The rise of Software Defined Instrumentation (SDI)

In a recent article, Prashanth Bhushan described his view of the Rise of Software Defined Instrumentation (SDI) https://www.thefastmode.com/author/3928-prashanthbhushan

The article discusses the emergence of Software-Defined Instrumentation. By decoupling the measurement functionality from hardware, SDI unlocks the possibility of getting rid of bulky dedicated hardware and opens up vast possibilities in terms of:
– quasi real-time test configurations
– remote management
– integration of generative AI into instruments
In addition to this, in the author’s view this decoupling gives also the opportunity to shift a portion or all of the processing part to the cloud for scalability. But is cloud dependency really necessary?

Modern SoC chips integrate AI accelerators, DSPs and CPUs capabilities that enable real-time processing on-device. Depending on the application, cost-efficient AI models can operate with the limited resources of embedded systems (so called Edge AI).

For many industrial sectors, including aerospace, telecom, medical devices and defense, using localized SDI over cloud-based SDI could mean ensuring security and privacy as well as minimizing the cost of cloud subscriptions, while benefiting from the flexibility of SDI.

While it is clear that the advantages of SDI will play a role in the future of industrial test and measurement, the question remains open on where companies will draw the line between edge and cloud based processing.