniki, Greece (in Greek). [13] H. G. Schantz, "A Near Field Propagation Law & A Novel Fundamental Limit to AntennaGain Versus Size," IEEE APS Conference, [14] C. E. Shannon, "Communication in the presence of noise," Proc. Institute of Radio Engineers, 1949, pp [15] J. Verschelde, "Notes on Signal Processing in MATLAB," in Introduction to Symbolic Computation, MCS 320, Dept of Math, Stat & SC, UIC, Spring [16] B. Ninness, "Spectral Analysis using the FFT," Department of Electrical and Computer Engineering, The University of Newcastle, Australia. [17] K. Moody, "Electromagnetic Noise Monitoring, Ministry of Commerce, Wellington, [18] C. A. Balanis, Antennas: An;alysis and Design, 2005.

Motivation. Approach. Requirements. Optimal Transmission Frequency for Ultra-Low Power Short-Range Medical Telemetry

Motivation. Approach. Requirements. Optimal Transmission Frequency for Ultra-Low Power Short-Range Medical Telemetry Motivation Optimal Transmission Frequency for Ultra-Low Power Short-Range Medical Telemetry Develop wireless medical telemetry to allow unobtrusive health monitoring Patients can be conveniently monitored