P. Siy

Wayne State University, Detroit, Michigan, United States

Are you P. Siy?

Claim your profile

Publications (58)55.81 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: Probes are the interface between microsystems and bio-cells. The ideal interface is one-to-one interface. Though various research groups have been able to establish some sort of interfaces after many years of research, they are very crude. Neurons are millions in numbers, whereas the prostheses successfully built so far have only a few hundred probes at best. Creating an effective interface is still far away. Though we have micro- and nano-technologies, we couldn't build a prosthesis with an effective resolution. Main reasons behind it are the type of probe being used and the poor design of the probe. To address this problem, we developed a methodology to design a probe and an array of probes with better resolution and less resistive donut probe. This methodology helps us to design a probe optimizing all the parameters. We presented our methodology through a design that is capable of 70 μm penetration inside the tissue. The tissue heating by our designed probe is only 0.411 ÛC. We also characterized the donut probe, which could be used by any research group to design a donut probe of their specific need.
    Journal of Bionic Engineering 09/2009; DOI:10.1016/S1672-6529(08)60124-9 · 1.33 Impact Factor
  • M. Akkal, Pepe Siy
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we present a generic sign detection algorithm based on mixed radix conversion algorithm, MRC-II [M. Akkal, P. Siy, A new mixed radix conversion algorithm MRC-II, Journal of System Architecture (2006)] and also we present an optimum algorithm for sign detection based on a special moduli set where mn is even. The described algorithm requires only one comparison for sign detection. A new moduli set will also be presented which simplifies MRC-II conversion algorithm by eliminating the need for table lookup normally used in MRC hardware implementation. The algorithm does not require ROM table like other algorithms. For a moduli set of four moduli that satisfies the special moduli set conditions, 0 tables are needed to do the conversion, while Szabo and Tanaka MRC algorithm [N.S. Szabo, R.I. Tanaka, Residue Arithmetic and Its Application to Computer Technology, McGraw-Hill, New York, 1967] requires 6 tables with a total table size of 4608bits; and Huang MRC algorithm [C.H. Huang, A fully ParallelMixed-radix conversion algorithm for residue number applications, IEEE Transactions on Computers c-32 (4) (1983)] requires 10 tables with a total table size of 3840bits.
    Journal of Systems Architecture 10/2008; 54(10):911-918. DOI:10.1016/j.sysarc.2008.03.003 · 0.69 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: By a retinal prosthesis, the researchers have been able to stimulate a limited number of neurons (around 100) by biphasic current stimulus to reproduce an image spatially by multiplexing technique. For functional restoration of sight (visual acuity 20/80), we are to stimulate at least 2500 neurons/mm(2). We are far behind that target. The time required by biphasic current stimulus, the sequential stimulation by multiplexing technique, and the conventional probes used for stimulation limits our ability to stimulate large number of neurons and maintains the persistence of vision. To address this problem, we have designed high resolution donut probes and used them in designing neural implant chip (NIC) applying parallel multiplexing technique, which helped us stimulate large number of neurons. This paper presents the design of a donut probe and an implantable CMOS-based 5x5 array NIC to demonstrate that idea. This 5x5 array NIC is scalable to any larger array. It uses external clock, which makes it suitable for any data rate. The programmable biphasic width controller (BWC) used in this design is capable of generating various kinds of stimulus for stimulation. NIC doesn't need address to route the pixels, which reduces the data transmission overhead. 0.5 microm CMOS technology is used to fabricate NIC with donut probes. Test results verify our technique of increasing the resolution.
    The Open Ophthalmology Journal 02/2008; 2(1):77-90. DOI:10.2174/1874364100802010077
  • Wei Zhang, Pepe Siy
    [Show abstract] [Hide abstract]
    ABSTRACT: An efficient residue to binary converter design for four moduli set (2 n -1,2 n +1,2 2n -2,2 2n+1 -3) is presented. The converter design is based on New Chinese Remainder Theorem 2 (New CRT II). This moduli set has more dynamic range than previous converter designs for four moduli set. The converter is adder-based and memory-less, enabling this design to achieve more speed and less hardware complexity.
    Information Sciences 01/2008; 178(1). DOI:10.1016/j.ins.2007.05.040 · 3.89 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: We have developed an electrical stimulator and diagnostic research microarray with wireless power and communications to facilitate spatial stimulation of retinal tissue. A third generation 32× 32 prototype of this retinal neural implant array has been developed. Integrated into the microarray is a functionally graded Ti/IrO2 microbump electrode system for interface with neural tissue with decreased impedance for stimulation. The microarray is designed for basic research to determine retinal tissue stimulation thresholds and spatial effects. The array is connected to a telemetry chip, which uses magnetic induction for wireless power with a digital overlay for communication. In our design, changes in the induced current in the telemetry coil are used to send information to the reading coil. Since the reading and telemetry coil are magnetically coupled, the current change can be sensed for bidirectional communication. Combined, this chip set provides a 1024 array that can stimulate neural tissue spatially, can sense neural signals spatially, and has wireless power and communication in a package of less than 2 mm size.
    09/2007: pages 169-186;
  • M. Akkal, Pepe Siy
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we present an efficient and simplified algorithm for the Residue Number System (RNS) conversion to weighted number system which in turn will simplify the implementation of RNS sign detection, magnitude comparison, and overflow detection. The algorithm is based on the Mixed Radix Conversion (MRC). The new algorithm simplifies the hardware implementation and improves the speed of conversion by replacing a number of multiplication operations with small look-up tables. The algorithm requires less ROM size compared to those required by existing algorithms. For a moduli set consisting of eight moduli, the new algorithm requires seven tables to do the conversion with a total table size of 519bits, while Szabo and Tanaka MRC algorithm [N.S. Szabo, R.I. Tanaka, Residue Arithmetic and its Application to Computer Technology, McGraw-Hill, New York, 1967; C.H. Huang, A fully parallel mixed-radix conversion algorithm for residue number applications, IEEE Transactions on Computers c-32 (4) (1983)] requires 28 tables with a total table size of 8960bits; and Huang MRC algorithm (Huang, 1983) requires 36 tables with a total table size of 5760bits.
    Journal of Systems Architecture 09/2007; 53:577-586. DOI:10.1016/j.sysarc.2006.12.006 · 0.69 Impact Factor
  • N. Abu-Khader, P. Siy
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we present a pipelined inversion/division circuit in Galois field using AB circuit technique (where both A and B are elements in the finite field). We use composite Galois fields in a multiple-valued logic (MVL) approach to minimize the inversion/division circuit needed for binary Galois fields. The overall design, which connects basic cells in a systolic manner, thereby making effective use of pipelining, is shown. The fact that less literals are used speeds up the calculation operation. Also, our circuit shows a significant amount of savings in both transistor count and connections, which is so important in VLSI.
    Multiple-Valued Logic, 2007. ISMVL 2007. 37th International Symposium on; 06/2007
  • Source
    M. Abtahi, P. Siy
    [Show abstract] [Hide abstract]
    ABSTRACT: The determination of the sign of RNS (Residue Number System) numbers is not a trivial task. In this work, we have proposed an algorithm called the “Factor-2 Sign Detection Algorithm” for this purpose. This algorithm uses the core value of a RNS number to determine its sign. On the basis of the core values, two areas are defined as ambiguous and unambiguous. The sign of a RNS number with its core in an unambiguous area can be determined immediately but if the core of a RNS number is in an ambiguous area, it may require two to several iterations of the algorithm to determine its sign.
    Computers & Mathematics with Applications 05/2007; DOI:10.1016/j.camwa.2006.03.034 · 2.00 Impact Factor
  • Nabil Abu-Khader, Pepe Siy
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we present two new parallel-in parallel-out systolic array architectures to compute the exponentiation operation for the field GF(2(k)) in a multiple-valued logic (MVL) approach, using the composite field GF((2(2))(m)). We compare both circuits with the circuit that use GF(2(k)) as its basis. The proposed circuits require much less amount of chip area, less clock cycles to generate the final output, and are highly regular, thus they are well suited for VLSI. (c) 2005 Elsevier B.V. All rights reserved.
    Integration the VLSI Journal 06/2006; 39(3):229-251. DOI:10.1016/j.vlsi.2005.06.001 · 0.53 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Probe used for neural stimulation has two parts- the stimulating and return electrodes. Most of the researches are evolving around designing stimulating electrode(s) to stimulate the neurons almost neglecting return electrode. Although some of the research group are using part of the stimulating electrodes as return electrode(s), while others are using a common electrode placed far away from the stimulating electrode as a return electrode. Designing return electrode is as important as the stimulating one is. It has a profound affect on the overall performance of a stimulator. The impedance of a probe is influenced equally by both the electrodes. So careful design of a return electrode can reduce the overall impedance and the required power of a probe dramatically. Here we are presenting a few designs of return electrode and their analysis to find out the best one. We have tested some of the fabricated designs in 0.9% saline solution and found that donut design has less resistance but suffers from high capacitance at higher frequency.
    Bioengineering Conference, 2006. Proceedings of the IEEE 32nd Annual Northeast; 05/2006
  • [Show abstract] [Hide abstract]
    ABSTRACT: Since retinal prosthesis has been able to stimulate a limited number of neurons (around 100) by the biphasic current stimulus to reproduce an image spatially by multiplexing technique. For functional restoration of sight (visual acuity 20/80) we are to stimulate at least 2500 neurons/mm2. We are far behind that target. The time required by biphasic stimulus, and sequential stimulation by multiplexing technique limits our ability to stimulate a large number of neurons, and maintain the persistence of vision. We have designed a stimulus circuit applying parallel multiplexing technique, which is capable of simultaneous stimulation of a large number of cells. This paper presents an implantable CMOS-based 5x5 array prosthesis to demonstrate that idea. This 5x5 array prosthesis is scalable to a larger array. It uses external clock for its timing making it suitable for any data rate. It has a programmable biphasic width controller (BWC), which can generate monophasic or biphasic current stimulus with or without interphase delay. The chip has been fabricated using 0.5 microm CMOS technology and tested.
    Conference proceedings: ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference 02/2006; 1:2816-9. DOI:10.1109/IEMBS.2006.259704
  • [Show abstract] [Hide abstract]
    ABSTRACT: Vital parts of biosensors are the probes. They are the interface between microsystems and bio-cells. The ideal interface is one-to-one interface. Though we have been able to establish some sort of interface after so many years of research, it is very crude. Neurons are in millions in numbers, whereas our successfully built prostheses have only a few hundred probes at best. We are still far from creating effective interface. Though we have micro- and nano-technologies, we couldn't build prosthesis with effective resolution. Main hurdles that we have found are the conventional probes that we use for this purpose. In our study, we have found that donut probe promises a better future. It has 33% better resolution and 84% lesser potential drop than the conventional probes. Detailed comparison of donut probes with other probes is presented here.
    Sensors Applications Symposium, 2006. Proceedings of the 2006 IEEE; 02/2006
  • Source
    Emad Attalla, Pepe Siy
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we are going to present a novel shape similarity retrieval algorithm that can be used to match and recognize 2D objects. The match process uses a new multi-resolution polygonal shape descriptor that is invariant to scale, rotation and translation. The shape descriptor equally segments the contour of any shape, regardless of its complexity, and captures three features around its center including the distance and slope relative to the center. All parameters are normalized relative to the max values. The novel shape matching algorithm uses the shape descriptor and applies it by linearly scanning a stored set of shapes and measuring the similarity using elastic comparisons of shape segments. Similarity measurement is achieved by the sum of differences distance measure. The multi-resolution segmentation provides flexibility for applications that have different time and space requirements while maintaining high accuracy results and the elastic matching adds an advantage when matching partially occluded shapes. We applied our algorithms on many test databases including the MPEG-7 shape core experiment and achieved the highest result reported with a score of 84.33% for the MPEG-7 Part B similarity test.
    Pattern Recognition 12/2005; 38(12-38):2229-2241. DOI:10.1016/j.patcog.2005.02.009 · 2.58 Impact Factor
  • Nabil Abu-Khader, Pepe Siy
    [Show abstract] [Hide abstract]
    ABSTRACT: A quaternary systolic product-sum computation circuit for GF((22)m) using voltage-mode vMOSFETs is presented. The design is composed of four basic cells connected in a pipelined fashion. Each basic cell is composed of 2 Galois field adders, 2 Galois field multipliers, and 7 flip-flops. The circuit was simulated using Affirma Analog Circuit Design Environment tool supplied by Cadence, and it has shown to perform correctly. The quaternary circuit for GF((22)2) shows a significant amount of savings in both transistor count and number of connections compared to the one that uses the binary field GF(24).
    Integration the VLSI Journal 09/2005; 39(1):29-47. DOI:10.1016/j.vlsi.2004.12.001 · 0.53 Impact Factor
  • Source
    M. Abtahi, P. Siy
    [Show abstract] [Hide abstract]
    ABSTRACT: This work is based upon the core function of a RNS (residue number system) number. In determination of the core of a RNS number, an ambiguity problem arises. In this study, we have proposed a new technique named SAS (scaled and shift) to eliminate the ambiguity problem existing in Akushkii's [1] core function. The gain is to compute the core value straightforward without utilizing any other subsystem to detect and remove the ambiguity. Also a new algorithm named WSA (weight selection algorithm) is introduced that gives us the optimum weight set for SAS technique. The optimum weights achieved from WSA provide us with the least complex (smallest possible) weights with the least nonlinearity of the core function.
    Computers & Mathematics with Applications 08/2005; 50(3):459-470. DOI:10.1016/j.camwa.2005.03.008 · 2.00 Impact Factor
  • Wei Zhang, P. Siy
    [Show abstract] [Hide abstract]
    ABSTRACT: The core function provides an efficient way to solve difficult residue number system (RNS) operations. The critical core may cause ambiguity during core function extraction. In this paper, an efficient FPGA design of core function extractor is presented which can form the basis for solving the difficult RNS operations.
    Fuzzy Information Processing Society, 2005. NAFIPS 2005. Annual Meeting of the North American; 07/2005
  • E. Setiaarif, P. Siy
    [Show abstract] [Hide abstract]
    ABSTRACT: Residue number system (RNS) offers a promising future because its carry-free operations in addition, subtraction and multiplication. This inherent property of RNS can be used to reduce the complexity of calculation in many applications, such as encryption and fuzzy systems. However, its adoption is still difficult due to its inherited complex operations such as sign detection and number comparison. We propose a new technique in selecting moduli set, which allows one to find a moduli set with any number of modulus. By using this moduli set, the complicated RNS operations such as sign detection and number comparison can be simplified.
    Fuzzy Information Processing Society, 2005. NAFIPS 2005. Annual Meeting of the North American; 07/2005
  • Source
    N. Abu-Khader, P. Siy
    [Show abstract] [Hide abstract]
    ABSTRACT: A unique encoding technique to perform arithmetic operations in Galois field using multiple-valued (MVL) is presented. The computations are done in a quaternary logic system that can replace the binary one. To show that both systems are equivalent, we illustrate computation examples of inversion and exponentiation in Galois field using AB<sup>2</sup> technique (where both A and B are elements in Galois field). Multi-level representation of signals lies between a hard decision binary system and fuzzy system whose value ranges from 0 to 1.
    Fuzzy Information Processing Society, 2005. NAFIPS 2005. Annual Meeting of the North American; 07/2005
  • M. Abtahi, P. Siy
    [Show abstract] [Hide abstract]
    ABSTRACT: Residue number system (RNS) has been recognized as a robust method to perform computations in a parallel fashion. RNS operations provide us with the capability of solving a precise or fuzzy system using a low resolution multi-moduli system. Despite of the advantages of RNS operations in parallel addition, subtraction, and multiplication, it suffers from some drawbacks such as RNS to binary conversion, sign detection, parity detection, overflow detection, scaling and division by Burgess (1997), and Szaho and Tanaka (1967). Several techniques have been developed to alleviate these drawbacks. For instance MRC (mixed radix conversion) by Szaho and Tanaka (1967), New CRT II (Chinese residue theorem II) by Wang et al. (1998), and core function by Burgess (1997) and Gonella (1991) are some well-known techniques used for RNS to binary conversion. Our study is oriented upon core function. Core function has a non-linear characteristic that causes an ambiguity in RNS to binary conversion as well as sign detection algorithms. Our work is a discussion about the nonlinear characteristic of the core function and its effects on these algorithms. Also, we point to some solutions to alleviate or resolve any ambiguity due to this non-linearity.
    Fuzzy Information Processing Society, 2005. NAFIPS 2005. Annual Meeting of the North American; 07/2005
  • N. Abu-Khader, P. Siy
    [Show abstract] [Hide abstract]
    ABSTRACT: In public key cryptosystems and error-correcting codes over Galois fields, the AB<sup>2</sup> operation is an efficient basic operation. The current paper presents the use of multiple-valued logic (MVL) approach to minimize the systolic architecture of AB<sup>2</sup> algorithm over binary Galois fields. The design is composed of four basic cells connected in a pipelined fashion. The circuit has been simulated using affirma analog circuit design environment tool supplied by Cadence, and it has shown to perform correctly. The quaternary circuit for GF((2<sup>2</sup>)<sup>2</sup>) shows a significant amount of savings in both transistor count and the number of connections compared to the one that uses the binary field GF(2<sup>4</sup>).
    Multiple-Valued Logic, 2005. Proceedings. 35th International Symposium on; 06/2005

Publication Stats

466 Citations
55.81 Total Impact Points

Institutions

  • 1984–2009
    • Wayne State University
      • Department of Electrical and Computer Engineering
      Detroit, Michigan, United States
  • 2005–2006
    • University of Detroit Mercy
      Detroit, Michigan, United States
  • 1986
    • Oakland University
      Rochester, Michigan, United States