The increasing relevance of big data applications in fields as the Internet of Things (IoT) and Industry 4.0 implies that sensors are requested to be secure and accurate. In the last years, sensors are evolving toward complex monitoring functionalities, increasing the complexity of data, meaning that the analysis stage is usually performed away from the sensor layer, i.e., the fog or the cloud. This separation entails issues for response time and security. As a possible way to address this data analysis closer to the edge, embedded machine-learning (ML) techniques have shown to be a good solution, leading to expert sensors. Feature extraction tools, as principal component (PC) analysis (PCA), might offer a solution to reduce the amount of data transmitted through the network, adding additional security because information is not transmitted as raw data. However, PCA is time-consuming and therefore, it should be carefully optimized according to the hardware used in the sensor device. This chapter proposes to embed the PCA inference stage in a low-cost field-programmable system on chip (SoC) (FPSoC) while performing a design space exploration for a general PCA inference problem. To this end, the authors analyze metrics, such as latency, scalability, and usage of hardware resources. The resulting architectures are compared to a multicore OpenMP approach to be executed in an ARM processor, analyzing the advantages of using the FPSoC implementation in speedup.

Chapter Contents:

• 6.1 Introduction
• 6.2 Related work
• 6.3 Principal component analysis: problem formulation
• 6.4 Workflow description
• 6.5 Embedded architecture
• 6.5.1 System-level architecture
• 6.5.2 PCA inference IP description
• 6.6 Experimental methodology
• 6.7 Experimental results
• 6.7.1 8- vs. 16-bit architectures
• 6.7.2 Hardware architecture vs. multicore approach
• 6.8 Conclusions
• Acknowledgments
• References

Inspec keywords: Big Data; field programmable gate arrays; data analysis; system-on-chip; learning (artificial intelligence); production engineering computing; feature extraction; telecommunication computing; Internet of Things; principal component analysis

Other keywords: general PCA inference problem; big data applications; data analysis; feature extraction tools; embedded machine-learning techniques; PCA inference stage; analysis stage; good solution; low-cost field-programmable system; response time; sensor device; additional security; complex monitoring functionalities; principal component analysis inference; raw data; sensor layer; increasing relevance; expert sensors

Subjects: Production engineering computing; Computer vision and image processing techniques; Data handling techniques; Other topics in statistics; Optimisation techniques; Communications computing; Logic circuits; Other topics in statistics; Knowledge engineering techniques; Industrial applications of IT