Outreach

INAOE develops the first dielectric permittivity sensor node connected to a short-range wireless network

On February 6, the article "Microwave-sensor-node integrated into a short-range wireless sensor network" was published in the prestigious journal Nature´s Scientific Reports. The authors are M. C. Miguel Hernández Águila, PhD student in Electronics at INAOE, and Dr. José Luis Olvera Cervantes, researcher of the Electronic Coordination of this Institute.
Sensor networks are a technological trend with manyt applications. It is estimated that by 2025 more than 75 billion devices will be connected to the Internet, and that by 2030 there will be 31 devices per person. Today there has been great interest in leveraging massive networks of devices connected to each other to make the concept of the Internet of Things (ioT) a reality. The IoT has multiple applications in health, transportation, agriculture, industry, home, education, weather forecasting, etc.

Miguel Hernández Águila. Credit: INAOE.

A fundamental piece for the ioT are wireless sensor networks (WSN). WSNs can be classified as short-range, long-range, and hybrid WSNs. Short-range wireless networks are usually implemented through different communication protocols such as ZigBee, RuBee, Wibree, Z-Wave, RFID, WiFi, etc. ZigBee-based short-range networks are widely used because they have a maximum data transfer rate of 250 Kb/s, can reach a range of up to 75 m, allow working with networks of up to 64,000 sensor nodes and They support mesh, ad hoc, and star topologies. ZigBee networks are made up of a client, a coordinator, and a sensor node. The client provides the user with an interface for network control while the coordinator is in charge of communicating the client's orders and collecting data from the sensor nodes.
A sensor node is mainly provided by a power unit, a processing unit, a transceiver and a sensing unit. The power unit is in charge of energizing the entire sensor node, the processing unit controls all the operations that are carried out by the sensor node, and the transceiver allows the sensor node to establish communication with the WSN. The sensing unit is the part of the sensor node responsible for capturing physical magnitudes such as humidity, gases, ph, temperature, sound, dielectric permittivity, etc.
Dielectric permittivity sensors are a type of sensors that relate dielectric permittivity at microwave frequencies to physical variables. Around the world many dielectric permittivity sensors have been proposed for applications in non-invasive glucose detection, milk quality monitoring, moisture measurement, liquid characterization, chromatography, meat quality, food adulteration, measurement of temperature, ph measurement, gas detection, diabetic foot ulcer detection, breast cancer detection, ultraviolet light detection, measurement of salt and sugar concentration in water, urine electrolyte characterization, tissue moisture evaluation, chemical reactions in gases, etc. These types of sensors are the ones being investigated in the group of Dr. José Luis Olvera Cervantes.

Dr. José Luis Olvera. Courtesy.

In the search for patents and scientific articles, only one sensor connected to a short-range WSN through the use of a vector network analyzer (VNA-laboratory equipment) integrated with a Bluetooth module has been found. The use of the VNA has two main disadvantages: the high cost of the system and the infeasibility to implement networks with multiple permittivity sensors, since a VNA is required for each sensor.
A sensor node connected to a WSN was developed at INAOE and was published in Scientific Reports (https://www.nature.com/srep/) under the title “Microwave-sensor-node integrated into a short-range wireless sensor network”. In this work, a prototype sensor node integrated into a short-range wireless network based on ZigBee technology is presented. The sensor node measures the opening of structural cracks through the reflection coefficient measured at the input of a one-port microwave sensor which is fixed between the crack slit. The sensor node uses a patch antenna-based displacement sensor and an analog Front-end circuit to read the sensor. The analog front-end circuit consists of a synthesizer, an isolator, two Wilkinson-type power dividers, and a mixer. The output signal of the Front-end is digitized with an Arduino UNO ADC. The processing unit and the transceiver are implemented through an Arduino UNO and an XBee module respectively. For its part, the network consists of a client PC made through a PC with Matlab 2018 communicated to the sensor node through a coordinator device implemented through an XBee module and a TTL-USB converter.
M. C. Miguel Hernández Águila explains the relevance of this research work: “In the last decade, different works have been reported in the literature that have proposed different models of permittivity sensors. Permittivity allows you to measure different physical magnitudes. Based on this, microwave sensors have been proposed to painlessly measure glucose, contaminants in water, alteration in food, there are even some works in which they have tried to detect breast cancer as well, all this is not invasive since, as electromagnetic waves are transmitted, they can interact with the tissue and damage it. The problem is that all these jobs require laboratory equipment for their measurement, a VNA, that is, a vector network analyzer, and this is the main problem because the equipment is expensive, bulky, and generally requires specialized staff for interpretation. In contrast, we propose a reflectometer that allows measuring this type of sensors without the need to use a VNA. We designed a reflectometer circuit with different microwave components that can be built at low cost, with small dimensions, and that can read this type of sensor."
Another contribution of the project is that this type of sensors can be integrated into a wireless sensor network. “The problem is that each sensor would need a VNA for its reading and the VNA can hardly connect to a wireless network because it would need a module with some type of technology such as Bluetooth, Zigbee, WiFi, to connect and, although there are VNA models that can do it, they are expensive. What we propose is to integrate a small Zigbee module into our reflectometer circuit that allows remote reading of this circuit and to monitor it through a simple computer or integrate it into a wireless network of Zigbee devices. The advantage of this technology is that you can connect up to 64 thousand devices at the same time and make a gigantic wireless network. These are the strengths of our work. It is the first work in the literature that enables permittivity sensors to be connected to a wireless network but does not require a VNA."