Development of a high-speed full digital processing phase detector for interferometry.

This study describes the development of a fully digital-type phase detector for plasma interferometry. This detector functions even in situations in which the phase changes rapidly or the input signal is too small to derive the correct phase shift from the intermediate frequency (IF) signal. The detector directly converts the IF signal waveform of the interferometer to the phase shift signal by means of data processing in a logic circuit. Thus, the phase is derived from the whole waveform of the IF signal. The IF signal of the interferometer is converted to in-phase and quadrature-phase signals by Hilbert transformation, processed by a digital low-pass filter, and converted to polar coordinates by a coordinate rotation digital computer algorithm to obtain the phase shift. A simulation of the high-speed full digital processing phase detector shows that a fringe jump does not occur unless the phase change rate exceeds 0.8 × 106 rad/s. This value is sufficiently large compared to the phase change velocity in rapid density increase resulting from a pellet injection. The phase conversion is simulated using a real IF signal from an interferometer measured with a Heliotron J device. The results show that the phase signal is correctly calculated by the full digital processing method from the IF signal, the phase derivation of which is typically difficult to obtain when using a conventional analog phase detector.

The synaptic targeting of mGluR1 by its carboxyl-terminal domain is crucial for cerebellar function.

The metabotropic glutamate receptor subtype 1 (mGluR1, Grm1) in cerebellar Purkinje cells (PCs) is essential for motor coordination and motor learning. At the synaptic level, mGluR1 has a critical role in long-term synaptic depression (LTD) at parallel fiber (PF)-PC synapses, and in developmental elimination of climbing fiber (CF)-PC synapses. mGluR1a, a predominant splice variant in PCs, has a long carboxyl (C)-terminal domain that interacts with Homer scaffolding proteins. Cerebellar roles of the C-terminal domain at both synaptic and behavior levels remain poorly understood. To address this question, we introduced a short variant, mGluR1b, which lacks this domain into PCs of mGluR1-knock-out (KO) mice (mGluR1b-rescue mice). In mGluR1b-rescue mice, mGluR1b showed dispersed perisynaptic distribution in PC spines. Importantly, mGluR1b-rescue mice exhibited impairments in inositol 1,4,5-trisphosphate receptor (IP3R)-mediated Ca(2+) release, CF synapse elimination, LTD induction, and delay eyeblink conditioning: they showed normal transient receptor potential canonical (TRPC) currents and normal motor coordination. In contrast, PC-specific rescue of mGluR1a restored all cerebellar defects of mGluR1-KO mice. We conclude that the long C-terminal domain of mGluR1a is required for the proper perisynaptic targeting of mGluR1, IP3R-mediated Ca(2+) release, CF synapse elimination, LTD, and motor learning, but not for TRPC currents and motor coordination.