Bernd-Peter Paris
George Mason University
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Bernd-Peter Paris.
IEEE Journal on Selected Areas in Communications | 2003
Song Xing; Bernd-Peter Paris
Measuring the size of the Internet via Monte Carlo sampling requires probing a large portion of the Internet protocol (IP) address space to obtain an accurate estimate. However, the distribution of information servers on the Internet is highly nonuniform over the IP address space. This allows us to design probing strategies based on importance sampling for measuring the prevalence of an information service on the Internet that are significantly more effective than strategies relying on Monte Carlo sampling. We present thorough analysis of our strategies together with accurate estimates for the current size of the Internet Protocol Version 4 (IPv4) Internet as measured by the number of publicly accessible web servers and FTP servers.
international conference on acoustics, speech, and signal processing | 1997
Bernd-Peter Paris; Geoffrey C. Orsak; Hongda Chen; Nirmal C. Warke
The problem of distinguishing reliably between signaling formats in the presence of noise, interference, unknown dispersive channel conditions, as well as timing and frequency mismatches is addressed. Methods based on a combination of blind equalization and universal classification are presented and their performance is assessed through simulations.
IEEE Transactions on Communications | 1996
Bernd-Peter Paris
The design of single-user decorrelating receivers employing finite-precision sequences for despreading is considered. The problem is formulated as a nonlinear bounded integer optimization problem which is shown to be network performance (NP)-hard. A branch-and-bound algorithm for finding the best finite-precision decorrelating sequence is described. Numerical examples demonstrate that the loss in performance between the optimum, infinite-precision, and the best finite-precision decorrelator is small even for large channel occupancies. Some suboptimum algorithms are investigated which greatly reduce the computational complexity associated with finding good finite precision decorrelator sequences.
global communications conference | 1993
Bernd-Peter Paris
The problem of estimating the most likely state sequence of a discrete-time finite-state Markov process with unknown parameters observed in independent noise arises in many important problems in digital communications, including self-adaptive equalization and adaptive multi-user detection. A maximum likelihood criterion over both the input sequence and the parameters is introduced for estimating the state sequence without using an embedded training sequence. Asymptotically, this estimator is close to the maximum-likelihood sequence estimator with completely known parameters. To facilitate the search for the most likely state sequence, we introduce computationally simple algorithms which are guaranteed to converge. Performance of the self-adaptive maximum-likelihood sequence estimator for the blind equalization problem is illustrated through numerical examples.<<ETX>>
international conference on computer communications and networks | 2003
Song Xing; Bernd-Peter Paris
The Internet is growing rapidly, causing concerns that the address space for IPv4 may soon be used up. Taking a snapshot of the Internet size can help in planning the future evolution and capabilities of the Internet as well as planning the implementation of the next generation IPv6. We extended our previous work on measuring the number of publicly accessible web servers to make more accurate measurements of the size of the Internet. An improved importance sampling approach is introduced which achieves a significant gain over Monte Carlo methods. The growth of the Internet is mapped by periodic measurements of the number of active web servers.
conference on information sciences and systems | 2007
Shyam Pandula; Bernd-Peter Paris
This paper considers spatial multiplexing (SM) systems with preceding. The precoder is derived from the singular value decomposition (SVD) of the available channel state information at the transmitter (CSIT) and the receiver is a function of the precoder and the current channel. With perfect CSIT, the MRxMT flat-fading MIMO channel can be decomposed into mm(MT,MR) parallel spatial subchannels. However in practice, the available CSIT suffers from delay-induced error due to the channel temporal variations. Using this outdated CSIT for precoding in SM systems causes interference among the subchannels. Performance of the decorrelator, minimum mean squared error (MMSE) and successive interference cancelation (SIC) receivers is analyzed as the reliability of the available CSIT varies. Explicit expressions for the signal to interference-plus-noise ratio (SINR) and the mean squared error (MSE) are derived. Simulation results are provided to illustrate the significant performance gain achieved by precoding even with a moderate amount of correlation between the available outdated channel estimate and the current channel.
international symposium on information theory | 1994
Bernd-Peter Paris
Linear decorrelating receivers for direct-sequence spread spectrum multiple access communication systems appear to be the most promising candidates for practical implementations of multi-user detectors. Decorrelating receivers compare favorably to other receiver structures because they are optimally near-far resistant, they do not require knowledge of the received signal amplitudes, and their computational complexity is on the same order as that of the conventional matched filter detector. However, in general optimum single user decorrelating receivers correlate the received signal with a sequence of real numbers and complexity comparisons did not reflect the need for a floating point multiplier in optimum decorrelating receivers. The authors focus on the problem of demodulating a single users signal in a multi-user environment. More specifically, they investigate the question whether it is possible to design single-user receivers which are identical in structure and complexity to the conventional matched filter receiver.<<ETX>>
conference on information sciences and systems | 2008
Shyam Pandula; Bernd-Peter Paris
This paper addresses the capacity optimization problem for MIMO wireless channels with a non-zero mean and transmit antenna correlation. With a decorrelator receiver, the capacity of the MIMO system is a function of the diagonal elements of an inverted noncentral Wishart distributed matrix. Hence, finding the average capacity is difficult. We simplify the problem by approximating the SNR of each spatial stream by a standard noncentral Chi-squared random variable. Using the moments of the SNR, we obtain a Taylor series approximation for the average capacity that is significantly better than the commonly used bound via Jensens inequality. The obtained Taylor series approximation is used to design a linear precoder that maximizes the total average capacity of the system. Simulation results are provided to illustrate the performance gain.
international conference on digital signal processing | 2006
S.G. Wood; Bernd-Peter Paris; Jill K. Nelson
We present a computationally efficient blind equalization technique to mitigate the effects of inter-symbol interference (ISI). Our technique is based entirely on context sorting of received ISI corrupted samples of a communications system. Techniques based on rank sorting and successive rank sorting provide context based analysis of the underlying state transitions of a finite memory source channel. We are able to exploit this information to effectively perform blind equalization and moreover recover the transmitted symbols of an ISI corrupted data stream
international symposium on information theory | 1998
Bernd-Peter Paris; Hongda Chen
A novel nonlinear blind equalizer for linearly modulated digital communication is proposed. The equalizer employs a ring decision device that feeds back a weighted sum of past ring decisions to approximately cancel intersymbol interference (ISI) in the present signalling interval. Simple adaptive algorithms for blindly adapting the coefficients of the equalizer are presented.