19 March 2015 in IEEE Transactions on Instrumentation and Measurement
Electrical impedance tomography (EIT) is used to determine the spatial conductivity distribution of a measurement environment that involves several applications in the areas of process engineering and medical diagnostics. In common EIT systems, the measurements are performed by means of current patterns which are injected into the measurement environment using a serial time-division-multiplexing (TDM) scheme for the excitation of different electrodes. The measurement rate can be increased by performing parallel excitation using orthogonal signals. In this paper, a code-division-multiplexing (CDM) measurement procedure is presented. To optimize the separation between different measurement channels and the dynamic range, orthogonal Walsh-Hadamard codes are used. The measurements are conducted with a fast EIT system with nine parallel excitation sources and 18 parallel measurement channels. The measured crosstalk rejection between different channels is larger than 98 dB. The maximum absolute deviation between different measurement sets for repeated measurements is less than 24 μV with a mean standard deviation of less than 8.2 μV. The dynamic range for impedance measurements based on different excitation procedures (TDM, frequency-division-multiplexing, and CDM) has been determined. Furthermore, reconstructed conductivity distributions based on measurements with the different excitation procedures have been compared with each other for different measurement scenarios (a root mean square difference of less than 1.2%). Finally, the influences of frequency-dependent measurement objects on the excitation procedures have been discussed.
2 April 2020 in IEEE Transactions on Instrumentation and Measurement
The measurement of the electromagnetic reflection characteristics of a device under test (DUT) is of interest in a large variety of applications. For this task, the established lowcost approach of using pulse-shaped excitation signals is more and more replaced by employing continuous binary coded excitation signals. The reason behind this is that reflection measurement systems with continuous signals offer a larger signal-to-noise ratio (SNR), which allows for a larger dynamic range compared to pulsed systems for the same measurement time. However, a correlator is needed to access the impulse response function (IRF) of the DUT. In this contribution, a reflection measurement system is presented using a hybrid correlation approach. In the latter, the correlation is performed using a combination of a hardware and a software correlator. The hardware of the developed measurement system has been optimized to achieve a high dynamic range and a high measurement accuracy. A demonstrator system has been realized using so-called almost perfect auto-correlation sequences (APAS) with a chip clock of 5 Gchip=s. With the optimized hardware, a dynamic range of 82 dB has been achieved. The measurement accuracy has been validated by means of distance measurements versus a high accuracy laser reference system achieving a maximum error of 25 μm. Furthermore, the measured reflection coefficient of a low-pass filter has been compared with the results obtained with a vector network analyzer (VNA).