Application of undersampling technique for the design of an NMR signals digital receiver

Abstract

ABSTRACT:In this article, undersampling technique theory and design, realization, and workbench test of a low-cost digital MRI (magnetic resonance imaging) receiver are reported. The authors first discuss classic analog receiver architecture, reporting its disadvantages, and then present a receiver with a complete description of the electronic circuit. The intended use of the subsystem should be as a radiofrequency (RF) receiver chain in a low-cost dedicated MRI scanner (e.g., suitable for musculoskeletal limbs studies). The digital receiver consists of a passband (antialiasing) filter, an ADC (analog to digital converter) to digitalize the signal with the undersampling technique, and a DDC (digital down converter) for the frequency translation and filtering to process the signal using a PC. The use of the DDC guarantees perfect signals quadrature and high-performance filtering; moreover, it is adaptable to each modification of the system because of its programmability. Detailed specifications and hardware design of the digital receiver that is designed to be used in a dedicated MR scanner is provided, guaranteeing top performance and low cost.

Measuring information gain for frequency-encoded super-resolution MRI

“…Often in NMR, the bandwidth of the antialiasing filter is much larger than the bandwidth of the measured signal [22]. In this case, the convolution integral may be approximated by s(k): Numerical simulations and experiments with real MRI data were carried out to study the extent to which this approximation is valid.…”

“…The bandwidth of the ADC is typically larger than that of s(k, T 1 , T 2 ) because oversampling is often used in MRI to increase the SNR of the measured data [22,48]. Since the bandwidth of A(k) is set to that of the ADC, we also expect BW A to be much greater than the bandwidth of s. Under these conditions, it is relatively easy to show (Appendix B) that dL(k m , r n )60, which is a result of the convolution integral in Eq.…”

“…The pixels of each acquisition are interleaved to create the merged data, which are used as an input to a deconvolution algorithm that calculates a high-resolution image. [19][20][21][22]. Via the Fourier convolution theorem [23], the application of the filter is modeled as a convolution in k-space.…”

“…For computational tractability, the filter A k Dk Á À is represented by a sinc function. Other filters[20,22,48] would require a significant amount of additional computational labor. The values of k that were used in our simulations were determined using typical parameters in an MR experiment.…”

Efficient multichannel coil data compression: A prospective study for distributed detection in wireless high‐density arrays

“…Both transmission methods can support analog or digital information. Analog transmission of MR signals with performance similar to that of coaxial cables requires meeting very stringent dynamic range and noise figure performance specifications (8,15,16) that test the limits of current technology. Conversely, digital transmission methods are sufficiently advanced that MR data could readily be transmitted if digitization were performed at the coil.…”