Building sensor networks using Powerline Communication - Embedded.com

Building sensor networks using Powerline Communication

This “Product How-To” article focuses how to use a certain product in an embedded system and is written by a company representative.

Sensors integrated into structures, machinery, and the environment, coupled with the efficient delivery of sensed information, could provide tremendous benefits to society. The power of wireless sensor networks lies in the ability to deploy large numbers of tiny nodes that assemble and configure themselves.

Usage scenarios for these devices range from real-time tracking and monitoring of environmental conditions to ubiquitous computing environments and real-time monitoring of the health of structures or equipment.

Sensors networks in general pose considerable technical problems in terms of data processing, communication, and sensor management. Because of potentially harsh, uncertain, and dynamic environments, along with energy and bandwidth constraints, sensor network applications will likely encounter one or more of the following challenges:

Scalability – The number of sensor nodes deployed may be of the order of a few hundred or thousand. A designer must be confident that the network technology being used is capable of scaling to meet the eventual system need.

Increasing the number of nodes in the system will impact either the lifetime or effective sample rate. Increasing the number of sensing points will require more data to be transmitted which will increase the power consumption of the network.

Topology – Knowledge of the network structure is essential for a sensor in the network to operate properly. Each node needs to know the identity and location of its neighbors to support processing and collaboration.

For ad hoc networks, the network topology has to be constructed in real time and updated periodically as sensors fail or new sensors are deployed. In the case of a mobile network, since the topology is always evolving, mechanisms should be provided for the different fixed and mobile sensors to discover each other.

Production Costs – In addition to one or more sensors, each node in a sensor network is typically equipped with a radio transceiver or other wireless communications device, a small microcontroller, and an energy source (usually a battery) which all add to the overall cost.

Since a sensor network consists of a large number of nodes, the cost of a single mode is very important to justify the overall cost of the network.

Transmission Media – Most sensor networks are linked by a wireless medium. This could be radio, infrared or optical based. Issues like spectrum license, robustness and communication distance, cost, ease of implementation, etc. need to be addressed while choosing the right medium for the application.

Power consumption – The single most important consideration for a wireless sensor network is power consumption. While the concept of wireless sensor networks looks practical and exciting on paper, if batteries are going to have to be changed constantly, widespread adoption will not be practical for most applications.

To address these challenging issues, we describe how to use Powerline Communication (PLC) based on the existing power line infrastructure in a home, office or other building, both indoor and outdoor, for networking and communication and thereby eliminating the expense and inconvenience of antenna-based networks.

It makes use of the Cypress CY8CPLC20 . It is an integrated PLC chip with the Powerline Modem PHY and Network Protocol Stack running on the same device to allow integration of multiple system functions in a single device, thereby reducing Bill of Materials (BoM) costs and providing a smarter way to implement command and control.

PLC nodes inherently require some form of coupling to prevent the input circuits from being destroyed by the high voltage powerlines and also to transfer the maximum possible energy between the low voltage network and the modem.

This should be applicable for all the signals within the bandwidth of the frequency used, and it should attenuate as much as possible all the signals off this band and among them especially the corresponding to the always present network voltage.

PLC as a Foundation
PLC technology is capable of transmitting data via the electrical supply network, and therefore can extend an existing power network through extension cables with the installation of specific units. This makes PLC networks inherently scalable.

Powerline networks also use a bus topology. The principle of PLC consists in superimposing a high frequency signal at low energy levels over the 50 Hz electrical signal. This second signal is transmitted via the power infrastructure and can be received and decoded remotely. Thus, the PLC signal can be received by any PLC receiver located on the same electrical network.

Multiple sensor nodes can send data along the power-lines to the same receiver. When designing a networked system on a bus topology, addressing is an important issue. Ideally, developers will want to utilize a network protocol that allows addresses to be manually assigned.

This enables more sophisticated implementations than is possible in those where the controller simply finds new nodes and assigns addresses to them without manual configuring.

The bus topology eliminates the need for complex routing algorithms that are normally associated with wireless sensor networks. In addition, the transmission distance of PLC is significantly further than that of the short-range radio communications and, if more distance is required, repeaters can be used.

PLC nodes inherently require some form of coupling to prevent the input circuits from being destroyed by the high voltage powerlines and also to transfer the maximum possible energy between the low voltage network and the modem.

This should be applicable for all the signals within the bandwidth of the frequency used, and it should attenuate as much as possible all the signals off this band and among them especially the corresponding to the always present network voltage.

There a many legal issues associated with using PLC communications as devices must co-exist with other equipment on the same network or on the same electromagnetic spectrum when signals are transmitted.

To facilitate coexistence, developers can use a narrowband PLC implementation, where narrowband refers to the frequency range used. In Europe, narrowband applications must comply with standards defining equipment emission, immunity, impedance, etc as defined by EN50065.

In the US, the FCC reserves a frequency between 9 and 490 kHz for signaling and communications on powerline as defined in Federal Communications Commission, Part 15 ” Radio Frequency Devices, Sec.15.113 “Power Line Carrier System. PLC implementations that support both of these narrowband ranges can enable a single product to serve for each of these markets.

Architecture
Described next is implementation of a sensor network that provides real time information of parameters such as rainfall, ambient light, temperature, airflow, salinity and soil moisture for a farm to a central monitoring station. Data collected from these boards can be used for crop management and spatial variability studies.

Data such as soil water availability, soil compaction, soil fertility, plant water status, local climate data, and so on can be inferred from these readings. A farmer could then easily monitor each area of the farmland in real-time to avoid frost, manage irrigation, determine fertilizer applications, and arrange harvest schedules.

A block diagram of a sensor node is shown in Figure 1 below . It shows only two digital and analog sensors connected to the chip but, because of the configurable nature of the core, it could quite easily be as many sensors as required.

The CY8CPLC20 family can have up to five I/O ports that connect to the global digital and analog interconnects, providing access to 16 digital blocks and 12 analog blocks. The digital blocks can be configured to be timers, counters, SPI, I2C, UART and USB interfaces.

The analog blocks can create the signal processing chain needed by the application and can be configured as band pass filters, instrumentation amplifiers with gains upto 48x, analog to digital converters with upto 14 bit resolution, and digital to analog converters.

Figure 1 Block Diagram of a Sensor Node

Implementation
The fundamental concept of collecting data and making decisions based on that data in agriculture has been around for many years. This was easier to do without technology on small plots. But as the size of farms grew, this no longer was possible.

The larger farms require new techniques and tools such as global positioning, sensors, and information management tools to assess and understand variations. Collected information may be used to more precisely evaluate optimum sowing density, estimate fertilizers and other inputs needs, and to more accurately predict crop yields.

In Figure 2 below we show an example where a powerline network can be used in a farm environment. A critical requirement for collecting data on the spatial variation of a parameter is an ability to accurately resolve positions in a field.

Figure 2. Powerline Network Bus Topology.

All data must be geo-referenced so that a representative field map may be built and for the purpose of correlating information on various parameters obtained from a field.

In general, there are two sampling strategies that can be used for data collection: grid and zone sampling. Grid sampling uses a systematic approach that divides the field into squares or rectangles of equal size. We have illustrated grid sampling in Figure 3, below .

The data collected by the base station combined with the yield of each monitoring node can be compared to maps of soil test data, chemical application maps, and other information, resulting in a recommendation for the next year's site-specific management plan.

When layered in a Geographic Information System with yield maps, a map of the data collected displays evidence of relationships between crop yield and field condition variables. Examples of maps that can be created are shown in Figure 3.

Figure 3. Example maps generated from the data. The figure on the top left shows elevation, top right shows the yield map and figure below shows the electrical conductivity of the soil. Images courtesy the Australian Centre for Precision Agriculture. )

Location specific applications can be developed for the system based on the data collected and the possibilities are endless. For example, the receiver unit of the power-line communications system could be placed at the mains switching board. It can then be connected to other already automated devices on the existing field bus and these devices can be controlled based on the sensor values.

( Rahul Parsani and Rohan Gandhi are applications engineers atCypress Semiconductor. )

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.