Feature Articles: Circuit and Device Technologies for Extremely Low-power Consumption for Future Communications Systems

Nanowatt Circuit Technology for an Ultrasmall Wireless Sensor Node with Energy Harvester

Shoichi Oshima, Kenichi Matsunaga, Toshishige Shimamura, Hiroki Morimura,
and Mitsuru Harada


This article describes an ultrasmall wireless sensor node (WSN) that is powered by an energy harvester. The small WSNs can be mounted on any kind of object without increasing their volume, and the energy harvester provides power without having to worry about battery life. However, it is difficult to achieve a WSN that is both ultrasmall and yet still has sufficient power; the smaller the energy harvester is, the less power it generates. We have developed two nanowatt circuit techniques to address this issue. One is a sensing circuit that converts signals with vibration energy. The other is a wireless circuit that uses a quick startup ring oscillator and pulse modulation with an intermittent-driving radio frequency circuit.


1. Introduction

Wireless sensor networks that can gather and analyze trivial information such as vibration, temperature, angles, or positions will lead to the development of new services [1]–[4]. For example, a current sensor attached to electrical appliances in a house can detect how much electricity is used or saved over time, and a sensor that can detect patterns in vibrations caused by wind can indicate cracks in concrete structures. In addition, wireless sensor networks make it possible to provide machine-to-machine (M2M) networks where appliances operate voluntarily without human instructions. An image of a wireless sensor network is shown in Fig. 1. The network consists of wireless sensor nodes (WSNs) that can be mounted on any object or living thing to monitor environmental or physical conditions, on master receivers that are connected to the network to collect information from the WSNs, and on servers to store and analyze data.

Fig. 1. Image of the wireless sensor network.

Our concept is to make the WSNs ultrasmall and to drive them with an energy harvester. The small WSNs can be mounted on any object without increasing the object’s volume. The energy harvester draws ambient energy from the surrounding environment. This means that the conventional battery is unnecessary as a power source, and consequently, there are no concerns about battery life. These two aspects also reduce the cost of the WSNs and make them less intrusive, which allows a greater range of applications. Unfortunately, there is a conflict between energy harvesting and node miniaturization; as the energy generator becomes smaller, so does the generated power. In this article, we describe ultralow power, nanowatt circuit technology that we developed to address this problem.

This article is organized as follows. Section 2 describes the energy harvester and nanowatt circuit technology. Section 3 covers the components and circuits of the WSN, and section 4 concludes the article.

2. Nanowatt circuit technology

The minimum configuration of the WSN energy harvester, sensor, and wireless integrated circuit (IC) is shown in Fig. 2(a). We first discuss the energy harvester and then explain the sensing circuit and the wireless configuration.

Fig. 2. (a) Minimum WSN configuration and (b) amount of power generated by various sizes of energy harvesters.

2.1 Energy harvester

The size of the energy harvester versus the energy generated from ambient heat, vibration, and light sources is shown in Fig. 2(b) [5]. The shape of the solar battery is considered as a square, and the others as cubes. Naturally, more energy degradation occurs as the size of the harvester decreases. The energy source that is selected therefore depends on the usage conditions and energy requirements.

For example, there is a good source of heat when WSNs are mounted on human skin or on a large-scale integrated circuit (LSI) of a computer. However, the energy conversion efficiency is low in this case, and a heat waste structure is needed to convert energy.

Vibration is a good energy source for WSNs mounted on humans or vehicles, but space is needed for the vibration generator. Further, the vibration frequency depends on the shape and the material of the generator, which limits the flexibility for modeling the generator.

Light is a good energy source only when the light is irradiated. The generated power is affected by the light source, and the intensity of light in different environments, for example, a dark wine cellar, and a sandy beach on a sunny day, can vary dramatically.

Whichever energy source is chosen, the generated power for a WSN less than 1 cm in size under stable conditions, such as when there is no one in a room, is at most 1 µW. To drive all blocks of the WSN with such low energy, nanowatt circuit technology is required.

2.2 Sensing circuit technology [6]

Here we introduce a vibration sensing circuit with nanowatt power consumption. The vibration structure is usually a passive device that consumes little power. However, the conversion of the vibration signal to an electrical signal is the most important point.

The configuration of the vibration sensing circuit is shown in Fig. 3. A movable component made of metal forms a capacitor that is connected to the power source. When this component vibrates, its capacitance vibration pattern changes. This results in current being pulled from the power source, which is the same as what happens in a charge pump operation. Thus, the vibration signal is changed into an electrical signal through the use of vibration energy rather than electrical energy, which consumes no electrical power. The measured power consumption of this circuit is only 0.7 nW.

Fig. 3. Vibration conversion circuit with movable metal component.

2.3 Wireless IC technology [7]

In this subsection we introduce an ultralow power wireless transmitter that uses intermittent pulse train modulation.

Let us start by first examining a conventional wireless system. The power consumption over time for one-time data transmission is depicted in Fig. 4(a). To transmit data, first, the wireless circuits turn on. However, some startup time is necessary before data can be transmitted. The main operations that make up the startup time are oscillator stabilization and IC reset and initialization. These operations require about 10 ms when the system uses a crystal oscillator to lock the phase-lock loop circuit. The wireless system then starts to transmit the data. During the data transmission, the power consumption consists of the power for driving the circuits and the power transmitted to the air. In theory, only the power transmitted to the air is required for wireless communications. As little power as possible should be used for oscillator stabilization and for driving the circuits.

Fig. 4. Power consumed to transmit data from startup to end of data block.
(a) Conventional system, (b) proposed system, and (c) total energy.

Our wireless system is shown in Fig. 4(b). We use two key techniques to reduce the total power consumption. One is a quick startup circuit to cut off the energy loss during the stabilization time. The other is the use of a 1-symbol intermittent pulse train instead of a continuous wave, so that when the pulse stops the radio frequency (RF) circuits also stop in order to reduce the energy needed to transmit 1-symbol data.

The total energy for transmitting data is shown in Fig. 4(c). When the length of the transmitting data is long, the 1-symbol energy is dominant in the total energy, and when the data length is short, the stabilization energy is dominant. The proposed system can reduce the energy used for both startup and data transmission and achieve lower energy than a conventional system.

A block diagram of the proposed wireless transmitter is shown in Fig. 5. It consists of a baseband ring oscillator for a data clock, an RF ring oscillator for a pulse generator, a pulse counter, an identification (ID) generator, and a driver amplifier. The ring oscillator is effective for achieving a quick startup. When the baseband oscillator starts, the ID signals are generated. While the signal is set to 1, the RF oscillator generates a pulse train. The pulse counter counts the number of pulses and stops the RF oscillator when the count reaches 32. After that, the RF circuits stop and consume no more power. The measured power consumption of 1-bit generation is 10 nJ/bit. This means that when we transmit 50 bits of “1” data in 1 s, the total power consumption is 500 nW.

Fig. 5. Block diagram of proposed wireless transmitter.

3. Fabricated WSN

Photographs of a fabricated WSN are shown in Fig. 6. The configuration is also shown. The WSN has a solar battery, vibration sensor, wireless IC, antenna, and other devices. The total volume is 1 cm3 with a small-wire dipole antenna. The 1-cm2 printed circuit boards are stacked three-dimensionally to reduce the total size. The frequency band is set to 300 MHz, which means that no usage license is necessary under low output power conditions in Japan. The data rate is 1 Mbit/s and the bandwidth is 10 MHz. The transmitted data length is 36 bits including the preamble, fixed pattern, and ID. When the WSN detects sufficient vibration, it transmits the ID.

The measurement setup and results are also shown in Fig. 6. The WSN was charged with solar energy, and the data signals were then transmitted when the vibrations were detected. The ID signals were strong enough for a receiver located 3 m away to detect them and to demodulate bit data.

Fig. 6. Photographs of fabricated WSN and representation of bit data modulated by receiver.

4. Concluding remarks

The ultrasmall WSN powered by an energy harvester is expected to lead to new services. In this article, we introduced nanowatt circuit technologies for sensing and wireless circuits. The fabricated 1-cm3 WSN successfully transmitted an ID a 3-m distance using only solar energy.

In future, we will attempt to fabricate a much smaller WSN that can transmit signals a much longer distance using less energy; such an achievement would allow any object in the world to be connected to a network.


[1] G. Werner-Allen, K. Lorincz, M. Ruiz, O. Marcillo, J. Johnson, J. Lees, and M. Welsh, “Deploying a Wireless Sensor Network on an Active Volcano,” IEEE Internet Computing, Vol. 10, No. 2, pp. 18–25, 2006.
[2] T. Fu, J. Paek, N. Nothari, S. Rangwala, J. Caffrey, R. Govindan, E. Johnson, and S. Masri, “Monitoring Civil Structures with a Wireless Sensor Network,” IEEE Internet Computing, Vol. 10, No. 2, pp. 26–34, 2006.
[3] T. Arampatzis, J. Lygeros, and S. Manesis, “A Survey of Applications of Wireless Sensors and Wireless Sensor Networks,” Proc. of the IEEE International Symposium on Intelligent Control, Mediterranean Conference on Control and Automation, pp. 719–724, Limassol, Cyprus, 2005.
[4] M. Hempstead, D. Brooks, and G. Wei, “An Accelerator-based Wireless Sensor Network Processor in 130 nm CMOS,” IEEE Emerging and Selected Topics in Circuits and Systems, Vol. 1, No. 2, pp. 193–202, 2011.
[5] J. A. Paradiso and T. Starner, “Energy Scavenging for Mobile and Wireless Electronics,” IEEE Pervasive Computing, Vol. 4, No. 1, pp. 18–27, 2005.
[6] T. Shimamura, M. Ugajin, K. Suzuki, K. Ono, N. Sato, K. Kuwabara, H. Morimura, and S. Mutoh, “Nano-watt Power Management and Vibration Sensing on a Dust-size Batteryless Sensor Node for Ambient Intelligence Applications,” Proc. of the 2010 IEEE International Solid-State Circuits Conference Digest of Technical Papers (ISSCC), pp. 504–505, San Francisco, CA, USA.
[7] S. Oshima, K. Matsunaga, T. Shimamura, H. Morimura, and M. Harada, “1-cm3 Event-driven Wireless Sensor Nodes,” Proc. of the 13th International Conference on Communication Systems (ICCS), Singapore, 2012.
Shoichi Oshima
Researcher, Wireless Communication Circuits Research Group, Smart Devices Laboratory, NTT Microsystem Integration Laboratories.
He received the B.S., M.S., and Ph.D. degrees in electrical engineering from Tohoku University, Miyagi, in 2002, 2004, and 2008, respectively. He has been with the NTT Microsystem Integration Laboratories since September 2008. His research interests include wireless transceivers and circuit design for low-power communication systems. He is a member of IEEE and the Institute of Electronics, Information and Communication Engineers (IEICE).
Kenichi Matsunaga
Wireless Communication Circuits Research Group, Smart Devices Laboratory, NTT Microsystem Integration Laboratories.
He received the M.E. degree in physical electronics from Tokyo Institute of Technology in 2010. He joined NTT Microsystem Integration Laboratories in 2010 and has been studying low-power CMOS circuit design for wireless sensor nodes. His research interests include RF CMOS and mixed signal circuits.
Toshishige Shimamura
Manager, NTT Intellectual Property Center.
He received the B.E. degree in physical electronics and the M.E. degree in advanced applied electronics from Tokyo Institute of Technology in 1995 and 1997, respectively. He joined NTT Microsystem Integration Laboratories in 1997 and engaged in research on sensing circuit techniques for fingerprint sensor LSIs, optical MEMS switch control, and millimeter-sized energy-harvesting wireless sensor nodes. During 2006–2007, he collaborated with HORIBA, LTD. in developing environmental measurement systems. He received the 2006 IEICE Best Paper Award and the Best Paper Award of the Symposium on Integrated MEMS Technology at the 26th Sensor Symposium. He served as an Associate Editor of the IEICE Transactions on Electronics during 2010–2012. He is a member of IEICE, the Japan Society of Applied Physics (JSAP), and IEEE.
Hiroki Morimura
Senior Research Engineer, Supervisor, Group Leader of Advanced Ubiquitous Communication Circuits Research Group, Smart Devices Laboratory, NTT Microsystem Integration Laboratories.
He received the B.E. degree in physical electronics, the M.E. degree in applied electronics, and the Dr.Eng. degree in advanced applied electronics from Tokyo Institute of Technology in 1991, 1993, and 2004, respectively. He joined NTT LSI Laboratories in 1993. He has been engaged in R&D of low-voltage, low-power SRAM and CMOS fingerprint sensors. His research interests include highly sensitive noise-tolerant circuits for short-range wireless communications, ultralow power circuit techniques for batteryless sensor nodes, and integrated CMOS-MEMS technology. He has served on the program committee of the Symposium on VLSI technology, ISADS, and the program subcommittee on Advanced Circuits and Systems of SSDM. He has also served on the special section editorial committee of the IEICE Transactions on Electronics. He received the 2004 CSS Best Paper Award, the 2006 IEICE Best Paper Award, and the Best Paper Award of the Symposium on Integrated MEMS Technology at the 26th Sensor Symposium. He is a member of IEEE, IEICE, and JSAP.
Mitsuru Harada
Senior Research Engineer, Supervisor, Group Leader of Wireless Communication Circuits Research Group, Smart Devices Laboratory, NTT Microsystem Integration Laboratories.
He received the B.S. and M.S. degrees in physics from Tsukuba University, Ibaraki, and the Ph.D. degree in electronic engineering from Tohoku University, Miyagi, in 1985, 1987, and 2003, respectively. In 1990, he joined the NTT Atsugi Electrical Communication Laboratories, where he engaged in research on thin-film SOI devices. Since 1997, he has been researching high-speed/low-power CMOS circuits for wireless terminals.