Peter A. Galko
University of Ottawa
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Peter A. Galko.
vehicular technology conference | 1994
Homayoun Hashemi; Gang Yun; Mohsen Kavehrad; Farbod Behbahani; Peter A. Galko
In a combination tutorial and research paper, propagation aspects of transmission at infrared (IR) frequencies for wireless in-building communications are explored. The tutorial section of the paper presents basic principles of propagation at IR, a comparison with indoor radio propagation, and the derivation of the channels baseband model. The research aspect of the paper reports on the results of recent frequency response measurements at eight different sites in a university building. A major result shows that the indoor wireless optical channel is very dynamic, with great variations in the channels characteristics for data collected in different rooms, in different locations within the same room, and for different orientations of the optical receiver at the same location of the same room. Numerical values of the channels relative path loss and 3 dB bandwidth, along with frequency response plots covering a wide range of conditions, are presented and discussed. Finally, on the basis of the results of measurements, schemes for improving the performance of future wireless in-building optical transceivers are proposed. >
IEEE Transactions on Communications | 1985
Peter A. Galko; Subbarayan Pasupathy
In this paper we show that many spectrally efficient modified MSK schemes, termed generalized MSK, although not representable as OQPSK, may nevertheless be (suboptimally) demodulated using an I-Q receiver with a proper choice of carrier-phase offset. Correlatively coded MSK schemes with I-Q receivers are studied, and it is concluded that duobinary MSK and (1 + 2D + D^{2})/4 MSK represent good performance-bandwidth tradeoffs among first- and second-order correlative coding polynomial schemes. The optimal design of these receivers are considered subject to the constraint of a finite duration impulse response, especially for asymptotic cases of arbitrarily small and large SNR. Filter design based on a zero-intersymbol interference constraint for PAM-based approximations of the signals is also considered. The optimized linear I-Q receivers for (1 + D)/2 MSK and (1 + D)^{2}/4 MSK are presented. These receivers are only 0.28 and 1.24 dB poorer than the optimal (Viterbi) receivers at high SNR.
IEEE Transactions on Information Theory | 1981
Peter A. Galko; Subbarayan Pasupathy
Many signals ocurring in data communications can be described by Markovian models. Expressions for the mean power spectral density function of a signal driven by a stationary discrete-parameter finite-state Markov chain are derived and discussed, both for irreducible and reducible chains. Two examples are presented to illustrate the application of the results.
IEEE Transactions on Information Theory | 1988
Peter A. Galko; Subbarayan Pasupathy
The optimization of receivers constrained to consist of a (linear) filter and level slicer is considered for general data communication signals in additive white Gaussian noise. Special consideration is given to the two limiting cases of arbitrarily low and high signal-to-noise ratios (SNRs). The low-SNR case results in an average matched filter receiver; the high-SNR filter is shown to be the solution to a specific nonlinear mathematical problem with a simple geometric interpretation. An efficient algorithm is developed to solve the problem. The results are illustrated for a duobinary minimum-shift-keyed signal observed over an interval of three-symbol-period duration. >
international zurich seminar on digital communications | 1994
Homayoun Hashemi; Gang Yun; Mohsen Kavehrad; Farbod Behbahani; Peter A. Galko
In this paper the results of frequency response measurements at eight different rooms in a university building at infrared (IR) frequencies are reported. A major result shows that the indoor wireless optical channel is very dynamic, with great variations in the channels characteristics for data collected in different rooms, in different locations within the same room, and for different orientations of the optical receiver at the same location of the same room. Numerical values of the channels relative path loss and 3 dB bandwidth along with frequency response plots covering a wide range of conditions are presented and discussed. The analysis shows that IR is a suitable candidate for high speed wireless data communication inside buildings.
Archive | 1997
Ahmad Atieh; P. Myslinski; J. Chrostowski; Peter A. Galko
Future optical communication systems will have to provide ultra-high speed transmission rates over long distances. An important limitation for such systems is imposed by the fiber nonlinearity typically described by the coefficient:
canadian conference on electrical and computer engineering | 2008
Walid Abediseid; Peter A. Galko
biennial symposium on communications | 2008
Walid Abediseid; Peter A. Galko
\gamma = \frac{{2\pi {n_2}}}{{\lambda {A_{eff}}}}
1998 International Conference on Applications of Photonic Technology III: Closing the Gap between Theory, Development, and Applications | 1998
Ahmad Atieh; P. Myslinski; J. Chrostowski; Peter A. Galko
international conference on communications | 1982
Peter A. Galko; Subbarayan Pasupathy
(1) where n2is the nonlinear refractive index of the fiber, A eff is the fiber effective core area, and X is wavelength of the propagating signal. To design systems for such applications, it is necessary to determine the ratio n 2/A eff with high accuracy. The nonlinear properties of the fibers have been determined recently by measuring the self-phase modulation (SPM) effect using propagation of a short pulse at 1.3 μn through the tested fibers [1]. A different technique was employed in reference [2], where cross-phase modulation effect between a low power probe signal at 1550 nm and a high power pump signal at 1540 nm was used to find the nonlinear characteristics of the fiber. The ratio n 2/A eff was also measured based on the generation of beating frequency components through four-wave mixing (4WM) of two DFB laser diodes operating around 1555 nm [3]. In this scheme, two cw laser beams with frequencies v1 and v2 are launched into the fiber under test. The two beams with different power levels generate new beat frequencies (2V1-V2) and (2 V2-2V1) through 4WM process. All these schemes measure the nonlinear ratio of the fiber and an independent measurement of the effective core area enables the estimation of the nonlinear refractive index.