Showing papers on "Fading distribution published in 1993"

Rayleigh fading compensation for QAM in land mobile radio communications

Abstract: A pilot symbol-aided Rayleigh fading compensation is investigated for M-ary quadrature amplitude modulation (QAM) to achieve highly spectrally efficient land mobile communication systems. The optimum parameters for fading compensation, bit error rate (BER) performance against E/sub b//N/sub 0/ (energy per bit to the noise power spectrum density), adjacent channel interference, and cochannel interference for 16-QAM, 64-QAM, and 256-QAM, and the spectral efficiencies for these modulation schemes in Rayleigh fading environments are investigated by computer simulation. To further verify the effect of pilot symbol-aided fading compensation from a a practical point of view, a 16-QAM modem is implemented, laboratory experiments are executed, and the impact of the dynamic range limitation due to the resolution of the analog-to-digital (A/D) converters is evaluated, along with the imperfection of the analog circuits. It is demonstrated by computer simulation and laboratory experiments that the pilot symbol-aided fading compensation can sufficiently compensate for fast varying Rayleigh fading, and 16-QAM gives the highest spectral efficiency in the case of cellular systems. >

Effect of fading correlation on adaptive arrays in digital wireless communications

Abstract: The effect of correlations among the fading signals at the antenna elements of an adaptive array in a digital wireless communication system is investigated. With an adaptive array, the signals received by multiple antennas are optimally weighted and combined to suppress interference and combat desired signal fading. A model of local scattering where the received multipath signals arrive within a given bandwidth is presented, and a closed-form expression for the correlation as a function of antenna spacing is derived. Results show that the degradation in performance with correlation in an adaptive array that combats fading and suppresses interference is only slightly larger than that for combating fading alone, i.e., with maximal ratio combining. This degradation is small, even with correlation as high as 0.5. >

Co-channel interference of microcellular systems on shadowed Nakagami fading channels

Abstract: The probability of cochannel interference is evaluated for a microcellular environment with log-normal shadowing and Nakagami fading. For the case when the interfering signals have the same shadowing and fading statistics, an exact mathematical expression is presented for the probability of cochannel interference. For the case when the interfering signals have different shadowing and fading statistics, the probability of cochannel interference is evaluated by approximating the individual interfering signals as having a log-normal distribution, and then applying Schwartz and Yeh's method to obtain a purely log-normal distribution for total interference, while maintaining a composite Nakagami log-normal distribution for the desired signal.

Coded Modulations for Fading Channels: An Overview

Abstract: It is by now well known that combined channel coding and digital modulation (in short coded modulation) yield power and bandwidth efficient digital transmission schemes for the Gaussian channel. These schemes perform closer to channel capacity than those where coding and modulation are treated separately. More recently, coded modulation methods have been investigated for transmission over Rayleigh fading channels. It has been discovered that systems optimized for the Gaussian channel are not optimum for fading channels. This is primarily due to the importance of built-in time diversity of the coded modulation scheme for fading channels. In this paper we give a brief overview of recent developments in signal design for Rayleigh fading channels. Mainly schemes based on phase shift keying are presented. Theoretical guidance for design and identification of the most important system parameters are given. A couple of detailed examples of code designs and their error probability performance are also given. Optimum and suboptimum receivers are discussed. An overview of analytical performance analysis tools, as well as code designs for both trellis coded modulation and block coded modulation is included in the paper. Methods for unequal error protection are given. Potential application areas are identified. An extensive reference list is supplied.

Innovations-based MLSE for Rayleigh fading channels

Abstract: Use of the innovations approach for developing MLSE (maximum likelihood sequence estimation) algorithms for signals transmitted over Rayleigh fading channels is considered. The derived receiver demodulates the signals recursively in a noncoherent fashion and is applicable to any linearly modulated signals and to both frequency-nonselective and frequency-selective Rayleigh fading channels. The algorithms can be implemented by a bank of FIR (finite impulse response) time-invariant filters followed by a VA (Viterbi algorithm) processor. A unified interpretation of MLSE of signals is obtained by using the innovations approach. >

Characteristics of a simulated fast fading indoor radio channel

Abstract: The channel characteristics of an indoor radio channel vary very slowly. A simple technique that introduces a fast fading effect on the channel is presented. This scheme requires two or more antennas at the transmitter and has to be used with a suitable channel code. Slow time varying phase offsets are provided to each transmit antenna. Because of the resulting fast fading effect, with this scheme, time diversity can be employed usefully in a coded system. The indoor radio channel is represented by a slow fading Rayleigh channel model, and analytical expressions are provided for the correlation of the received signal in the presence of this simulated fast fading scheme. Expressions are derived for the level crossing rate and the average fade duration of the envelope of the received signal.

RAKE receiver structures for differential and pilot symbol-assisted detection of DS-CDMA signals in frequency-selective Rayleigh fading channels

Abstract: The capacity of a CDMA system is a function of the bit error rate performance of individual users. Therefore it is important to optimize the individual links before proceeding to system level analysis. This is particularly true for operating in a fading channel where the performance without diversity reception is rather poor. This paper compares the BER performance of differential detection and pilot symbol-assisted coherent detection of a direct-sequence spread spectrum signal on a frequency-selective Rayleigh fading channel using RAKE reception. Both equal gain and maximal ratio combining are considered, and the effect of convolutional coding with interleaving is studied. >

The cutoff rate of time correlated fading channels

Abstract: The cutoff rate of Rician and Rayleigh channels characterized by time correlated fading is derived for M-ary phase shift keying (MPSK) modulation. It is useful for situations where perfect interleaving cannot be achieved. It indicates the practical achievable information rate of the channel when coding is employed. It is used to determine the potential coding gains of coded modulation over the correlated fading channel. >

Diversity improvement for 16-DAPSK in Rayleigh fading channel

Abstract: Postdetection combining has been analysed together with fully differential 16-APSK (16-DAPSK) as a potential method for improving digital transmission in a frequency nonselective Rayleigh fading channel. Results are presented for the theoretical and computer simulated BER performance both with and without the use of diversity combining in the receiver.

Average error probability for quadriphase modulated DS-SSMA communications through Nakagami fading channel

Abstract: An asynchronous direct sequence spread spectrum multiple access system using quadrature phase shift keying modulation through Nakagami'sm-distributed fading channel is considered for nondiversity reception. Approximation to the average error probability is evaluated in two steps. Using the Gauss quadrature rule, the moments of the multiple interferences are used to evaluate the conditional probability conditioned on a fixed fading amplitude of the desired signal. The probability density function of the desired signal is Nakagami distributed. The average error probability which is the expected value of the conditional probability is evaluated by the trapezoidal integration method. This method provides a good approximation to the average error probability for a small to a fairly large number of users. Numerical results are presented for a set of Gold code of code length 127.

A Comparison of Differentially Coherent and Coherent Multiuser Detection with Imperfect Phase Estimates in a Rayleigh Fading Channel

Abstract: Multiuser detectors have superior performance over their singleuser counterparts in a multiple-access channel, assuming perfect knowledge of system parameters [l-31. In this paper we extend the analysis of multiuser detectors in fading channels by incorporating the effects of imperfect parameter estimates on symbol error probability. This type of analysis should be useful in designing multiuser receivers, showing the error rate sensitivity to channel parameter mismatch. We focus on a synchronous CDMA channel shared by I< users where the signal of each user arrives at the central receiver through an independent flat Rayleigh fading cha.nne1. The central receiver has the knowledge of the signature waveforms of all users, and the outputs of a matched filter ba.nk provide the sufficient statistics. We also assume that a state space description of the fading distortion is available. In a single-user situation an optimum receiver structure in the flat fading channel [4] consists of an adaptive estiinator of fading distortion, and a detector which utilizes these estimates. The estimation of the complex channel distortion performs the ta.sk of the carrier recovery. Given the state-space model of the channel distortion, the Kalman filter is the optimal, minimuin variance state estimator. A suboptimum, realizable receiver can be implemented using the decision-directed approach where the data dependence is removed from the matched filter output. Multiuser carrier recovery can be accomplished in two ways. The first approach is proposed in [ 5 ] in which the matched filter outputs are decorrelated and each user employs phase estimators which assume isolated tra.nsmission. In this case the multipleaccess interference is removed from the matched filter outputs a t the expense of noise enhancement and correlation, which affects the performance of the carrier recovery circuit. We also consider the vector generalization of the receiver proposed for the single-user channel [4]. Due to synchronism among the users, the data dependency can be removed in a decision-directed manner and the joint phase estimates are obtained by using a multi-input multi-output Kalman filter. We focus our analysis on two low-complexity suboptimum multiuser detectors, the coherent and differentially coherent decorrelating detector. The coherent decorrelating detector utilizes phase estiniates obtained by the aforementioned carrier recovery techniques. In this case, the probability of error can be calculated using Stein’s unified analysis [7]. Assuming perfect symbol phase elimination, the lower hound on the error probability is given by

Evaluation of channel capacity in a generalized fading channel

Abstract: Channel capacity in a Nakagami fading channel is evaluated. The probability density function of the channel capacity and its mean and variance are derived in order to examine channel fading effects. The diversity combining effect and the cochannel interference effect on channel capacity are studied.

Adaptive correlator receiver performance in fading multipath channels

Abstract: The behavior of a least mean square (LMS) adaptive correlator operating on a fading multipath channel is studied. The channel is modeled as a time-varying tapped delay line transversal filter. The adaptive correlator is seen to be effective for slowly varying channels. Its performance degrades in fast fading channels.

Performance of coherent data transmission in frequency-selective Rayleigh fading channels

Abstract: The error probability of digital transmission through a frequency-selective channel with random multipath fading due to uncorrelated scatterers is calculated. The error probability is averaged with respect to the randomly varying delays, with a given delay power spectrum. Two coherent receivers, suboptimum without any knowledge of the channel and optimum with perfectly known channel, are compared. The results are applied to continuous-phase-modulation (CPM) signals. >

Receivers for CDMA communication over time-varying Rayleigh fading channels

Abstract: We study multiuser receiver design and analysis for synchronous CDMA channels with time-varying Rayleigh fading. Starting from an error probability criterion, we first derive a receiver for this channel that admits a detector-estimator decomposition, has certain asymptotic optimality properties and a complexity which is horizon independent. The performance of this detector is analytically characterised and contrasted with that of the single-user (or conventional) detector as well as the optimal multiuser detector for the time-invariant CDMA Rayleigh fading channel when it is implemented over the time-varying channel. Notable among our conclusions is the fact that, unlike the time-invariant channel multiuser detector, the time-varying channel detector is able to withstand not only the estimated interference from the other system users, but also, the residual (unestimatable) interference arising out of imperfect estimation of the interferer fading parameters. >

Pilot symbol assisted modulation in fading and delay spread

Abstract: The performance of pilot symbol assisted modulation (PSAM) is analyzed in intersymbol interference (ISI) due to frequency selective fading. PSAM is shown to be sensitive to RMS delay spread, though it always gives better performance than differential detection.

Capacities of hybrid FDMA/CDMA systems in multipath fading

Abstract: A hybrid FDMA/CDMA scheme is described and analyzed. The performance of the hybrid system is then compared with that of a wideband CDMA system occupying the same total bandwidth. It is found that, while the hybrid system has properties which may be desirable in certain situations, such as the flexibility of using a non-contiguous spectrum, in a multipath fading environment a wideband CDMA system has greater capacity than the hybrid. >

Performance of a multidimensional BCM scheme with fading estimation based on time correlation

Abstract: The combination of the channel fading compensation technique and the coded modulation technique is discussed. The system presented is considered to operate under correlated Rayleigh or Rician fading channels. The channel correlation property has been used to carry out channel fading compensation with the aid of periodically inserted pilot symbols. A multidimensional BCM (block coded modulation) scheme was used to improve the system error performances. On decoding, a corresponding trellis is used; therefore, the Viterbi algorithm and channel state information become available. The error performances of the system are investigated through computer simulations. Combined with the fading compensation, the multidimensional BCM scheme gives good error performances under various fading channels. For the Rayleigh fading channel, 7.5-dB coding gain has been achieved at a bit error rate (BER) of 10/sup -3/. For the Rician fading channel with K=10 dB, the coding gain is 3.7 dB at BER=10/sup -4/. >

Fading compensation device

Abstract: PURPOSE:To attain estimate of fading at high accuracy by estimating fading fluctuation through the use of not only fading fluctuation in a pilot symbol but also fading fluctuation in an information symbol. CONSTITUTION:A pilot symbol detection part 6 detects a pilot symbol from a reception signal converted into a base band at an orthogonal demodulation part 3 and sampled by an A/D converter 4. A fading estimate part 7 estimates fading fluctuation in an information symbol based on a pilot symbol detected by the pilot symbol detection part 6 and a re-modulation result obtained from the discrimination result at a re-modulation part 10. The fading fluctuation in the information symbol is compensated by using an estimate value 24 of the fading fluctuation in the information symbol obtained by the estimate part 7 and multiplying its inverse characteristic with a reception signal to obtain a complex base band signal 25 in which the fading fluctuation is compensated. Then the complex base signal is regenerated from the signal discriminated by a discrimination part 9 in the rem-modulation part 10.

Statistics of sampled Rician fading

Abstract: : Realizations of the channel impulse response function generated with Rician amplitude statistics have been used for many years to evaluate system performance in the regime between full Rayleigh fading and ambient non-fading conditions. While it is well known that Rician distributed realizations fail to match observed phase fluctuations, Rician statistics provide a worst case description of the occurrence of deep fades. Thus the use of Rician statistics is attractive in slow fading situations where system performance has been rendered insensitive to phase variations. The purposes of this report are to provide new information on the temporal statistics of Rician fading (mean fade duration and separation) and to determine sampling requirements on realizations of the channel impulse response function with Rician amplitude statistics. It is found that 400 decorrelation time tau sub 0 realizations generated with 10 samples per tau sub 0 and sampled at (tau sub 0)/40 accurately reproduce the temporal statistics of Rician fading.. Temporal statistics, Scintillation, Rician fading, Rayleigh fading.

Low-complexity and high-performance multilevel coded modulation for the AWGN and Rayleigh fading channels

Abstract: The multilevel coding method is a powerful technique for constructing bandwidth efficient modulation codes systematically with arbitrarily large distance parameters from Hamming distance component (block or convolutional) codes in conjunction with proper bitsto-signal mapping through signal-set partitioning. The multilevel modulation codes constructed by this method allow the use of multi-stage decoding procedures that provide good trade-off between performance and decoding complexity.

Adaptive methods for equalization of rapidly fading channels

Abstract: Basis expansion ideas are employed in order to equalize frequency-selective, rapidly fading channels. It is shown that for certain fading channels (like the mobile radio channel), the coefficients' time variation can be modeled as a combination of a small number of complex exponentials, which leads to novel adaptive and decision-feedback equalization algorithms. By interpreting ideas from cyclostationary signal analysis, batch and recursive methods are also developed for estimating the frequencies of the exponentials. Finally, some illustrative simulations are presented. >

Effect of multipath fading on millimetre wave propagation: a field study

Abstract: Analysis of clear air fading encountered on millimetric wave and infrared radio links are presented. The analysis is based on data obtained during a four-year field study in the city of Riyadh, Saudi Arabia. The region can be con- sidered as a typical arid climate where the rate of evaporation is higher than the rate of precipi- tation. Maximum rain rate is of the order of 30 mm/h for 0.001% of the year. A brief descrip- tion of the experimental setup is presented, together with the results of measuring clear air fading experienced by the radio links. Statistical characterisation of fading is given for both the millimetric wave links operating near 40 GHz and the infrared radio link at a wavelength of 0.88 pm. It is shown that fading is dominated by multipath, having a Rayleigh amplitude distribution and an occurrence factor similar to the microwave band. Fades were highly correlated for the two links sharing the same path and separated in frequency by 1%. Multipath fading was also measured on the infrared link. Even at the short hop length at 0.75 km, the occurrence factor was about 1%. Time duration of fades are also analysed, and fade durations were essentially exponentially distrib- uted. Small fades, however, have normal duration distribution.

16-state nonlinear equalizer for IS-54 digital cellular channels

Abstract: A simple receiver that consists of a Nyquist filter matched to the transmitter filter followed by a 16-state sequence estimator has been considered for a two-beam, frequency-selective fading channel with inter-beam delay up to one symbol. 16-state sequence estimator compensates for the two-beam fading model that represents the channel between transmitter and receiver. Performance has been evaluated for perfect channel state information regarding fading and was found to be much superior to using a linear equalizer rather than a 16-state sequence estimator. The results compare favorably with earlier results that did not involve Nyquist pulse shaping. North American Digital Cellular fading parameters have been used in the performance analysis. Results are presented for both root-raised-cosine filtered /spl pi//4-DQPSK (differential quadrature phase shift keying) and QPSK modulation formats. >

Maximum-likelihood sequence estimation for fading channels via the EM algorithm

Abstract: We investigate the use of the expectation-maximization (EM) algorithm to the problem of estimating sequences of two-dimensional constellation signals transmitted over a Rayleigh fading channel. The resulting iterative algorithm does not require a carrier recovery loop, and is shown to have a simple linear form. Simulation results for binary signaling indicate that it performs significantly better than previously proposed algorithms. >

A coherent demodulator for fast-frequency-hopping-spread-spectrum signal employing an adaptive algorithm

Abstract: The authors propose a coherent demodulator for a fast-frequency-hopping (FFH)-spread-spectrum signal in a frequency selective fading channel, which has been considered impossible to realize, employing a method in which the received signal is noncoherently downconverted and phase-adjusted by an adaptive algorithm. Computer-simulation shows that the proposed system brings great improvement to bit-error-ratio performance.

Comments on multi-carrier and single-carrier digital modulation in a multipath radio channel

Abstract: An analytical comparison of single-carrier (SC) and multi-carrier (MC) transmission is presented for binary signalling on slow fading mobile radio channels defined by Rayleigh, Rician, and one-sided normal distributions. For the same rate of transmission and signal bandwidth, it is demonstrated that the SC transmission system is superior to the MC system on the slow fading multipath channel. >

A performance analysis of trellis coded modulation schemes in non-independent Rician fading channels

Abstract: The paper presents a new upper bound on the pairwise error probability (PEP) of trellis-coded modulation (TCM) schemes over nonindependent Rician fading channels. The bound is quite general, e.g., applicable to ideal coherent and pilot-tone aided detection, and differential detection, etc. Being quite accurate, the bound can be used in conjunction with a truncated union bound to estimate the bit error probability when ideal interleaving is not possible. >