Search

Block turbo codes (BTCs) are constructed by serially concatenating linear block codes and iteratively decoded by letting each component code be decoded in two stages. The Chase algorithm is employed in the first stage to make a list of candidate codewords by generating a fixed number of test sequences (TSs) and algebraically decoding them, regardless of the signal-to-noise ratio or the iteration number. In the second stage, the extrinsic information is generated for iterative decoding. In this letter, we propose a low-complexity decoding algorithm for BTCs. The proposed algorithm first checks whether an algebraic hard-decision decoder outputs a codeword for a given decoder input vector, and then adaptively applies one of the two estimation rules. Based on these two rules, the number of TSs in the proposed algorithm can be made monotonically decreasing with iterations. Numerical results demonstrate that the proposed algorithm has much lower computational complexity with a negligible performance loss, compared with the conventional decoding scheme based on the Chase algorithm.

In this study, the coverage of downlink cellular orthogonal frequency-division multiple-access networks transmitting one quadrature-amplitude modulation symbol per hop is analysed. The coverage probability is first defined by using an upper bound on the bit error rate (BER) of a bit-interleaved coded modulation system. Moreover then, the moment generating functions of the legacy Gaussian and Laplacian metrics (LMs) for soft-decision decoding of an employed error-correcting code are derived for computing the upper bound on the BER and the corresponding coverage probability of the networks. Numerical results demonstrate that due to the impulsive nature of the inter-cell interference for moderate-to-small cell loads, the networks using the LM for soft-decision decoding significantly outperform those employing the legacy Gaussian metric in terms of the achievable cell radius and the amount of the average transmit power reduction.

Multitone-frequency and quadrature-amplitude modulation (MFQAM) is a combination of multitone frequency-shift keying and quadrature-amplitude modulation (QAM). In this study, the authors consider MFQAM with active subcarriers grouped in balance, where the active subcarriers are divided into a specified number of groups and the same QAM symbol is conveyed by every subcarrier in each group. This grouping may make MFQAM have frequency diversity in itself. They first derive the joint probability density function of the correlator outputs of an MFQAM receiver. They then compute the channel capacity of a coded modulation (CM) system employing MFQAM over Rayleigh fading channels. As a result, they demonstrate that for a moderate code rate between 1/3 and 2/3, its normalised throughput is close to the channel capacity limit in the low signal-to-noise ratio region, contrary to a CM system employing the conventional modulation. They also evaluate the frame error rate (FER) performance of a CM system employing MFQAM when a non-binary error-correcting code is employed. Numerical results demonstrate that the normalised throughput of the CM system computed via the FER shows a similar trend to that obtained via the channel capacity.

A systematic, granulized piecewise linear approximation to the Jacobian logarithm is developed. This approximation reverts to the Jacobian approximation and the exact Jacobian logarithm at the far extremes. The utility of the derived approximation is demonstrated by applying it to the decoding of Turbo coded, MIMO/BICM-ID systems.

In this paper, union upper bound on convolutional coded bit error rates (BER) is derived for downlink orthogonal frequency division multiple access (OFDMA) networks. According to the numerical results, for the small network loads, the BER performance with Laplacian decoding metric outperforms the BER performance with Gaussian decoding metric under downlink OFDMA networks with Viterbi decoder.

The performance of uplink cellular OFDMA networks transmitting multiple QAM symbols per hop is evaluated under log-normal shadowing and Rayleigh fading without resorting to a Gaussian approximation of the inter-cell interference. The derived interference statistic is observed to deviate significantly from the Gaussian distribution which may be exploited to dramatically enhance the network performance. It is also observed that the network performance is highly dependent on the number and the geometry of the QAM symbols transmitted within a hop, especially for large code frame lengths.

Using a Gaussian approximation to the distribution of the message vectors under density evolution, we design irregular weighted nonbinary repeat-accumulate codes over GF(q) with q-ary orthogonal modulation. The resulting codes achieve a frame error rate of 10-1 within 0.56 to 0.91 dB of channel capacity under the AWGN channel.

An approximate upper bound is derived to the maximum likelihood decoding threshold of weighted nonbinary repeat-accumulate (WNRA) codes over GF(q) with q-ary orthogonal modulation and coherent detection over the additive white Gaussian noise channel. As with binary RA codes, WNRA codes show the potential for achieving the channel capacity with decreasing code rate for large enough q.

A very simple design rule for parallel concatenated Turbo codes over GF(q) with q-ary orthogonal modulation is proposed. Unlike previously proposed exhaustive search methods, the proposed design rule allows straightforward and systematic code designs for any value of q ≥ 2 and any constraint length greater than 2.

An iterative algorithm employing decision feedback provided by either an equalizer or a channel decoder is proposed in order to compensate for the phase noise resulting from imperfect oscillators in orthogonal frequency-division multiplexing (OFDM) systems. In the proposed algorithm, the received OFDM symbol is partitioned into subblocks in the time domain and the estimate of the time-average of the phase noise at each subblock is used to compensate for the phase noise. A formula for the signal-to-interference-plus-noise ratio (SINR) after phase noise compensation is derived and its values are evaluated under a variety of conditions in order to demonstrate the efficiency of the proposed algorithm. Numerical results show that the proposed algorithm achieves performance close to that of an OFDM system without phase noise over a wide range of conditions, while requiring only one-half the complexity of the partial intercarrier interference reduction algorithm.

Algorithms based on the QR decomposition of the equivalent space-time channel matrix have been proved useful in the detection of V-BLAST systems. Especially, the parallel detection (PD) algorithm offers ML approaching performance up to 4 transmit antennas with reasonable complexity. We show that when directly applied to STBCs, the PD algorithm may suffer a rather significant SNR degradation over ML detection, especially at high SNRs. However, simply extending the PD algorithm to allow p ≥2 candidate layers, i.e. p -PD, regains almost all the loss but only at a significant increase in complexity. Here, we propose a simplification to the p-PD algorithm specific to STBCs without a corresponding sacrifice in performance. The proposed algorithm results in significant complexity reductions for moderate to high order modulations.

In this paper, we propose a novel class of unitary space-time constellations based on quaternary quasi-orthogonal sequences (QOSs), minimizing the maximum correlation between the signal matrices. The proposed design, owing to the underlying algebraic structure of the QOSs, allows for an efficient decoder structure while offering performance virtually identical to the unitary space-time constellation design by Hochwald et al.

In this paper, a distribution matching technique is proposed to process the extrinsic information produced by practical multiple-antenna detectors with iteratively demodulated, bit-interleaved coded modulation (BICM-ID). It is shown that significant gains may be achieved by appropriately processing the extrinsic information generated by practical multiple-antenna detectors before feeding it to the channel decoder. The validity of the proposed matching technique is shown via simulation results applied to Turbo-coded BICM-ID systems using (QRD)-M and list sphere detectors.

The probability density function (PDF) of the intercell interference for downlink cellular orthogonal frequency division multiple access networks under log-normal shadowing and multipath Rayleigh fading is derived. The derived PDF is observed to highly deviate from the Gaussian distribution, especially at moderate to small network loads. Based on the derived inter-cell interference statistics, the decoder performance can be significantly improved compared to the case when the traditional decoder optimized for the AWGN channel is employed. A simple suboptimum decoding algorithm that closely approximates the derived optimum decoder performance is also proposed.

In this correspondence, we propose a simplified QR-decomposition-M algorithm for coded multiple-input multiple-output systems based on the idea of selective branch extension. The proposed algorithm results in significant complexity reductions without observable performance losses, especially for large modulation orders where complexity reduction is most important.

The need for real-time computation of the Euclidean norm of a vector arises frequently in many signal processing applications such as vector median filtering, vector quantization and multiple-input multiple-output wireless communication systems. In this correspondence, we examine the properties of a linear combination of the 1-norm and the infinity norm as an approximation to the Euclidean norm of real-valued vectors. The approximation requires only two multiplications regardless of the vector length and does not require sorting of the absolute values of the vector entries. Numerical results show that the considered approximation incurs negligible performance degradations in typical applications.

The authors optimize system parameters including modulation order M, diversity order L, and traffic density lambda to maximize the throughput of frequency-hopping-multiple-access networks with multilevel frequency-shift keying. It is shown that optimal lambda is close to ln 2 with the exception that optimal lambda=1 for the erasures coding channel with M=2. For the erasures coding channel, optimum L is also derived, and for the hard-decision coding channel, optimum L is numerically searched. For the region of M 64, the hard-decision coding channel achieves slightly higher throughput than the erasures coding channel. While the maximum throughput for the hard-decision coding channel monotonously increases as M increases, the maximum throughput for the erasures coding channel shows a local increase at M=2, and it is significantly higher than that for the hard-decision coding channel with M=2. Tradeoffs between system parameters to achieve the maximum throughput under a fixed bandwidth constraint are also presented. It is shown that the optimal set for (M,L,N(=number of available frequencies)) is approximately (G,log2G,G) for both coding channels, where G is the total system bandwidth expansion normalized by bit rate

Optimum values for the modulation order M, code rate r, and the number of frequency-hop slots q maximizing the network throughput are obtained based on simulations for frequency-hopped spread-spectrum multiple-access networks, where L Q-ary Reed-Solomon (RS) code symbols are transmitted per hop, and each Q-ary RS code symbol is transmitted using log/sub M/Q M-ary frequency-shift-keying-modulated signals. Network throughput is evaluated under additive white Gaussian noise and Rayleigh fading channels. For the case when the received RS symbol is not interfered by multiple-access interference (MAI), a closed-form expression for the symbol-error probability is derived, and for the case when the symbol is interfered by MAI, simulated symbol-error probabilities are used. It is shown that the optimum M is four or eight, irrespective of the channel environment and the number of users. The optimum code rate is determined primarily based on the channel environment and does not show much dependence on M or Q. It is also shown that for the case of synchronous hopping under Rayleigh fading at high signal-to-noise ratios, the difference in instantaneous power among the interfering users significantly improves the performance, compared with the case when there is no fading. We also consider the case when the receiver erases the symbols that are interfered and compare the performance with the case of the hard decisions receiver.

A novel frequency-hop spread-spectrum multiple-access network employing M-ary orthogonal Walsh sequence keying with noncoherent demodulation is proposed. The transmitted Walsh sequence is overlaid by a user-specific pseudonoise sequence to reduce the effect of multiple-access hits. Two Gaussian approximations for the multiple-access interference from both the dehopped slot and its neighboring slots are developed and are used to analyze the performance of the proposed network for synchronous and asynchronous hopping under nonfading and Rayleigh fading channels. The effect of imperfect hop timing synchronization at the receiver is also analyzed. It is shown that the proposed network offers significantly improved network throughput compared to networks based on traditional M-ary frequency-shift keying modulation.

An adaptive serial search pseudonoise (PN) code acquisition scheme is proposed, in which the detection threshold is scaled by the instantaneous received power measured prior to PN code correlation. We observe that the proposed scheme achieves significantly improved mean acquisition times compared to the conventional nonadaptive schemes under Rayleigh fading and pulsed Gaussian noise jamming. Furthermore, the proposed scheme is shown to be optimum under pulsed Gaussian noise jamming in the sense that it forces the worst case jamming fraction to unity.

Unlike traditional trained channel equalizers, not much work has been done to theoretically characterize the convergence properties of blind channel equalizers due to their inherent nonlinearity. It is only recently that convergence properties of some well-known algorithms such as the generalised Sato algorithm (GSA) and the constant modulus algorithm (CMA) have been analytically derived. In this paper, the convergence properties of the stop-and-go algorithm proposed by Picchi and Prati (1987) are analyzed. The derived mean squared error and the coefficient trajectories are compared with simulation results to verify the validity of the analytical results.

Recently, Liu et al., developed the /spl Sigma//sub 0/-rank criteria for space-time codes. It provides a sufficient condition on codeword and generator matrices defined over finite rings /spl Zopf//sub 2k/(j) to ensure full spatial diversity with 2/sup 2k/-QAM modulation. Here, we generalize the /spl Sigma//sub 0/-rank criteria and derive a sufficient condition on the generator matrices defined over finite rings /spl Zopf//sub 2l/(j) to ensure full spatial diversity with 2/sup 2k/-QAM modulation for any positive integer l/spl les/k. We also show that generator matrices defined over GF(2) satisfying the BPSK stacking construction constraint of Mammons and El Gamal achieve full spatial diversity when used with 2/sup 2k/-QAM modulation.

In this paper, we derive error probability expressions for binary phase-shift keying (BPSK) and quaternary phase-shift keying (QPSK) spread direct-sequence spread-spectrum (DSSS) systems employing random spreading sequences with RAKE receivers. The derived expressions accurately take into account the effect of interpath interference which usually has been neglected in previous analyses. Selection, equal gain, and maximal ratio techniques are considered for diversity combining. Two possible finger assignment strategies, one based on the instantaneous amplitudes and another based on the average powers of the multipath components, are considered for the assignment of multipath components to the available demodulating fingers in the RAKE receiver. Also, various simple, and in many cases, closed-form approximations for the error probabilities are derived and their accuracies are assessed.

Space-time and space-frequency coded orthogonal frequency division multiplexing systems employing the Alamouti scheme were proposed by Lee and Williams. Straightforward implementation of these systems requires separate inverse discrete Fourier transform (IDFT) processing blocks for each of the two transmit antennas. In this letter, DFT/IDFT symmetry properties are utilized to approximately halve the computational load of these systems at the transmitter.

New full-rate space-time block codes achieving full diversity for QAM modulation using an even number of transmit antennas over quasistatic Rayleigh fading channels are proposed. The proposed codes are constructed by serially concatenating unitary rotating precoders in X. Giraud et al. (1997) and Y. Xin et al. (2001) with the Alamouti code of S. M. Alamouti (1998). The coding advantage (V. Tarokh et al., 1998) of the proposed code for a codeword pair corresponding to any distinct input pair is shown to be greater than or equal to that of the ST-CR code (Y. Xin et al., 2001).

A novel modulation scheme suitable for noncoherent demodulation based on quaternary quasi-orthogonal sequences is proposed. Compared to orthogonal modulation, the controlled quasi-orthogonality between the sequences allows significantly increased bandwidth efficiency with little or no degradation in power efficiency. A hardware efficient demodulator structure using fast Walsh transforms is also presented.

A maximum-likelihood (ML) coarse symbol synchronization algorithm for OFDM systems in multipath channels is derived. Several simplifying assumptions reduce the rather impractical ML algorithm to more practical forms of minimum mean-squared-error and maximum correlation algorithms. All of the derived algorithms exploit the cyclic prefix and thus do not require a known training sequence. Estimating the position of the first arriving path, the algorithms allow maximum utilization of the guard interval thus minimizing intersymbol interference.

The performance of robust metrics (metrics that can be computed from the outputs of the matched filters only) with convolutional coding and diversity under worst-case partial-band noise jamming is analyzed. Both binary and dual-k convolutional codes employing these metrics with diversity are compared via Union-Chernoff bounds. The performances of metrics considered in the literature that assume perfect side-information are given for comparison purposes. It is found that there exist very good robust metrics that provide performance comparable to metrics using perfect side-information. Among the robust metrics considered, the self-normalized metric offers the best performance and achieves performance practically identical to that of the square-law-combining metric with perfect side-information for M=8. >

Expressions are developed for the probability of error for asynchronous frequency-hop spread-spectrum multiple-access networks using Markov hopping patterns and binary frequency shift keying (BFSK) with one symbol transmitted per hop. The expressions are exact when there is one interfering user and orthogonal BFSK is used. They provide excellent approximations when there are more than one interfering user. It is also shown that the error probability when Markov hopping patterns are used is a good approximation to the error probability when memoryless hopping patterns are used. By computing the channel capacity and the associated throughput, a simple hard decision receiver is shown to perform much better than a receiver using perfect side-information to erase the symbols transmitted on hops that were hit when all the users have the same power and one binary symbol is transmitted per hop. >

An efficient synchronization scheme based on a repetitive preamble structure for orthogonal frequency division multiplexing systems was proposed by Schmidl and Cox. However, standard autocorrelation based estimation algorithms result in a plateau in the timing metric due to the cyclic prefix (CP), resulting in a region of uncertainty. Several modifications, either to the preamble structure itself or the estimation algorithm, have been proposed to remedy this problem. In this paper, we present a maximum-likelihood (ML) symbol timing estimation algorithm taking into account the presence of the CP. The performance of its practical modification is evaluated via computer simulations.

Recently, a variety of full-rate, full-diversity quasi-orthogonal space-time block codes (QO-STBCs) have been designed. Though the quasi-orthogonality does reduce the decoding complexity from order of MQN to MQN/2 for Q-ary QAM modulation with N transmit and M receive antennas, there is still need for improvement. Several simple suboptimum decoding algorithms were proposed but suffer quite severe performance losses over maximum likelihood (ML) decoding. Here, we propose a class of decoding algorithms for QO-STBCs based on interference cancellation with complexity on the order of MQ N/2 -1 and performance virtually identical to that of ML decoding. A modification to the base algorithm is also developed with complexity approximately 1/radic(Q) times that of the base algorithm.

An efficient synchronization scheme based on a repetitive preamble structure for orthogonal frequency division multiplexing systems was proposed by Schmidl and Cox. However, standard autocorrelation based estimation algorithms result in a plateau in the timing metric due to the cyclic prefix (CP), resulting in a region of uncertainty. Several modifications, either to the preamble structure itself or the estimation algorithm, have been proposed to remedy this problem. In this paper, we present a maximum-likelihood (ML) symbol timing estimation algorithm taking into account the presence of the CP. The performance of its practical modification is evaluated via computer simulations.

A novel frequency-hop spread-spectrum multiple-access (FHSS-MA) network employing M-ary orthogonal Walsh sequence keying with noncoherent demodulation is proposed. The Walsh sequence is overlaid by a user specific PN sequence to reduce the effects of multiple-access hits. Two Gaussian approximations for the multiple-access interference from both the dehopped slot and its neighboring slots are developed. These are used to analyze the performance of the proposed network for synchronous and asynchronous hopping under nonfading and Rayleigh fading channels. It is shown that the proposed network significantly improves the maximum achievable normalized throughput compared to networks employing traditional MFSK.