This chapter gives an introduction to signal processing techniques that are relevant to many of the biomagnetic studies described in earlier chapters. We first consider the process of averaging a repeated signal to enhance the portion that is time-locked to a reference in comparison with the uncorrelated portion, or “noise.” A special example is a sinusoidal signal of known frequency, for which the technique of phase-sensitive detection may be used to improve the signal-to-noise ratio. We then discuss the process of digitizing analog data, as an introduction to the subsequent section that describes how a signal can be analyzed in terms of its individual frequency components and what the resulting “Fourier transform” signifies. Here the special techniques of digital signal processing play a crucial role: Special attention is devoted to how the problem of aliasing can be avoided by choosing a sufficiently rapid sampling rate for recording the data. In the last section we turn to filtering techniques that permit a signal to be processed by the selective attenuation or enhancement of some of its frequency components. More sophisticated methods for digital filtering will also be addressed, including nonrecursive, recursive, and adaptive filters.
KeywordsImpulse Response Discrete Fourier Transform Sine Wave Digital Filter Adaptive Filter
Unable to display preview. Download preview PDF.