Font Size: a A A

Research, The Key Technologies Of The Near-infrared Detector Based On Dsp + Fpga

Posted on:2013-01-20Degree:MasterType:Thesis
Country:ChinaCandidate:Y L QianFull Text:PDF
GTID:2218330371460361Subject:Circuits and Systems
Abstract/Summary:PDF Full Text Request
In this paper, based on principles of design, I have designed a fast real-time detection of liquor alcohol portable near-infrared spectroscopy instruments, and focus on its key technologies.Control system combined with DSP+FPGA design, DSP achieves algorithm,temperature compensation, and the management system is running; FPGA to implement some interface, timing control, simplify the external circuit to improve the reliability of the system, better solution to the high-speed real-time low level light signal processing circuit requirements.Signal acquisition circuit consists of low-cost and low-power near-infrared LED and the optical frequency converter TSL245 composition, high stability, less susceptible to interference and high accuracy, more compact circuit design. Through the acquisition of experimental data, complete hardware and software design of the temperature compensation to establish a mathematical model of the frequency correction and temperature, linear regression analysis, and simulation.Hardware and software combination of weak signal detection approach, through the pre-signal processing circuit and the appropriate detection signal amplification, the optical signal is more easy to handle, and then by the rules of the square wave shaping circuit filter noise output, so that the frequency acquisition module read. Then the use of wavelet analysis on the signal to noise spectrum analysis, combined with the signal contained in the analysis of the characteristics of the noise spectrum to achieve the targeted processing f the signal noise, remove harmonic noise signal acquisition, excluding mutations, to improve under the premise of de-noising the computational speed can retain the details of the signal better to improve the accuracy of measurement instruments. Finally, the use interpolation segmentation establish the alcohol and the frequency of the fitted curve, in order to achieve more accurate processing of spectral data.Combination of these design features, the paper studied the design of the main features of the near-infrared spectral detector of small size, high integration, easy to carry, suitable for on-site for rapid detection.
Keywords/Search Tags:DSP+FPGA, TMS320F2812, near infrared spectroscopy, wavelet analysis, weak-signal detection, food testing
PDF Full Text Request
Related items