Review of Single-User Space-Time Matrix Modulation. First, we briefly review space-time matrix mo... more Review of Single-User Space-Time Matrix Modulation. First, we briefly review space-time matrix modulation for the single-user case. The single-user transmission system is shown in Figure 1.1(b). We consider N d < MULTIUSER MIMO CHANNEL EQUALIZATION 3 User k Receiver d k,1 [n] b k,1 [n] r1[n]d k,1 [n] d k,N d,k [n] d k,N d,k [n] b k,N T [n] rN R [n]
Conference Record of the Thirty-Sixth Asilomar Conference on Signals, Systems and Computers, 2002.
We propose an efficient iterative algorithm for demodulating/decoding linear dispersion (LD) code... more We propose an efficient iterative algorithm for demodulating/decoding linear dispersion (LD) codes in the case of an unknown MIMO channel. The algorithm is based on the POCS (projections onto convex sets) technique and consists of a~teniating pmjections of an estimate of the transmit matri.r onto hvo linear subspaces. In the noise-free case, the iteration is giraranteed to converge to the transmitted data whenever the LD code theoretically allows perfect data reconstruction. In the presence of noise, simulation results demonstrate the algorithm's good performance. We also propose a DFT-matrix based code design that yields a further substantial reduction of complexiry, and a modified algorithm that exploits the presence of pilot symbols.
Most transmission schemes for MIMO channels assume (i) a block fading channel without delay sprea... more Most transmission schemes for MIMO channels assume (i) a block fading channel without delay spread and (ii) availability of channel state information at the receiver. Here, we extend the space-time motrir modulation (STMMj scheme and the iterative demodulation algorithm that we introduced previously to unknown, doubly selective MlMO channels, i.e., delayDoppler-spread MlMO channels that are unknown to the receiver. We show that the structure inherent in STMM allows perfect reconstruction of the data when transmitting over an unknown doubly selective channel (apparently, this is not currently possible with other transmission schemes). Numerical simulations demonstrate significant performance advantages of STMM over Cayley differential unitary space-time modulation.
Efficient POCS algorithms for deterministic blind equalization of time-varying channels
Globecom '00 - IEEE. Global Telecommunications Conference. Conference Record (Cat. No.00CH37137), 2000
We present two POCS (projections onto convex sets) algorithms for deterministic blind equalizatio... more We present two POCS (projections onto convex sets) algorithms for deterministic blind equalization of linear time-varying (LTV) channels. Our approach is based on a multichannel LTI representation of LTV channels. We prove a theorem on unique reconstruction/equalization, formulate a POCS algorithm for the resolution of a matrix ambiguity, and show how to cope with unequal subchannel lengths. We also present
IEEE International Conference on Communications, 2003. ICC '03., 2000
We present space-time methods for channel estimation in the downlink of a UMTS/TDD system. The ch... more We present space-time methods for channel estimation in the downlink of a UMTS/TDD system. The channels associated to all base stations near the mobile receiver are estimated by a multiuser, multi-antenna technique. We develop a minimum mean-square error (MMSE) channel estimator that incorporates midamble detection/estimation and least-squares channel estimation as preparatory stages. Combination of this MMSE channel estimator with a successive interference cancellation scheme allows to cope with strong co-channel interference. Simulation results indicate the good performance of our space-time channel estimator for various realistic propagation scenarios.
GLOBECOM '03. IEEE Global Telecommunications Conference (IEEE Cat. No.03CH37489), 2000
We propose a space-time detection algorithm for a UMTS/TDD downlink scenario with strong interfer... more We propose a space-time detection algorithm for a UMTS/TDD downlink scenario with strong interference. Conventional space-time detection schemes based on MMSE equalization require accurate estimation of the correlation matrix of the received signal vector, which is difficult because of the time variation of the mobile radio channel. We thus develop a decision feedback detection scheme in which the correlation matrix is "detected" rather than estimated. For detection of the active data channels, a maximal invariant detection statistic is presented. Simulation results demonstrate that the proposed space-time detection algorithm features good performance for various realistic propagation scenarios even at low SINR.
Global Telecommunications Conference, 2002. GLOBECOM '02. IEEE, 2000
Thb paper prcsroii mlbods for synchronkailon ta the downlink of D U M T M D D ryrtpm. Spacetime p... more Thb paper prcsroii mlbods for synchronkailon ta the downlink of D U M T M D D ryrtpm. Spacetime proc-tag based on mullipk rerdvr antennas b ired lo soppren poteniidly slrong coshpnnel blerfmnce c~v s e d by other base nadons. We develop a dererdon-based synchronlzstlon pmcednre exploiting the IIIDCi m r e of UMTMDD. and we present three sltemsilve dewtion 4gorilhm. Sholation msdll Othp v~r i o~l prOpsgall0. t c p n d 0 5 tadkale ihe good performance of our iynchmnizatlon melhodr In Ibe presence o f strong rc-chnnmrl taterference.
2004 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2004
The conventional nulling-and-cancelling (NC) detection scheme for MIMO systems uses the layerwise... more The conventional nulling-and-cancelling (NC) detection scheme for MIMO systems uses the layerwise post-detection mean-square errors (MSEs) as reliability measures for layer sorting. These MSEs are average measures that do not depend on the received vector. In this paper, we propose the novel dynamic nulling-and-cancelling (DNC) detector that performs "dynamic" layer sorting based on the current received vector. Approximate a-posteriori probabilities (constructed by means of a Gaussian approximation for the postdetection interference) are used as measures of layer reliability. This results in an MMSE nulling technique that uses a simple layersorting rule with significantly improved performance. Our simulation results show that the DNC scheme can yield near-ML performance for a wide range of system sizes and signal-to-noise ratios.
GLOBECOM '03. IEEE Global Telecommunications Conference (IEEE Cat. No.03CH37489), 2003
It is well known that suboptimal detection schemes for MIMO spatial multiplexing systems (equaliz... more It is well known that suboptimal detection schemes for MIMO spatial multiplexing systems (equalization-based as well as nulling-and-cancelling detectors) cannot exploit all of the available diversity. In this paper, we show that this inferior performance is primarily caused by poorly conditioned channel realizations. We then present the novel sphere-projection algorithm (SPA) that is robust to poorly conditioned channels. The SPA is a computationally efficient add-on to standard suboptimal detectors. Simulation results show that the SPA is able to achieve near-ML performance and significantly increased diversity gains. The SPA's computational complexity is comparable to that of nulling-and-cancelling detectors and only a fraction of that of the Fincke-Phost sphere-decoding algorithm for ML detection.
Channel dependent tree pruning for the sphere decoder
IEEE International Symposium on Information Theory, 2004
We reduce the complexity of sphere decoding for space-time transmission by "guessing" t... more We reduce the complexity of sphere decoding for space-time transmission by "guessing" the first few levels of the search tree a priori. These guesses are based on a simplified channel model. The resulting complexity is only a fraction of the original complexity while performance remains close to optimal.
Recent work on wireless beamforming has focused on multiuser diversity effects, where terminals i... more Recent work on wireless beamforming has focused on multiuser diversity effects, where terminals in a cellular system are chosen opportunistically from a pool of terminals based on their responses to test beams. Orthogonality between the test beams is typically chosen to aid subsequent interference-free transmission to the corresponding terminals. However, much of this work assumes that i) the pool of terminals on a given time-frequency resource is large enough that a subset of terminals can be found whose spatial signatures match the orthogonal test beams; ii) the responses to the test beams of the entire pool are known to the basestation; iii) beamforming considerations can drive traffic scheduling and resource allocation. These conditions are not always met. Rather, we examine orthogonalizing a given set of data-carrying beams for terminals that are scheduled on distinct time-frequency resources. The scheduled terminals are not chosen for their spatial signatures, but rather for their need to receive data. Our orthogonalization is chosen to maximize the beamforming gain to the scheduled terminals, whose channels are generally not spatially orthogonal. Unscheduled terminals may then "eavesdrop" on pilots embedded in the orthogonal beams and thereby learn their own channel state information. This process eliminates the need for separate dedicated orthogonal training signals. We show how this scheme may be deployed in frequency-division duplex and time-division duplex systems. We quantify the price in beamforming gain paid for orthogonalizing a set of beams. One result analytically shows that the average price is at most 1.4 dB in SNR.
It is known that conventional nulling-and-canceling (NC) detection for multiple-input/multiple-ou... more It is known that conventional nulling-and-canceling (NC) detection for multiple-input/multiple-output (MIMO) systems cannot exploit all of the available diversity, and, thus, its performance is significantly inferior to that of maximum likelihood (ML) detection. Conventional NC employs the layerwise postequalization signal-to-noise ratios (SNRs) as reliability measures for layer sorting. These SNRs are average quantities that do not depend on the received vector. In this paper, we propose the novel dynamic nulling-and-canceling (DNC) technique that uses approximate a posteriori probabilities as measures of layer reliability. The DNC technique is a minimum mean-square error (MMSE) nulling scheme combined with an improved "dynamic" layer sorting rule that exploits the information contained in the current received vector. We calculate the error probability of DNC for a simple special case and show that it is upper bounded by the error probability of conventional NC. Simulation results are presented for spatial multiplexing systems and for systems using linear dispersion codes. It is demonstrated that the DNC technique can yield near-ML performance for a wide range of system sizes and channel SNRs at a fraction of the computational complexity of the sphere-decoding algorithm for ML detection.
It is well known that suboptimal detection schemes for multiple-input multiple-output (MIMO) spat... more It is well known that suboptimal detection schemes for multiple-input multiple-output (MIMO) spatial multiplexing systems (equalization-based schemes as well as nulling-and-cancelling schemes) are unable to exploit all of the available diversity, and thus, their performance is inferior to ML detection. Motivated by experimental evidence that this inferior performance is primarily caused by the inability of suboptimal schemes to deal with "bad" (i.e., poorly conditioned) channel realizations, we study the decision regions of suboptimal schemes for bad channels. Based on a simplified model for bad channels, we then develop two computationally efficient detection algorithms that are robust to bad channels. In particular, the novel sphere-projection algorithm (SPA) is a simple add-on to standard suboptimal detectors that is able to achieve near-ML performance and significantly increased diversity gains. The SPA's computational complexity is comparable with that of nulling-and-cancelling detectors and only a fraction of that of the Fincke-Phost sphere-decoding algorithm for ML detection.
We propose two methods for the estimation of scattering functions of random time-varying channels... more We propose two methods for the estimation of scattering functions of random time-varying channels. In contrast to existing methods, our methods exploit the underspread property of these channels to achieve good estimation performance and low computational complexity. The first method uses a dedicated sounding to measure the channel. The second method uses the data signal of an ongoing data transmission as sounding signal and thus allows estimation without dedicated sounding. Both methods are effectively unbiased and can be implemented efficiently using the Zak transform. The performance of our scattering function estimators is studied both analytically by means of variance bounds and experimentally through numerical simulation, and their superiority over existing methods is demonstrated.
Review of Single-User Space-Time Matrix Modulation. First, we briefly review space-time matrix mo... more Review of Single-User Space-Time Matrix Modulation. First, we briefly review space-time matrix modulation for the single-user case. The single-user transmission system is shown in Figure 1.1(b). We consider N d < MULTIUSER MIMO CHANNEL EQUALIZATION 3 User k Receiver d k,1 [n] b k,1 [n] r1[n]d k,1 [n] d k,N d,k [n] d k,N d,k [n] b k,N T [n] rN R [n]
Conference Record of the Thirty-Sixth Asilomar Conference on Signals, Systems and Computers, 2002.
We propose an efficient iterative algorithm for demodulating/decoding linear dispersion (LD) code... more We propose an efficient iterative algorithm for demodulating/decoding linear dispersion (LD) codes in the case of an unknown MIMO channel. The algorithm is based on the POCS (projections onto convex sets) technique and consists of a~teniating pmjections of an estimate of the transmit matri.r onto hvo linear subspaces. In the noise-free case, the iteration is giraranteed to converge to the transmitted data whenever the LD code theoretically allows perfect data reconstruction. In the presence of noise, simulation results demonstrate the algorithm's good performance. We also propose a DFT-matrix based code design that yields a further substantial reduction of complexiry, and a modified algorithm that exploits the presence of pilot symbols.
Most transmission schemes for MIMO channels assume (i) a block fading channel without delay sprea... more Most transmission schemes for MIMO channels assume (i) a block fading channel without delay spread and (ii) availability of channel state information at the receiver. Here, we extend the space-time motrir modulation (STMMj scheme and the iterative demodulation algorithm that we introduced previously to unknown, doubly selective MlMO channels, i.e., delayDoppler-spread MlMO channels that are unknown to the receiver. We show that the structure inherent in STMM allows perfect reconstruction of the data when transmitting over an unknown doubly selective channel (apparently, this is not currently possible with other transmission schemes). Numerical simulations demonstrate significant performance advantages of STMM over Cayley differential unitary space-time modulation.
Efficient POCS algorithms for deterministic blind equalization of time-varying channels
Globecom '00 - IEEE. Global Telecommunications Conference. Conference Record (Cat. No.00CH37137), 2000
We present two POCS (projections onto convex sets) algorithms for deterministic blind equalizatio... more We present two POCS (projections onto convex sets) algorithms for deterministic blind equalization of linear time-varying (LTV) channels. Our approach is based on a multichannel LTI representation of LTV channels. We prove a theorem on unique reconstruction/equalization, formulate a POCS algorithm for the resolution of a matrix ambiguity, and show how to cope with unequal subchannel lengths. We also present
IEEE International Conference on Communications, 2003. ICC '03., 2000
We present space-time methods for channel estimation in the downlink of a UMTS/TDD system. The ch... more We present space-time methods for channel estimation in the downlink of a UMTS/TDD system. The channels associated to all base stations near the mobile receiver are estimated by a multiuser, multi-antenna technique. We develop a minimum mean-square error (MMSE) channel estimator that incorporates midamble detection/estimation and least-squares channel estimation as preparatory stages. Combination of this MMSE channel estimator with a successive interference cancellation scheme allows to cope with strong co-channel interference. Simulation results indicate the good performance of our space-time channel estimator for various realistic propagation scenarios.
GLOBECOM '03. IEEE Global Telecommunications Conference (IEEE Cat. No.03CH37489), 2000
We propose a space-time detection algorithm for a UMTS/TDD downlink scenario with strong interfer... more We propose a space-time detection algorithm for a UMTS/TDD downlink scenario with strong interference. Conventional space-time detection schemes based on MMSE equalization require accurate estimation of the correlation matrix of the received signal vector, which is difficult because of the time variation of the mobile radio channel. We thus develop a decision feedback detection scheme in which the correlation matrix is "detected" rather than estimated. For detection of the active data channels, a maximal invariant detection statistic is presented. Simulation results demonstrate that the proposed space-time detection algorithm features good performance for various realistic propagation scenarios even at low SINR.
Global Telecommunications Conference, 2002. GLOBECOM '02. IEEE, 2000
Thb paper prcsroii mlbods for synchronkailon ta the downlink of D U M T M D D ryrtpm. Spacetime p... more Thb paper prcsroii mlbods for synchronkailon ta the downlink of D U M T M D D ryrtpm. Spacetime proc-tag based on mullipk rerdvr antennas b ired lo soppren poteniidly slrong coshpnnel blerfmnce c~v s e d by other base nadons. We develop a dererdon-based synchronlzstlon pmcednre exploiting the IIIDCi m r e of UMTMDD. and we present three sltemsilve dewtion 4gorilhm. Sholation msdll Othp v~r i o~l prOpsgall0. t c p n d 0 5 tadkale ihe good performance of our iynchmnizatlon melhodr In Ibe presence o f strong rc-chnnmrl taterference.
2004 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2004
The conventional nulling-and-cancelling (NC) detection scheme for MIMO systems uses the layerwise... more The conventional nulling-and-cancelling (NC) detection scheme for MIMO systems uses the layerwise post-detection mean-square errors (MSEs) as reliability measures for layer sorting. These MSEs are average measures that do not depend on the received vector. In this paper, we propose the novel dynamic nulling-and-cancelling (DNC) detector that performs "dynamic" layer sorting based on the current received vector. Approximate a-posteriori probabilities (constructed by means of a Gaussian approximation for the postdetection interference) are used as measures of layer reliability. This results in an MMSE nulling technique that uses a simple layersorting rule with significantly improved performance. Our simulation results show that the DNC scheme can yield near-ML performance for a wide range of system sizes and signal-to-noise ratios.
GLOBECOM '03. IEEE Global Telecommunications Conference (IEEE Cat. No.03CH37489), 2003
It is well known that suboptimal detection schemes for MIMO spatial multiplexing systems (equaliz... more It is well known that suboptimal detection schemes for MIMO spatial multiplexing systems (equalization-based as well as nulling-and-cancelling detectors) cannot exploit all of the available diversity. In this paper, we show that this inferior performance is primarily caused by poorly conditioned channel realizations. We then present the novel sphere-projection algorithm (SPA) that is robust to poorly conditioned channels. The SPA is a computationally efficient add-on to standard suboptimal detectors. Simulation results show that the SPA is able to achieve near-ML performance and significantly increased diversity gains. The SPA's computational complexity is comparable to that of nulling-and-cancelling detectors and only a fraction of that of the Fincke-Phost sphere-decoding algorithm for ML detection.
Channel dependent tree pruning for the sphere decoder
IEEE International Symposium on Information Theory, 2004
We reduce the complexity of sphere decoding for space-time transmission by "guessing" t... more We reduce the complexity of sphere decoding for space-time transmission by "guessing" the first few levels of the search tree a priori. These guesses are based on a simplified channel model. The resulting complexity is only a fraction of the original complexity while performance remains close to optimal.
Recent work on wireless beamforming has focused on multiuser diversity effects, where terminals i... more Recent work on wireless beamforming has focused on multiuser diversity effects, where terminals in a cellular system are chosen opportunistically from a pool of terminals based on their responses to test beams. Orthogonality between the test beams is typically chosen to aid subsequent interference-free transmission to the corresponding terminals. However, much of this work assumes that i) the pool of terminals on a given time-frequency resource is large enough that a subset of terminals can be found whose spatial signatures match the orthogonal test beams; ii) the responses to the test beams of the entire pool are known to the basestation; iii) beamforming considerations can drive traffic scheduling and resource allocation. These conditions are not always met. Rather, we examine orthogonalizing a given set of data-carrying beams for terminals that are scheduled on distinct time-frequency resources. The scheduled terminals are not chosen for their spatial signatures, but rather for their need to receive data. Our orthogonalization is chosen to maximize the beamforming gain to the scheduled terminals, whose channels are generally not spatially orthogonal. Unscheduled terminals may then "eavesdrop" on pilots embedded in the orthogonal beams and thereby learn their own channel state information. This process eliminates the need for separate dedicated orthogonal training signals. We show how this scheme may be deployed in frequency-division duplex and time-division duplex systems. We quantify the price in beamforming gain paid for orthogonalizing a set of beams. One result analytically shows that the average price is at most 1.4 dB in SNR.
It is known that conventional nulling-and-canceling (NC) detection for multiple-input/multiple-ou... more It is known that conventional nulling-and-canceling (NC) detection for multiple-input/multiple-output (MIMO) systems cannot exploit all of the available diversity, and, thus, its performance is significantly inferior to that of maximum likelihood (ML) detection. Conventional NC employs the layerwise postequalization signal-to-noise ratios (SNRs) as reliability measures for layer sorting. These SNRs are average quantities that do not depend on the received vector. In this paper, we propose the novel dynamic nulling-and-canceling (DNC) technique that uses approximate a posteriori probabilities as measures of layer reliability. The DNC technique is a minimum mean-square error (MMSE) nulling scheme combined with an improved "dynamic" layer sorting rule that exploits the information contained in the current received vector. We calculate the error probability of DNC for a simple special case and show that it is upper bounded by the error probability of conventional NC. Simulation results are presented for spatial multiplexing systems and for systems using linear dispersion codes. It is demonstrated that the DNC technique can yield near-ML performance for a wide range of system sizes and channel SNRs at a fraction of the computational complexity of the sphere-decoding algorithm for ML detection.
It is well known that suboptimal detection schemes for multiple-input multiple-output (MIMO) spat... more It is well known that suboptimal detection schemes for multiple-input multiple-output (MIMO) spatial multiplexing systems (equalization-based schemes as well as nulling-and-cancelling schemes) are unable to exploit all of the available diversity, and thus, their performance is inferior to ML detection. Motivated by experimental evidence that this inferior performance is primarily caused by the inability of suboptimal schemes to deal with "bad" (i.e., poorly conditioned) channel realizations, we study the decision regions of suboptimal schemes for bad channels. Based on a simplified model for bad channels, we then develop two computationally efficient detection algorithms that are robust to bad channels. In particular, the novel sphere-projection algorithm (SPA) is a simple add-on to standard suboptimal detectors that is able to achieve near-ML performance and significantly increased diversity gains. The SPA's computational complexity is comparable with that of nulling-and-cancelling detectors and only a fraction of that of the Fincke-Phost sphere-decoding algorithm for ML detection.
We propose two methods for the estimation of scattering functions of random time-varying channels... more We propose two methods for the estimation of scattering functions of random time-varying channels. In contrast to existing methods, our methods exploit the underspread property of these channels to achieve good estimation performance and low computational complexity. The first method uses a dedicated sounding to measure the channel. The second method uses the data signal of an ongoing data transmission as sounding signal and thus allows estimation without dedicated sounding. Both methods are effectively unbiased and can be implemented efficiently using the Zak transform. The performance of our scattering function estimators is studied both analytically by means of variance bounds and experimentally through numerical simulation, and their superiority over existing methods is demonstrated.
Uploads
Papers by H. Artes