Delayed Decision Feedback Sequence Estimation Research Articles

Iterative Processing Structure for the Single-Channel Mixture of Digital-Modulated Adjacent-Frequency Source Signals

Considering the high cost and complexity of multiple-channel blind source separation, single-channel solutions have elicited a significant amount of interest in source separation. Joint maximum likelihood sequence estimation implemented by the Viterbi algorithm (VA) is a seemingly ultimate solution to the problem of data transmission over dispersive channels. However, both joint reduced-state sequence estimation (JRSSE) and joint delayed-decision feedback sequence estimation (JDDFSE) have numerous shortcomings when applied to single-channel blind signal separation. Hence, joint double-direction DDFSE (JDD-DDFSE) is proposed in this article. JDD-DDFSE employs joint VA (JVA) of the low-order channel twice. The estimated symbols from the initial JVA are utilized to equalize the causal and non-causal taps of the channel beyond the trellis state in the second JVA. A method that can derive the log likelihood ratio of the estimated symbols is provided; the complexity of this method is considerably lower than that of soft-output VA or the BCJR algorithm. Hence, the entire computation complexity of the iterative separation structure basically remains unchanged compared with that of the no-iteration structure. For the non-causal channel, the proposed algorithm has the lowest computation complexity and demonstrates a separation performance similar to that of conventional JVA, JRSSE, and JDDFSE algorithms.

Nonlinear Equalization Approaches for Physical Layer Network Coding

We consider a two-way relaying system employing physical layer network coding in channels suffering from frequency-selective fading. We study decision-feedback equalization (DFE), a technique based on delayed decision-feedback sequence estimation (DDFSE), and Tomlinson-Harashima precoding (THP) approaches for mitigating the distortions introduced by the channel. For DFE, we introduce transmit filtering schemes that generate identical overall source-to-relay channel impulse responses for both source nodes, while achieving the maximum signal-to-noise ratio after equalization for the zero-forcing and the minimum mean-squared error criterion, respectively. The advocated DDFSE approach also relies on transmit filtering at the source nodes. For the THP scheme, we derive the optimal precoding filters that guarantee an intersymbol interference free source-to-relay transmission. In order to obtain a two-way relaying system, which uses the same equalization techniques in both transmission directions, we design a THP-based compromise precoding scheme for the relay-to-destination transmission. The performance of the proposed techniques is compared with that of benchmark schemes with simpler filtering/precoding, where identical overall source-to-relay channel impulse responses are enforced by straightforward but suboptimum choices for the transmit filters. Our results reveal that the developed schemes enable significant gains compared with the benchmark schemes.

Iterative Equalization With Soft Feedback With a Subsequent Stage Employing Error Search and Correction

In this paper, a second stage for algorithms applying iterative soft decision interference cancellation (ISDIC) for channel equalization is proposed, which performs error search and correction. Analysis by simulations shows that matched filter (MF)-based ISDIC (MF ISDIC) with subsequent second stage can outperform more complex minimum-mean-square-error (MMSE) based ISDIC. Furthermore, the latter can be improved by the proposed second stage by up to 2 dB. It is shown that the ISDIC schemes with subsequent second stage can reach the MF bound up to a fraction of a decibel for all analyzed channels and 4-ary quadrature amplitude modulation (QAM) transmission, and up to 1 dB for 16-ary QAM transmission. In the work of Gerstacker , it has been reported that MF ISDIC performs very well for highly dispersive channels, e.g., with 60 symbol-spaced taps, but for channels of moderate dispersion, an error floor occurs. By utilizing performance bounds for delayed decision-feedback sequence estimation (DDFSE), we show that now for channels of moderate length, e.g., 20 symbol-spaced taps, even low-complexity MF ISDIC with the proposed subsequent second stage outperforms a DDFSE approach with states.

Single antenna interference cancellation (SAIC) for cellular TDMA networks by means of joint delayed-decision feedback sequence estimation

In this letter, single antenna co-channel interference cancellation for cellular time-division multiple access (TDMA) networks by means of joint delayed-decision feedback sequence estimation is studied. The performance is increased by a novel adaptive state allocation technique.

Decision feedback MLSE for spatially multiplexed MIMO frequency selective fading channel

A new decision feedback maximum likelihood sequence estimation (DF-MLSE) scheme for spatially multiplexed MIMO frequency selective fading channel has been proposed. A single trellis processor is required in the proposed receiver instead of multiple NT parallel processors as required in existing low complexity receivers, where NT is the number of transmitter antennas. The bit error rate (BER) performance of the proposed algorithm is the same as standard MLSE, but the computational complexity has been reduced significantly. For a model MIMO system (four transmitter and four receiver antennas) the computational complexity has been reduced by 95% in comparison to standard MLSE exploiting the proposed detection algorithm. Unlike the decision feedback equaliser (DFE) and the delayed decision feedback sequence estimation (DDFSE), the proposed algorithm has no signal-to-noise ratio (SNR) penalty while achieving the same BER performance as standard MLSE.

Low-Complexity Viterbi Equalizer for MBOK DS-UWB Systems

This paper presents a low-complexity equalization for M-ary biorthogonal keying based direct sequence ultra wideband (MBOK DS-UWB) systems. We focus on a Viterbi equalizer, which is based on maximum likelihood sequence estimation (MLSE). To reduce the computational complexity of MLSE-based equalizer, we use two strategies. One is the use of delayed-decision feedback sequence estimation (DDFSE), which is a hybrid estimation between MLSE and decision feedback estimation (DFE). And the other is the truncation of state transition in MLSE by considering MBOK pulse mapping. The reduced complexity sequence estimation is named as reduced state (RS)-DDFSE. By the use of RS-DDFSE, the complexity of Viterbi equalizer for MBOK DS-UWB is significantly reduced, by comparison with that of MLSE. The performance of RS-DDFSE based equalizer is evaluated on multipath fading channel models provided by IEEE802.15.3a. An analysis on trellis diagram of RS-DDFSE and simulation results show that the impact on error rate performance generated by the complexity lower is slight.

Reduced-state MIMO sequence detection with application to EDGE systems

In this paper we propose a joint reduced-state sequence detector (JRSSD) for a multiple-input multiple-output (MIMO) system. The proposed JRSSD incorporates the set-partitioning principle to obtain a reduced-state trellis and is the space-time extension of the RSSD proposed in for a single-input single-output (SISO) equalization problem. We show that two elements are essential in achieving the desired complexity and performance tradeoff for the proposed JRSSD algorithm: 1) a proper multisymbol set-partition and 2) an efficient space-time structure that effectively decouples the spatial and temporal processing. We propose a simple multisymbol uniform set-partition (USP) that retains certain geometric symmetry in the partitioned subsets. We also develop two suboptimal algorithms, namely the decorrelating (DC) and ordered successive (OS) algorithms, to decouple the spatial multisymbol detection problem inherent in the processing of parallel transitions, into single-symbol detection problems. The symmetry in USP, together with the DC or OS algorithm, guarantees efficient processing of parallel transitions and leads to a low-complexity JRSSD. Furthermore, numerical simulations show that the performance of JRSSD is near that of the more complex delayed decision feedback sequence estimator (DDFSE).

Spatio-Temporal Equalization for Space-Time Block Coded Transmission over Frequency Selective Fading Channel with Co-channel Interference

In this paper, we propose a spatio-temporal equalizer for the space-time block coded transmission over the frequency selective fading channels with the presence of co-channel interference (CCI). The proposed equalizer, based on the tapped delay line adaptive array (TDLAA), performs signal equalization and CCI suppression simultaneously using the minimum mean square error (MMSE) method. It is to show that our scheme outperforms the previous two-stage combined adaptive antenna and delayed decision feedback sequence estimator (DDFSE) approach. We also show that performance can be further improved if the synchronization between the preceding and delayed paths is achieved.

Layered space-time equalization for wireless mimo systems

We investigate layered space-time equalization (LSTE) architectures for multiple-input multiple-output (MIMO) frequency-selective channels. At each stage or layer of detection, a MIMO delayed decision feedback sequence estimator (MIMO-DDFSE) is used to tentatively detect a group of selected data streams, among which a subgroup of data streams are output and are canceled from the received signals. With the proposed architectures, the numbers of the tentatively detected data streams and output data streams can be different at different LSTE stages, while the MIMO-DDFSE can also reduce to the special cases of multiple-input single-output decision feedback equalizer (MISO-DFE), MISO-DDFSE, and MIMO-DFE, allowing tradeoffs between performance and complexity. We also derive the equalizer coefficients, discuss timing recovery, and consider channel estimation. Simulation results demonstrate the performance of the proposed LSTE structures and the tradeoffs between performance and complexity of the multistage structure and the single-stage version. We also demonstrate the impact of imperfect channel estimation, imperfect interference cancellation, the number of receive antennas, filter length, and oversampling on performance.

Equalization concepts for EDGE

An equalization concept for the novel radio access scheme Enhanced Data rates for GSM Evolution (EDGE) is proposed by which high performance can be obtained at moderate computational complexity. Because high-level modulation is employed in EDGE, optimum equalization as usually performed in Global System for Mobile Communications (GSM) receivers is too complex and suboptimum schemes have to be considered. It is shown that delayed decision-feedback sequence estimation (DDFSE) and reduced-state sequence estimation (RSSE) are promising candidates. For various channel profiles, approximations for the bit error rate of these suboptimum equalization techniques are given and compared with simulation results for DDFSE. It turns out that a discrete-time prefilter creating a minimum-phase overall impulse response is indispensable for a favorable tradeoff between performance and complexity. Additionally, the influence of channel estimation and of the receiver input filter is investigated and the reasons for performance degradation compared to the additive white Gaussian noise channel are indicated. Finally, the overall system performance attainable with the proposed equalization concept is determined for transmission with channel coding.

Improved spatial-temporal equalization for EDGE: a fast selective-direction MMSE timing recovery algorithm and two-stage soft-output equalizer

The radio interface EDGE (Enhanced Data rates for Global Evolution) is currently being standardized as an evolutionary path from GSM and TDMA-IS136 for third-generation high-speed data wireless systems. For the EDGE system with multiple antennas, spatial-temporal equalization (STE) can reduce intersymbol interference and co-channel interference, thereby increasing the capacity and range. In this paper, we propose two new techniques to improve the performance of a previously proposed STE: a fast timing recovery algorithm for a selective time-reversal equalizer and a two-stage soft-output equalizer. The new timing recovery algorithm determines the estimated burst timing and processing direction by computing the minimum mean-square error (MMSE) for decision feedback equalizers in both the forward and reverse time directions. The two-stage soft-output equalizer is the cascade of a delayed decision-feedback sequence estimator (DDFSE) and maximum a posteriori probability (MAP) estimator. The DDFSE provides better noise variance estimation and channel truncation for the following MAP. The performance of the new STE is evaluated for the EDGE. At 10% block error rate, the two-branch receiver requires a 3-7-dB lower signal-to-interference ratio (SIR) than the previous approach. Compared with the one-branch receiver, the two-branch receiver requires a 4-dB lower SNR with noise only, and a 10-27-dB lower SIR with a single interferer.

Practical reduced-complexity equaliser for EDGE

EDGE uses the 8PSK modulation scheme. It is difficult to employ optimum equalisation due to the high computation complexity required. Consequently, a sub-optimum technique offering an approximation to maximum likelihood sequence estimation performance is required. A simple, reduced complexity fractional spaced equaliser structure involving the delayed decision feedback sequence estimation is therefore presented. The fractional spaced approach does not require additional pre-processing creating a minimum phase overall channel impulse. It is suitable for unknown channels and has near optimum performance.

Joint equalization and interference suppression for high data rate wireless systems

Enhanced Data Rates for Global Evolution (EDGE) is currently being standardized as an evolution of GSM in Europe and of IS-136 in the United States as an air interface for high speed data services for third generation mobile systems. In this paper, we study space-time processing for EDGE to provide interference suppression. We consider the use of two receive antennas and propose a joint equalization and diversity receiver. This receiver uses feedforward filters on each diversity branch to perform minimum mean-square error cochannel interference suppression, while leaving the intersymbol interference to be mitigated by the subsequent equalizer. The equalizer is a delayed decision feedback sequence estimator, consisting of a reduced-state Viterbi processor and a feedback filter. The equalizer provides soft output to the channel decoder after deinterleaving. We describe a novel weight generation algorithm and present simulation results on the link performance of EDGE with interference suppression. These results show a significant improvement in the signal-to-interference ratio (SIR) performance due to both diversity (against fading) and interference suppression. At a 10% block error rate, the proposed receiver provides a 20 dB improvement in SIR for both the typical urban and hilly terrain profiles.

A receiver with iterative soft decision and forward error correction

As a receiving method suitable for channels with intersymbol interference such as frequency selective fading, a method is proposed in which the soft decision using the hard decision sequence generated from the data after forward error correction (FEC) is alternated with FEC. In the conventional receiver with a serial configuration of the equalizer, deinverter, and decoder, the performance of the equalizer significantly affects the overall receiver characteristics. Hence, the performance of the equalizer is emphasized. On the other hand, in the proposed method the error generated in the initial equalization can be corrected by repetitive soft decision and FEC. Therefore, the method does not strongly depend on the performance of the equalizer used in the initial equalization. Hence, even if an equalizer requiring a smaller amount of process with a somewhat degraded performance is used in the initial equalization in place of a high-performance equalizer, the amount of process for the entire receiver can be decreased without degrading the characteristics of the proposed method very much. In this paper, the characteristics of the proposed method are evaluated by computer simulation. The following are the confirmed conclusions. (1) The proposed method using the maximum-likelihood sequence estimation (MLSE) as the equalization has superior BER characteristics over the conventional method using a maximum a posteriori (MAP) equalizer and MLSE. (2) In a transmission channel with a large delay spread, the degradation of the BER characteristics of the proposed method using the delayed decision-feedback sequence estimation (DDFSE) in place of MLSE is very slight while the amount of processing can be reduced substantially from that in the conventional method using MLSE. © 2000 Scripta Technica, Electron Comm Jpn Pt 1, 83(9): 76–85, 2000

Equalisation for EDGE mobile communications

An equalisation concept for the novel radio access scheme enhanced data rates for GSM evolution (EDGE) is proposed, by which high performance can be obtained at moderate computational complexity. Because high-level modulation is employed in EDGE, optimum equalisation as performed in most GSM receivers is too complex, and suboptimum schemes have to be considered. It is shown that delayed decision-feedback sequence estimation (DDFSE) is a promising candidate. For various channel profiles, approximations for the bit error rate of DDFSE are given and compared with simulation results. It turns out that a discrete-time allpass prefilter creating a minimum-phase overall impulse response is indispensable for obtaining a favourable tradeoff between performance and complexity. Additionally, the influence of the receiver input filter is investigated.

Improving differential detection of MDPSK by nonlinear noise prediction and sequence estimation

A new technique is proposed to improve the performance of differential detection (DD) of M-ary differential phase-shift keying (MDPSK) significantly, applying sequence estimation. In order to obtain an appropriate representation of the received signal, a nonlinear time-variant finite impulse response or infinite impulse response prediction-error filter is used. For both filter structures the optimum coefficients are derived, assuming transmission over an additive white Gaussian noise (AWGN) channel. Delayed decision-feedback sequence estimation (DDFSE) is employed to estimate the transmitted symbol sequence. It is shown by simulations that even for decision-feedback equalization, which is a simple special case of DDFSE, a significant performance improvement of conventional DD under AWGN conditions results. In contrast to other noncoherent low-complexity receivers proposed in literature, this receiver is very robust under flat fading (Rayleigh and Ricean) conditions.

Reduced-complexity equalization techniques for broadband wireless channels

This paper presents reduced-complexity equalization techniques for broadband wireless communications, both outdoors (fixed or mobile wireless asynchronous transfer mode (ATM) networks) and indoors [high-speed local-area networks (LANs)]. The two basic equalization techniques investigated are decision-feedback equalization (FE) and delayed decision-feedback sequence estimation (DDFSE). We consider the use of these techniques in highly dispersive channels, where the impulse response can last up to 100 symbol periods. The challenge is in minimizing the complexity as well as providing fast equalizer start-up for transmissions of short packets. We propose two techniques which, taken together, provide an answer to this challenge. One is an open-loop timing recovery approach (for both DFE and DDFSE) which can be executed prior to equalization; the other is a modified DFE structure for precanceling postcursors without requiring training of the feedback filter. Simulation results are presented to demonstrate the feasibility of the proposed techniques for both indoor and outdoor multipath channel models. The proposed open-loop timing recovery technique plays a crucial role in maximizing the performance of DFE and DDFSE with short feedforward spans (the feedforward section of DDFSE is a Viterbi sequence estimator). A feedforward span of only five is quite sufficient for channels with symbol rate-delay spread products approaching 100. The modified DFE structure speeds up the training process for these channels by 10-20 times, compared to the conventional structure without postcursor precancellation. The proposed techniques offer the possibility of practical equalization for broadband wireless systems.

A reduced-complexity online state sequence and parameter estimator for superimposed convolutional coded signals

This paper develops a reduced-complexity online state sequence and parameter estimator for superimposed convolutional coded signals. Joint state sequence and parameter estimation is achieved by iteratively estimating the state sequence via a variable reduced-complexity Viterbi algorithm (VRCVA) and the model parameters via a recursive expectation maximization (EM) approach. The VRCVA is developed from a fixed reduced-complexity Viterbi algorithm (FRCVA). The FRCVA is a special case of the delayed decision-feedback sequence estimation (DDFSE) algorithm. The performance of online versions of the FRCVA, VRCVA, and the standard Viterbi algorithm (VA) are compared when they are used to estimate the state sequence as part of the reduced-complexity online state sequence and parameter estimator.

Combined partial response decision feedback equalization and coding for the Lorentzian channel

This paper considers a combined coding and equalization technique applied to the magnetic recording channel. The optimization of the coded-equalization scheme for this channel requires balancing the coding rate with its associated distance gain and equalization loss. A major concern in a coded-equalization scheme using decision feedback equalization (DFE) is error propagation. A new indicator to measure the effect of error propagation when using soft decision decoding is proposed. For medium to high normalized recording densities (S>1.5), it is shown that error propagation significantly degrades the theoretical performance of a detection scheme which combines a partial response DFE and simple trellis codes. In fact, this combined system performs worse than the uncoded one. The addition of delayed decision feedback sequence estimation (DDFSE) to limit the effect of error propagation is then investigated. With a moderate decoding complexity, positive coding gains have been achieved at such densities when choosing the partial response polynomial according to the derived indicator.