Embedded Platform for Gas Applications Using Hardware/Software Co-Design and RFID
ABSTRACT
This paper presents the development of a wireless low power reconfigurable self-calibrated multi-sensing platform for gas sensing applications. The proposed electronic nose (EN) system monitors gas temperatures, concentrations and mixtures wirelessly using the radio-frequency identification (RFID) technology. The EN takes the form of a set of gas and temperature sensors as well as multiple pattern recognition algorithms implemented on the Zynq system on chip (SoC) platform. The gas and temperature sensors are integrated on a semi-passive RFID tag to reduce the consumed power. Various gas sensors are tested including an in house fabricated 4_4 SnO2 based sensor and 7 commercial Figaro sensors. The Data is transmitted to the Zynq based processing unit using a RFID reader where it is processed using multiple pattern recognition algorithms for dimensionality reduction and classification. Multiple algorithms are explored for optimum performance including principal component analysis (PCA) and linear discriminate analysis (LDA) for dimensionality reduction while decision tree (DT) and k-nearest neighbors (KNN) are assessed for classification purpose. Different gases are targeted at diverse concentration including carbon monoxide (CO), ethanol (C2H6O), carbon dioxide (CO2), propane (C3H8), ammonia (NH3) and hydrogen (H2). An accuracy of 100% is achieved in many cases with an overall accuracy above 90% in most scenarios. Finally, the hardware/software heterogeneous solution to implementation PCA, LDA, DT and KNN on the Zynq SoC shows promising results in terms of resources usage, power consumption and processing time.
EXISTING SYSTEM :
Many researchers have already tried to integrate sensors in an RFID tag such as in where a carbon nano tube baseg as sensor is integrated into a passive and low cost RFID tag. The sensor has been successfully tested to detect NH3 gas. However, considerable optimization and improvement is to be done. Another similar work where a gas sensor based on carbon nano tube is integrated into an RFID tag has been presented in and used for the detection of small quantities sensor have been integrated into a passive RFID tag in the tag is tested and it achieves real-time wireless communication with RFID readers and the operating temperature is between 20 _C and 30 _C. Integrating sensors with RFID requires meeting very challenging specifications and constraints. Firstly, the sensors need to operate at extremely low power to avoid affecting the tag lifetime or limiting its coverage range. For instance, reported power consumption of temperature sensors for RFID applications is around tens to hundreds microwatts, which does not suite RFID applications. However, gas sensors power demand increases to reach a few hundred watts making the integration of gas sensors on RFID tags extremely challenging. This increase is due to the requirement of heating a sensing film at high temperatures . Secondly, the sensors deployed in RFID devices require minimal or preferably no human intervention such as battery replacement or sensors calibration. For instance, temperature sensors often require gain and offset calibration . Chemical and gas sensors typically suffer from drift issues and sensor poisoning, which is a gain and offset shift of the sensors response after exposure to target species . Current commercially available electronic nose (EN) systems require frequent supervised calibration (sometimes on a daily basis). This would be prohibitively expensive for RFID applications involving a very large number of deployed sensors. In addition, current chemical and gas sensors have limited sensitivity, an in ability to deal with non-stationary background odors and are unable to discriminate between odor quality and odor concentration.
PROPOSED SYSTEM
In this paper, a new generation of ultra-low power and self-calibrated multi-sensing platform for RFID applications is proposed and developed. The platform mainly focuses on temperature and gas measurements that could be used in several applications including gas processing, security and logistics and shipment industries. The applications of this multi-sensing platform can also be extended to include environmental monitoring and pollution control, which are important applications globally. These power and calibration issues are real obstacles that need to be addressed before the birth of sensor tags and their deployment in large scale RFID applications can truly be witnessed. A new generation of temperature and gas sensors is therefore required in order to meet the future demand of a low cost, reliable, real-time, portable and self calibrated multi-sensing platform.
CONCLUSION :
A low power reconfigurable self-calibrated multi-sensing platform for gas applications have been presented in this paper. The platform is made of two main parts, the sensing part and the processing part. The processing part takes the form of a heterogeneous and reconfigurable platform which is the Xilinx Zynq SoC. Various pattern recognition algorithms including PCA, LDA, DT and KNN are implemented on this platform for hardware acceleration and to perform gas identification in real-time. Finally, the heterogeneous hardware/software codesign solution to implement PCA, LDA, DT and KNN on the Zynq SoC as well as the integration of temperature and gas sensors in an RFID tag show promising results in terms of resources usage, power consumption and processing time. It is worth mentioning that the best solution in terms of the classifier to use or the choice of the number of principal components as well the hardware directives will depend on the final application and resources availability. The best solution is the trade-off between the best accuracy, the processing time and the resources available on the hardware. Someone may choose the best accuracy scenario with the fastest implementation for critical and vital applications while compromising on the accuracy and speed is possible for other applications to be able to save on hardware resources and power consumption.