Research Interests:
Amin has conducted multiple research projects in colaboration with Concordia University, Montreal, McGill University, Montreal and Colorado State University, Colorado. All the projects outcomes were published as standard IEEE conference proceedings. Amin’s research interest includes:
– Wide gain band-width circuit and system implementation in CMOS technology
– Multi-mode analog filter implementation using universal amplifier module (UAM)
– Modeling and simulation of high-speed transmission line networks
– Numerical algorithms and techniques
– Design Automation of VLSI circuits and High speed analog (RF) circuits
– Stochastic circuit simulation of high-speed passive distributed networks
Research @YorkU:
At Concordia, Amin has mainly worked on 5-Axis CNC Micromilling for Rapid, Cheap, and Background-Free NMR Microcoils. he superior mass sensitivity of microcoil technology in nuclear magnetic resonance (NMR) spectroscopy provides potential for the analysis of extremely small-mass-limited samples such as eggs, cells, and tiny organisms. For optimal performance and efficiency, the size of the microcoil should be tailored to the size of the mass-limited sample of interest, which can be costly as mass-limited samples come in many shapes and sizes. Therefore, rapid and economic microcoil production methods are needed. One method with great potential is 5-axis computer numerical control (CNC) micromilling, commonly used in the jewelry industry. Most CNC milling machines are designed to process larger objects and commonly have a precision of >25 μm (making the machining of common spiral microcoils, for example, impossible). Here, a 5-axis MiRA6 CNC milling machine, specifically designed for the jewelry industry, with a 0.3 μm precision was used to produce working planar microcoils, microstrips, and novel microsensor designs, with some tested on the NMR in less than 24 h after the start of the design process. Sample wells could be built into the microsensor and could be machined at the same time as the sensors themselves, in some cases leaving a sheet of Teflon as thin as 10 μm between the sample and the sensor. This provides the freedom to produce a wide array of designs and demonstrates 5-axis CNC micromilling as a versatile tool for the rapid prototyping of NMR microsensors. This approach allowed the experimental optimization of a prototype microstrip for the analysis of two intact adult Daphnia magna organisms. In addition, a 3D volume slotted-tube resonator was produced that allowed for 2D 1H–13C NMR of D. magna neonates and exhibited 1H sensitivity (nLODω600 = 1.49 nmol s1/2) close to that of double strip lines, which themselves offer the best compromise between concentration and mass sensitivity published to date.
Research @Concordia:
At Concordia, Amin has mainly worked on Universal Amplifier Module (UAM) implemented in latest CMOS technology, that can be configured to function as a VCVS, CCCS, VCCS, CCVS and CCII. As a result, the amplifier can be used to produce four kinds of basic circuit transfer functions, i.e., voltage-, current-, transadmittance- and transimpedance- transfer functions. This is validated by presenting simulation results for second order voltage, current, transadmittance and transimpedance bandpass filter (BPF) transfer functions using only three UAM devices. The device is expected to be useful in a complex VLSI system environment where interfacing between sub-systems with varied impedance levels is required.
Research @McGill:
At McGill, Amin has mainly worked on stochastic simulation of high-speed passive distributed networks. Stochastic distributed networks can be characterized in the frequency-domain by augmented multiport Y-parameter sampled data based on a stochastic Galerkin’s formulation of the network equations. In his research he proposed a Loewner Matrix approach towards generating time-domain macromodels from the augmented multiport data. The key benefit of his research is that the superior scaling of the computational complexity of the Loewner Matrix approach with respect to number of network ports is utilized to generate the macromodel much more efficiently than the traditional Vector Fitting approach.