Wireless sensor networks are a hot topic in wireless technology research right now. In this blog post, we describe the basic technology, talk about some applications and challenges, and explain why Nutaq's ZeptoSDR is ideal for prototyping the sensors that form the network.
The Internet of Things (IoT) is an idea where any physical object can be connected to the Internet: from your kitchen fridge to a road sign indicating the speed limit in your neighborhood. Such a reality will require the installation of many wireless sensors and the streaming of data to processing nodes. Cognitive radio and wireless spectrum analysis technologies will enable the efficient use of spectrum. Wireless sensor networks will become part of our daily lives.
The principle behind wireless sensors is simple: many devices, distributed in different locations, are used to monitor physical conditions such as temperature, pressure, or light. This information needs to be transmitted from the different devices to a main location. Sensors are linked together, forming a machine-to-machine (M2M) network. The individual sensors are called nodes. A gateway sensor node provides a link from the network to the main location, where a central processing unit like a computer can record and process the data.
Figure 1: A wireless sensor network
The potential applications are many, ranging from healthcare monitoring, where sensor in the human body transmit information useful for health purposes, to air pollution monitoring, where sensors gather information for environmental purposes. The applications are numerous – some people even use wireless sensors to produce home-made wine!
The sensor itself needs to be able to communicate. Therefore, it must contain a processing unit, communication resources like a radio front-end with antenna, and an operating system able to coordinate the numerous functions. Depending on the sensor type, some processing can be performed on the sensor itself. On-sensor processing however leads to trade-offs in the size, weight, and cost of the units. In applications where sensors need to be small, cheap, and light, the amount of processing that can be performed on the sensor is limited.
Another challenge is power consumption. For the sensor to be wireless, energy must be provided to obtain the necessary electricity to run it – be it a battery or an energy-harvesting device that uses the surrounding light, heat, electromagnetic waves, or other source. Obviously, the less power the sensor consumes, the longer its lifetime.
The perfect prototyping tool
As sensors require a low-power, low-cost processing unit and a radio interface, the ZeptoSDR software-defined radio (SDR) platform is an excellent choice for a prototyping tool. The ZeptoSDR is based on the low-power Zynq system-on-a-chip (SoC) hybrid architecture: a dual-ARM Cortex-A9 processor allied to an Artix-7 field-programmable gate array (FPGA). The system includes a powerful radio front-end that is suitable for all 3GPP standards and most IEEE ones. The RF front-end supports frequencies from 300 MHz to 3.8 GHz and bandwidths from 1.5 MHz to 28 MHz. Its numerous connectors, including HDMI, USB, and 3.5-mm audio jacks enable the connection of physical measurement devices to the processing unit for prototyping the sensor systems. The ZeptoSDR could be used to design the algorithm that goes into the SoC. Once the algorithm is developed, a sensor design could re-use the IP. Nutaq offers engineering services and could provide the RF front-end for the final product, which would reproduce a state-of-the-art design similar to the ZeptoSDR platform.
Radio420 (RF front end): http://www.nutaq.com/products/radio420x
ZeptoSDR (prototyping platform): http://www.nutaq.com/products/zeptosdr