Many sensors require better data acquisition, analysis
In the age of Big Data, we often link value to the number of terabytes generated, but engineers should really evaluate whether they are better informed. Teams should be asking, "Can we make the right decisions and can we make them faster?" IBM CEO Ginni Rometty refers to data as the "new natural resource," meaning data needs to be processed, refined, and presented in a meaningful way that leads to a decision. For engineers, this means instrumentation must be smarter, and sensors, measurement hardware, data busses, and application software need to work together to provide actionable data at the right time.
Consider the Internet of Things (IoT). By 2020, some estimate that the number of connected sensors and devices will exceed 1 trillion. Without intelligence distributed throughout these systems, the amount of data produced will far exceed our ability to understand and process it, not to mention the network bandwidth required to send this magnitude of data.
Filtering actionable data
By employing smarter devices, only actionable data will get transmitted along with the context required to understand the information. By moving processing and intelligence from a centralized host to the data source, data can be refined and reduced, leading to faster decisions and less reliance on data transfer. Similar to the server/thin-client model of computing from the 90s, consider the big data trend as a transitional phase while sensing and processing models catch up.
The relevance of this trend in engineering cannot be overstated. As systems become more complex, the amount of data required to describe those systems has grown beyond comprehension, which inevitably results in longer project schedules and less efficiency in development. More advanced tools and smarter measurement systems will be essential to managing this explosion of data and help engineers make informed decisions faster. In conventional measurement systems, sensors are often considered simply as a source-meaning a sensor changes a physical signal into an electrical signal to be sampled down the line with an analog to digital converter.
Smaller, smarter sensors
With emerging sensor, processor, and battery technologies, intelligence can move much farther down the signal chain. With intelligence built into the sensor, data does not have to wait until reaching the instrumentation to be processed. For example, some intelligent accelerometers can digitize and perform fast Fourier transfer (FFT) analysis on vibration data before forwarding that data to instrumentation. Sending frequency data and not raw time domain data reduces the burden on the data transfer and reduces the processing required by the instrumentation. Right now, microelectromechanical systems, MEMS technologies, are playing a major role in this trend. By using new fabrication techniques, entire sensors can now be implemented in a small silicon chip. These new sensors use much less power, require much less space, and are orders of magnitude cheaper to manufacture. Coupled with lower power, lower cost processors, and improved power management techniques, engineers can integrate sensing and intelligence at a scale that would be impossible with traditional analog sensors.
Already, MEMS technology has revolutionized the mobile world. Consider your smartphone, for example. In a matter of years, accelerometers, gyroscopes, photos sensors, and microphones have transformed the cell phone from a device that makes calls to a system that interacts with the world around it, whether by tracking user health or alerting users about inclement weather. This explosion of sensors in the consumer world has driven down the cost and driven up the performance of MEMS sensors to the point where some measurements are approaching the performance of their analog counterparts at a fraction of the cost.
Better sensors, smarter decisions
Temperature, sound, and vibration have seen much wider adoption into test, measurement, and control systems. A recently introduced sound camera uses new MEMS microphone technology with field-programmable gate array (FPGA) processing and signal synchronization to offer a higher performance, lower cost, smaller sound camera than previous analog sensing devices.
As sensor and processing technologies continue to decrease in size and cost while increasing in performance, engineers will see a wider set of smart sensor options for better understanding of the systems they are testing. Combined with smarter instrumentation, smarter networks (buses), and smarter software, engineers will be able to better manage the collection, processing, and analysis of this data and therefore make more informed decisions faster.
– Jim Schwartz is group manager – data acquisition at National Instruments; edited by Mark T. Hoske, content manager, CFE Media, Control Engineering, firstname.lastname@example.org.
- Advances in smart sensors and processors create more data collection opportunities to create knowledge.
- Data collection and analysis needs to be smarter.
- Big data isn’t just more data, but better decisions, more quickly.
Where in your world do you need faster, smarter decisions, and how can smarter sensors and better data acquisition and analysis tools help?
- A smarter, portable sensors example
- Links to NI and IBM information on data acquisition, Big Data
- Links to related NI articles: 5 red flags before choosing configuration-based data acquisition software, Controller designed to reduce measurement system cost
- More about the author.
Smarter, portable sensors
Temperature, sound, and vibration have seen much wider adoption into test, measurement, and control systems, explains Jim Schwartz, group manager – data acquisition at National Instruments. "One example, the SeeSV-S205 sound camera from THP Systems, uses new MEMS microphone technology with FPGA processing and signal synchronization to offer a higher performance, lower cost, smaller sound camera than previous analog sensing devices."
The sound camera can capture 25 images per second. While sound cameras have traditionally been heavy and expensive instruments, THP Systems said: "Now you can carry your sound camera anywhere to perform measurements."
More about the author
Jim Schwartz manages the Data Acquisition Product Team at National Instruments (NI). In his current role, he is responsible for steering product development and positioning of data acquisition platforms to align with market trends and company vision. Schwartz began his career at NI in January 2009 as a member of the Engineering Leadership Program, where he served as a team leader and provided technical support for data acquisition and modular instrumentation at top accounts. He graduated with a bachelor’s degree in electrical engineering from Kansas State University.
See additional related articles linked below.