The paper presents MIMO field performance results observed using a ultra mobile broadband (UMB) testbed network. We evaluate metrics such as antenna correlations and channel condition number to characterize the MIMO channel. Results show that low condition numbers, which are beneficial to MIMO, are prevalent for a majority of the coverage area in our network. We demonstrate that the use of MIMO provides gains of the order of 20-40% over SIMO transmissions. These gains are made possible by the use of cross-polarized transmit antennas and advanced UMB features that allow dynamic MIMO vs. SIMO transmission selection based on channel conditions. These results are obtained in a truly mobile, wireless wide-area deployment, which makes them unique. Our results point to the viability and value of MIMO in future mobile wireless networks.

The problem of transmission of information over arbitrarily permuted parallel channels is studied here. The transmitter does not know over which channel a certain code-sequence will actually be transmitted, however the receiver knows how the sequences are permuted. The permutation is arbitrary but constant during the (simultaneous) transmission of the code-sequences via the parallel channels. It is shown first that the sum of the capacities of each channel is achievable for such a communication system in the special case where the capacity achieving input distributions of all channels are identical. More important is that this sum-capacity can also be achieved using a single channel code for all channels combined with a sequential decoding method. The construction of a rate-matching code based on maximum distance separable (MDS) codes turns out to be crucial. Finally, the case where the parallel channels have different capacity-achieving input distributions is investigated. Also for this case the capacity is determined. Again, this capacity is achievable with a sequential decoding procedure.

In mobile reception, the reliability of orthogonal frequency division multiplexing (OFDM) is limited because of the time-varying nature of the channel. This causes intercarrier interference (ICI) and increases inaccuracies in channel tracking. We model the ICI using derivatives of the channel amplitude. This allows us to design a relatively simple receiver scheme that iteratively cancels the ICI. The design of the canceler aims at maximizing the signal-to-noise-plus-ICI ratio at the detector input. We also propose a new channel estimator, and we show that it achieves reliable mobile reception in practical situations that are relevant to terrestrial Digital Video Broadcasting (DVB-T). Extensive simulations for a receiver with one or two antennas show that a small number of iterations between ICI cancellation and channel estimation allow a reliable reception at vehicle speeds above 100 km/h.

The paper discusses antenna subset selection in MIMO wireless systems. The subsets of transmit and receive antennas are selected so as to maximise the channel capacity. First of all, we establish the relationship between the multiplexing gain and the diversity gain achievable with adaptive antenna subset selection. Second, we indicate a selection rule that allows a full diversity advantage to be achieved with a reduced computational effort, by decoupling the combined transmit/receive selection into the separate selection of transmit/receive subsets. Finally, we study the performance of practical systems with antenna selection in the context of high throughput MIMO-OFDM WLAN.

This paper discusses the packet error rate (PER) performance of multiple-input multiple-output (MIMO) wireless systems. We focus our discussion on communication systems based on the IEEE 802.11a/g standard. In particular, we study the performance of spatial multiplexing systems with joint encoding at the transmitter and linear detection at the receiver. We show that spatial multiplexing systems based on minimum mean square error (MMSE) or zero forcing (ZF) demultiplexing benefit greatly from antenna subset selection. These results agree with recent analytical results showing the equivalence in diversity order between a full system (all receive antennas) and a system with antenna selection.

In orthogonal frequency-division multiplexing, time variations of a multipath channel lead to a loss of orthogonality between the subcarriers, and thereby limit the achievable throughput. This paper proposes a general framework for a controlled removal of intercarrier interference (ICI) and channel acquisition. The core idea behind our method is to use a finite power series expansion for the time-varying frequency response, along with the known statistical properties of mobile radio channels. Channel acquisition and ICI removal are accomplished in the frequency domain and allow for any desired tradeoff between the residual ICI level, the required training for channel acquisition, and processing complexity. The proposed approach enables a high spectral efficiency (64-quadrature amplitude modulation mode) of digital video broadcasting-terrestrial in highly mobile environments.

This paper discusses antenna sub-set selection in multiple-input multiple-output wireless systems. The antennas are selected so as to maximize the channel capacity. We present a near-optimal selection algorithm that enables tractable statistical analysis of the selection gain. We show that with antenna selection the capacity is statistically lower bounded by the capacity of the set of parallel independent SIMO channels each with selection diversity. We leverage this result to prove the equivalence in diversity order between the full system (all receive antennas) and the system with antenna selection. Simulations validating analysis and illustrating algorithm performance are also presented.

The paper describes a novel encoding strategy for the existing layered space-time architectures used in multiple-element antenna (MEA) transmission systems. Layered architectures like V-BLAST regard a MEA channel as a set of independently encoded mutually interfering scalar channels (layers) which are decoded subsequently, so that a hefty fraction of the total MEA capacity may be achieved with a moderate complexity proportional to the number of transmit antennas. Practically achievable rates are, however, limited by the minimum of the outage throughputs taken over all layers. Usually, this limitation is overcome due to a joint encoding/decoding of different layers, resulting in a substantial increase in decoding complexity. The encoding strategy described allows leveraging this limitation in some cases of practical interest, without altering the existing space-time processing schemes. Numerical analysis confirms the high potential of this approach in the case of small outage rates.

This correspondence discusses the problem of the receive antenna subset selection in multiple-element antenna (MEA) transmission systems. The antennas are selected so as to maximize the channel capacity. A set of near-optimal selection algorithms is presented. The first algorithm in particular allows statistical analysis of selection gains. We present tight analytic lower bounds on the outage capacity achievable through antenna selection. Extensive simulations validating analysis and illustrating performance of the selection algorithms are also presented.

The authors recently introduced a novel approach for detection-estimation of more uncorrelated Gaussian sources than sensors in sparse linear antenna arrays that is critically dependent on accurate likelihood-ratio (LR) maximisation over the set of candidate model covariance matrices. Here we introduce a non-asymptotic lower bound for the maximised LR. Any solution could now be tested against this bound with a high probability of correctly identifying a non-optimal solution. We demonstrate that existing techniques, based not on the exact LR criteria, but on related criteria, fail to give solutions that approach this lower bound for typical sparse array applications. A novel LR optimisation technique is introduced; for such sensitive sparse array problems, this technique is shown to generate a class of solutions that statistically exceed the lower bound. Some properties of solutions belonging to this class are discussed.

The use of multiple element antennas (MEA) at the transmitter and receiver offers a possibility to substantially increase the throughput of a fading channel. Recent publications show that for a moderate number of transmit/receive subchannels, the outage throughput of MEA transceivers improves substantially when a big set of transmit or/and receive antennas is available to select a suitable subset of antennas which is subject to channel realisation. Optimal subset selection is, however, numerically intensive and hardly feasible for a big number of antennas and a moderate number of subchannels. In this contribution, we present a suboptimal selection algorithm which yields an important complexity reduction and may enable a real-time selection. Numerical results suggest that such a sub-optimal selection yields a negligible loss compared to the optimal strategy in the case of frequency flat channel. A possible extension for frequency selective channels is also indicated.

In orthogonal frequency division multiplexing (OFDM), time variations of a fading multipath environment lead to a loss of orthogonality between the subcarriers and thereby limit the achievable throughput. This paper suggests a general framework for a controlled removal of intercarrier interference (ICI) and channel acquisition. The core idea behind our method is to use a finite Taylor expansion for the time-varying frequency response along with the known statistical properties of mobile radio channels. Channel acquisition and ICI removal are accomplished in the frequency domain and allow for any desired tradeoff between the residual ICI level, the required training for channel acquisition and processing complexity.

The reception of the DVB-T signal on a mobile equipment is strongly affected by Doppler spread. Most of the current proposals require the use of two or more antennas at the receiver and a computationally expensive signal processing. In this paper for a DVB-T receiver with a single antenna we propose new schemes based on the interference cancellation principle. With respect to previously studied techniques, our solution is able to deliver DVB-T services at higher speeds, while a more flexible architecture allows a wider range of trade-off between complexity and performance. In order to improve the performance in terms of maximum achievable speed, an iterative scheme is proposed which performs both the interference cancellation and the channel estimation. A decoding and re-encoding of the inner convolutional code is inserted into the iteration process in order to boost the performance. Moreover, by applying the interference cancellation principle also for the estimation of the channel parameters, we obtain a scheme with a reasonable complexity for the DVB system. Simulation results for various DVB-T modes, constellations and code rates show that the proposed schemes allow the correct reception of DVB-T for moderate and high speeds.

The capacity of a wireless link with multiple transmit antennas is compared to the average per user capacity of a multiuser wireless network with space division multiple access (SDMA). Under the assumption of a rich scattering environment and a narrow-band link, the propagation medium is modeled as a Rayleigh flat fading with a good receive diversity. By contrast to the previous works, the assumption of full decorrelation of the transmit antennas is relaxed. This enables a study of the transmit diversity scenarios where the transmit antennas occupy a limited space. The new analytic results give rise to a comparison of the capacity achievable with multiple transmit antennas within a limited transmit volume on one hand, and the capacity of an SDMA network with spatially spread users, on the other hand. The capacity simulations corresponding to an indoor wireless LAN are presented along with the theoretical results.

The paper proposes and analyses a new receiver structure to mitigate the effect of Doppler on the reception of OFDM signals. A discrete-frequency channel representation is developed for the link between the input of the transmit I-FFT and the output of the receive FFT. It is based on a Taylor expansion of the time variations of the received subcarrier amplitudes. The model realistically addresses the correlation of fading at neighboring subcarriers. We study a new type of receiver which estimates not only amplitudes but also derivatives of subcarriers amplitudes. An adaptive MMSE filter is proposed to cancel the intercarrier interference (ICI) resulting from Doppler. This results in a substantial improvement of the link performance.

This paper considers the direction-of-arrival (DOA) estimation identifiability problem for uncorrelated Gaussian sources and nonuniform antenna arrays. It is now known that sparse arrays always suffer from manifold ambiguity, which arises due to linear dependence amongst the columns of the array manifold matrix (the "steering vectors"). While the standard subspace DOA estimation algorithms such as MUSIC fail to provide proper unambiguous estimates under these conditions, we demonstrate that in most cases involving uncorrelated Gaussian sources, manifold ambiguity does not necessarily imply nonidentifiability. An effective manifold ambiguity resolution algorithm is introduced. A superior number of uncorrelated Gaussian sources (more than sensors) may also be unambiguously localised by sparse arrays under specified identifiability conditions. While manifold ambiguity does not apply to superior scenarios, a similar "co-array manifold ambiguity" phenomenon may compromise DOA estimation. The proposed algorithm can also resolve such ambiguity in all identifiable cases.

Maximum-likelihood sequence estimation is often used to recover digital signals transmitted over finite memory convolutive channels when an estimate of the channel is available. We study the impact of channel estimation errors on the quality of sequence detection. The general case of single input multiple output (SIMO) channels is considered. An asymptotic upper bound for the symbol error rate is presented which allows to treat channel estimation errors as equivalent losses in signal-to-noise ratio (SNR). This relationship is studied and numerically validated for the standard least squares channel estimate and for the semi-blind estimator which makes use of the empirical subspace of the observed data.

Second-order blind deconvolution of single input multiple output (SIMO) finite impulse response (FIR) channels is considered. A major drawback of several blind diversity techniques using antenna arrays/temporal-oversampling is high sensitivity to the choice of the model order. In this contribution, a robust method using only the second-order statistics is described. It provides high and robust estimation accuracy even for a limited sample size and an unknown model order. In contrast to other suggested approaches, that often exploit properties valid only in the large sample case, the proposed method is applicable both in the large sample and high signal-to-noise ratio (SNR) scenario.

Maximum-likelihood sequence estimation is often used to recover digital signals transmitted over finite memory convolutive channels when an estimate of the channel is available. In this letter, we study the impact of channel estimation errors on the quality of sequence detection. The general case of single input multiple output (SIMO) channels is considered. An asymptotic upper bound for the symbol error rate is presented, which allows us to treat channel estimation errors as equivalent to losses in signal-to-noise ratio (SNR). This relationship is studied and numerically validated for the standard least squares channel estimate.

We investigate DOA (direction-of-arrival) estimation for arbitrary linear arrays, where the antenna positions may be non-integer values in half-wavelength units. We introduce an approach based on arbitrary virtual linear arrays to resolve manifold ambiguity and estimate DOA's in the superior case. These virtual arrays adopt the set of covariance lags specified by the original array and so themselves have an incomplete set of covariance lags. A maximum entropy completion algorithm for the partially-specified Hermitian covariance matrix is proposed. This is followed by an algorithm which searches for a fixed number of plane wavefronts ("generalised Pisarenko completion"). The variety of possible virtual array geometries also permits a "randomised" approach, whereby the DOA estimates are determined as the stable point of partial solutions calculated over the set of particular virtual geometries. Numerical simulations demonstrate the high efficiency of manifold ambiguity resolution, and a remarkable proximity to the Cramer-Rao bound for DOA estimation.

International audience

This paper considers the problem of DOA (direction-of-arrival) estimation for multiple uncorrelated plane waves incident upon so-called "fully-augmentable" sparse linear arrays. When the number of waves, m, is less than the number of antenna sensors, M, a new approach is presented which constructs a full-sized positive definite (p.d.) Toeplitz matrix from the M/spl times/M direct data covariance matrix R/spl circ/. Alternating projections are then applied to find a p.d. Toeplitz matrix with truncated signal subspace. When m/spl ges/M, direct augmentation is used to form an initial Toeplitz estimate which is then treated as in the previous case. Results on simulated data demonstrate that this new approach has superior DOA-estimation performance compared with existing methods.

Multipath propagation appears to be a typical limitation in mobile digital communications where it leads to severe intersymbol interference. The classical techniques to overcome this problem use either periodically sent training sequences or blind approaches. This paper addresses the blind identification of FIR multiple input multiple output (MIMO) transfer functions in the case where the number of inputs is strictly less than the number of outputs. We consider a second order identification providing signals extraction up to a regular instantaneous mixture matrix. A novel method of second order channel estimation is presented. Basically exploiting the simultaneous MA and AR nature of the observations, it displays however a considerable improvement as compared to the original linear prediction approach. Performance study and comparison with the existing approach is provided by computer simulations.

Blind identification of single-input multiple-output (SIMO) FIR systems based on second order statistics has attracted a lot of research efforts. We focus on subspace estimation procedures, which exploit the structure of the range space of certain matrix-valued statistics constructed by arranging in a prescribed order the covariance coefficients of the observations. General subspace identifiability results are obtained, based on the properties of minimal polynomial bases of rational subspaces. Several subspace estimation procedures are then derived. These estimators are all based on the (possibly weighted) least-square solution of an overdetermined system of linear equations and are thus well-suited for practical implementations.

A new class of nonstationary adaptive techniques is proposed to put up with clutter distortions originated from the current pattern readjustment, caused by the temporal perturbations of the jammer spatial covariance matrix. To introduce this approach we describe, by using the scalar-type vector-valued autoregressive model the inter-repetition sequence of the clutter-only antenna input snapshots vectors. Examples of joint jammer rejection and decontamination of F2 layer-propagated signals strongly perturbed by phase-path variations prove the efficiency of the proposed approach. The new adaptive processing algorithms yield an adequate approach for frequency management, oriented towards operational adaptive jammer rejection facilities.>