#### DMCA

## Extended MLSE Receiver for the Frequency-Flat, Fast Fading Channel (1997)

Venue: | IEEE Trans. Veh. Technol |

Citations: | 5 - 0 self |

### Citations

335 |
An analysis of pilot symbol assisted modulation for Rayleigh fading channels[J].
- Cavers
- 1991
(Show Context)
Citation Context ...iver operates successfully whenever high quality CSI is available. Suitable systems for estimating CSI include pilot tones [1,2] and pilot symbols (when the transmitter pulse shaping is full Nyquist) =-=[4]-=-. The need for CSI is not restrictive, since any high performance receiver must accurately estimate the channel. In the absence of pilot signals, the MLSE receiver structure uses predictors in a Per-S... |

220 | A Statistical Theory of Mobile-Radio Reception”, - Clarke - 1968 |

196 |
Estimation and Modulation Theory, Part I,
- Trees, Detection
- 1968
(Show Context)
Citation Context ...2 1 2 (4) where the overbar denotes complex conjugation. The noise autocovariance is given by ( ) ( ) ( ) ( ) R E n t n t nn t t = + 1 2 (5) and ( ) R nn -1 t is the limiting form of a matrix inverse =-=[16]-=-, so it must satisfy ( ) ( ) ( ) R R nn nn t t d t * = -1 (6) where * denotes convolution. Substituting (1) and (2) into (4) leads to ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) l y z y t h t irT z t R t t y t h ... |

101 |
The characteristic function of Hermitian quadratic forms in complex normal variables
- Turin
- 1960
(Show Context)
Citation Context ...6 a a x y = 1 if x and y are samples from the same symbol period. With these definitions, R gg is completely described. The characteristic function of a Gaussian quadratic form, k u,v,w , is given by =-=[21]-=- ( ) P j u v w u v w x x x , , , = - 1 2 I R G gg (32) which is an inverse polynomial in x, and can expanded as simple partial fractions to obtain ( ) P r p u v w cf i cf i i x x x , , , , , = - - 1 (... |

84 |
Adaptive maximum-likelihood receiver for carrier-modulated data-transmission systems,”
- Ungerboeck
- 1974
(Show Context)
Citation Context ...that assumes ideal channel state information (CSI) is available at the receiver but not the transmitter. The derivation follows Ungerboeck's Extended MLSE receiver for the frequency-selective channel =-=[10] and is kn-=-own as the "Extended MLSE receiver for the time-selective channel," or EMLSE-t. Although the derivation assumes ideal CSI, the receiver operates successfully whenever high quality CSI is ava... |

54 | Detection, Estimation and Modulation Theory, Part 1 - Trees - 1968 |

53 |
Maximum likelihood sequence estimation of CPM signal transmitted over Rayleigh flat-fading channels
- Lodge, Moher
- 1990
(Show Context)
Citation Context ...ive, since any high performance receiver must accurately estimate the channel. In the absence of pilot signals, the MLSE receiver structure uses predictors in a Per-SurvivorProcessing (PSP) structure =-=[7,11,14,15]-=-. These jointly estimate the channel and make decisions. For ease of analysis, these can be reformulated as two stage receivers, where the first stage uses predictors and PSP to provide CSI rather tha... |

15 |
Generating a fading process for the simulation of land-mobile radio communications
- Verdin, Tozer
- 1993
(Show Context)
Citation Context ...transmitter employed an r = 3 multi-sinc pulse, sending differentially encoded QPSK. The fading channel was simulated using filtered white noise, passed through a 192-tap FIR filter, with tap weights =-=[18]-=-, ( ) f J f T l lT l D r r = 1 4 1 4 2p Hanning(lT r ). The analytic curves also used these windowed taps in calculating R zz , R ze and R ee . White noise was added to the sampled signal, obviating t... |

12 |
On the Bit Error Probability of QAM Modulation
- FITZ, SEYMOUR
- 1994
(Show Context)
Citation Context ...t 10 -4 BER. The union bound is particularly loose for 16-QAM due to the large constellation. For 16-QAM, the lower bound considers two disjoint error events. In slow fading, the power penalty is 3dB =-=[17]-=-. Finally, we compare the performance of systems with and without pilot tones. When pilot tones are transmitted, they are exploited; when there is no pilot information, a predictor-based PSP receiver ... |

12 |
On the validity of the slow and moderate fading models for matched filter detection of Rayleigh fading signals
- Cavers
- 1992
(Show Context)
Citation Context ...ot Nyquist pulse, a fast fading process introduces ISI, since in general the channel's complex gain changes along the pulse's length, and the pulse is not simply a scaled version of itself. Reference =-=[19]-=- addresses the boundaries between slow and fast fading. In the time-invariant channel, the ISI is fixed, so Ungerboeck's EMLSE-f receiver is able to precompute s lr,kr . This is not possible in the ti... |

7 |
Mobile digital communications via tone calibration
- Davarian
- 1987
(Show Context)
Citation Context ...ive channel," or EMLSE-t. Although the derivation assumes ideal CSI, the receiver operates successfully whenever high quality CSI is available. Suitable systems for estimating CSI include pilot t=-=ones [1,2]-=- and pilot symbols (when the transmitter pulse shaping is full Nyquist) [4]. The need for CSI is not restrictive, since any high performance receiver must accurately estimate the channel. In the absen... |

5 |
Digital Communications," 2nd Ed
- Proakis
- 1989
(Show Context)
Citation Context ...sonable SNRs, and so does an error event. Thus the upper bound is tight and easily calculated for fast fading and high SNR only. For slow fading, the conventional slow-fading approximations are valid =-=[1,2,4,13]-=-. For BPSK and QPSK, the union of E = 1 error events due the nearest neighbours of the errored symbol only is particularly useful. Clearly, the BER of this nearest neighbours union is asymptotically c... |

3 |
An Adaptive Maximum Likelihood Receiver for Rayleigh Fading Channels
- Dam
- 1990
(Show Context)
Citation Context ...wever they are not optimum, since they only take one sample per symbol and this is below the Nyquist sampling rate. In fact they fail to exploit the implicit Doppler diversity of fast fading channels =-=[9]-=-. This paper derives an optimal receiver configuration that assumes ideal channel state information (CSI) is available at the receiver but not the transmitter. The derivation follows Ungerboeck's Exte... |

2 |
Dual Pilot-Tone Calibration Technique
- Simon
- 1986
(Show Context)
Citation Context ...ive channel," or EMLSE-t. Although the derivation assumes ideal CSI, the receiver operates successfully whenever high quality CSI is available. Suitable systems for estimating CSI include pilot t=-=ones [1,2]-=- and pilot symbols (when the transmitter pulse shaping is full Nyquist) [4]. The need for CSI is not restrictive, since any high performance receiver must accurately estimate the channel. In the absen... |

2 |
D.P.Taylor, "An Adaptive Maximum Likelihood Receiver for Correlated Rayleigh-Fading Channels
- Dam
- 1994
(Show Context)
Citation Context ...g r. In the limit as rs, the multi-Nyquist pulses converge on rect(t/T). Using g(t) = sinc(t/T), the one-sided Doppler-spread bandwidth of the received signal is r/2T+f D . The Dam, Vitetta and Taylor=-=[6,11]-=- receivers take r/T samples per second, so the aliasing is small for slow fading or large r. The error floor of these receivers found for extremely fast fading and low values of r is due to this alias... |

2 |
A.Polydoros and Ching-Kae Tzou, "Per Survivor Processing: a General Approach to MLSE in Uncertain Environments," submitted to
- Raheli
- 1992
(Show Context)
Citation Context ...ive, since any high performance receiver must accurately estimate the channel. In the absence of pilot signals, the MLSE receiver structure uses predictors in a Per-SurvivorProcessing (PSP) structure =-=[7,11,14,15]-=-. These jointly estimate the channel and make decisions. For ease of analysis, these can be reformulated as two stage receivers, where the first stage uses predictors and PSP to provide CSI rather tha... |

2 |
The performance of reference-based Mary PSK with trellis coded modulation in Rayleigh fading
- Chan, Bateman
- 1992
(Show Context)
Citation Context ...s doubled for these receivers also, thus doubling the noise spectral density. For pilot tone CSI, we arbitrarily choose r a D f T = + 1 1 2 which is asymptotically optimum for r = 1 sample per symbol =-=[20]-=-. For the predictor-based receivers, four curves are plotted in Figures 6 and 7, namely . the analytic lower bound, assuming coherent QPSK. . twice the analytic lower bound, to approximate the effect ... |

1 | M.K.Simon, "The Design of Trellis Coded MPSK for Fading Channels: Performance Criteria - Divsalar - 1988 |

1 | Comments on Tone Calibration in Time-Varying Fading - Fitz - 1992 |

1 |
D.P.Taylor, "Maximum Likelihood Detection of Differentially Encoded PSK Signals Transmitted over Rayleigh, FrequencyFlat Fading Channels
- Vitetta
- 1994
(Show Context)
Citation Context ...l properties are quite different except in very slow fading. In fast fading, the pulse distortion due to the multiplicative channel leads to non-root-Nyquist pulses, ISI, and thus a BER floor. Others =-=[7,8]-=- have employed full-Nyquist pulse shaping at the transmitter, and a zonal filter at the receiver. These receivers are effective, in that there is no ISIinduced error floor. However they are not optimu... |

1 |
D.P.Taylor, "Maximum Likelihood Sequence Estimation of Uncoded and Coded PSK Signals Transmitted over Rayleigh, Frequency-Flat Channels," Supercomm/ICC
- Vitetta
- 1994
(Show Context)
Citation Context ...ive, since any high performance receiver must accurately estimate the channel. In the absence of pilot signals, the MLSE receiver structure uses predictors in a Per-SurvivorProcessing (PSP) structure =-=[7,11,14,15]-=-. These jointly estimate the channel and make decisions. For ease of analysis, these can be reformulated as two stage receivers, where the first stage uses predictors and PSP to provide CSI rather tha... |