This paper presents a prototype pair of dual-layer electromagnetic bandgap slotted circularly pol... more This paper presents a prototype pair of dual-layer electromagnetic bandgap slotted circularly polarized patch 8x8Tx/8x8Rx orthogonally-polarized antenna arrays designed for in-band full-duplex massive MIMO. The design processes of the antenna array, beamforming feeding network, and beam steering algorithms are discussed. The proposed prototype operates in a bandwidth of 250 MHz (or 7.1% at 3.5 GHz) from 3.35 GHz to 3.6 GHz. Using orthogonal Tx/Rx polarization, an average mutual coupling between any two Tx and Rx antenna elements of-49.2 dB can be achieved. Further improvement in Tx-to-Rx isolation at the beam level can be achieved with the two proposed beamforming schemes. Illustrative results using Tx/Rx 1x4 sub-arrays indicate the formed beams with an average mutual coupling of-65.6 dB between any two Tx and Rx beams (i.e., an improvement of 16.4 dB in Tx/Rx isolation). The sample Tx and Rx farfield beam patterns, radiation null depth and position, and corresponding Tx-to-Rx mutual coupling levels are investigated, which gives insight into the beamforming isolation improvement design. Within a steering range from-50 to 50 degrees, a directivity over 10.7 dB, a 3dB-beamwidth narrower than 33.0 degrees, and a normalized sidelobe level lower than-9.0 dB are achieved. Simulation and measurement results on the mutual coupling, beamforming directivity, beamwidth, and radiation patterns in various scenarios are investigated and found to be in good agreement.
The University of Danang - Journal of Science and Technology, Aug 31, 2016
A thesis submitted to McGill University in partial fulfillment of the requirements for the degree... more A thesis submitted to McGill University in partial fulfillment of the requirements for the degree of Master of Engineering.
A multistage interference cancettation (MIC) technique with Rake diversity (MIC-Rnke) for QPSK as... more A multistage interference cancettation (MIC) technique with Rake diversity (MIC-Rnke) for QPSK asynchronous dwect sequence code division multiple access (DS/CDMA) system over frequency selective multipath Rayleigh fading channels is presented. Unlike the conventiouatMIC, which tries to subtract the lump sum of the multipleaccessinterference (MAI) and the seW-interference(S1),the MIC-Rake attempts to cancel the MAI and the partial S1, and to treat the residual S1 as useful siguat for symbol decision. The Rake combining is employed to collect signat replicas over multiple fading paths, The upper and lower bounds on the bit error probability are derived by using a Gaussian approximation. Furthermore, the effect of the chaunel estimation error is studied, Analysis and simulation show that the MIC-Rake can provide a performance close to the ideal performance of single-user system, and outperforms the conventional MIC even in the presence of channel estimation error.
This paper considers the down-link transmission of an OFDMA-based multi-cell virtualized wireless... more This paper considers the down-link transmission of an OFDMA-based multi-cell virtualized wireless network (VWN) to serve users belonging to different service providers (slices) with the quality-of-service (QoS) requirements in terms of each slice's minimum reserved rate. In order to improve energy efficiency, we formulate a joint base station (BS) assignment, sub-carrier and power allocation problem to minimize the total transmit power of all the BSs subject to the QoS constraints. This problem is inherently a non-convex optimization problem. To tackle its computational complexity, we apply successive convex approximation (SCA) and complementary geometric programming (CGP) to convert the problem into a computationally efficient formulation and propose an iterative algorithm for solving the problem. We introduce a variable called user association factor (UAF) for jointly assigning users to both BSs and sub-carriers. Simulation results illustrate the performance enhancement of the VWN achieved through our formulation for different network scenarios.
A thesis submitted to McGill University in partial fulfillment of the requirements of the degree ... more A thesis submitted to McGill University in partial fulfillment of the requirements of the degree of Master of Engineering.
In this paper, we present a joint channel estimation and synchronization technique for burst-mode... more In this paper, we present a joint channel estimation and synchronization technique for burst-mode OFDM systems. The technique uses the non-linear recursive least squares algorithm (NL-RLS) to jointly estimate the channel impulse response (CIR), the carrier frequency offset (CFO), the sampling clock offset (SCO), and the timing offset, and to correct for each of these in the digital domain and without the use of a delay-locked loop (DLL). The need for pilot symbols in the payload data is eliminated by a decision-directed operation mode. The time-domain NL-RLS algorithm needs a small number of parameters to be estimated and can suppress the effect of decision feedback errors. Simulation results confirm its improved performance and relatively low variance.
An adaptive multistage parallel interference cancellation technique based on a partial interferen... more An adaptive multistage parallel interference cancellation technique based on a partial interference cancellation (IC) approach [1] was proposed in [2] for multipath fading channels. In this paper, the proposed technique is applied to develop a receiver structure in an AWGN environment. Unlike the scheme in [1], the weighting factors in this proposed scheme are derived by minimizing the meansquare error between the received signal and its estimate through an LMS algorithm. Neither training sequence nor pilot signal is needed. The complexity of the proposed adaptive multistage PIC structure is much lower than that of linear multiuser detectors. Simulation results show superior performance of the proposed receiver structure over an AWGN channel and in various conditions.
EURASIP Journal on Advances in Signal Processing, Jun 6, 2014
This paper is concerned with linear precoding designs for multiuser downlink transmissions. We co... more This paper is concerned with linear precoding designs for multiuser downlink transmissions. We consider a multiple-input single-output (MISO) system with multiple single-antenna user equipment (UE) experiencing nonhomogeneous average signal-to-noise ratio (SNR) conditions. The first part of this work examines different precoding schemes with perfect channel state information (CSI) and average SNR at the base-station (eNB). We then propose a weighted minimum mean squared error (WMMSE) precoder, which takes advantage of the non-homogeneous SNR conditions. Given in a closed-form solution, the proposed WMMSE precoder outperforms other well-known linear precoders, such as zero-forcing (ZF), regularized ZF (RZF), while achieving a close performance to the locally optimal iterative WMMSE (IWMMSE) precoder, in terms of the achievable network sum-rate. In the second part of this work, we consider the non-homogeneous multiuser system with limited and quantized channel quality indicator (CQI) and channel direction indicator (CDI) feedbacks. Based on the CQI and CDI feedback models proposed for the Long-Term Evolution Advanced standard, we then propose a robust WMMSE precoder in a closed-form solution which takes into account the quantization errors. Simulation shows a significant improvement in the achievable network sum-rate by the proposed robust WMMSE precoder, compared to non-robust linear precoder designs.
In this paper, we consider a down-link transmission of multicell virtualized wireless networks (V... more In this paper, we consider a down-link transmission of multicell virtualized wireless networks (VWNs) where users of different service providers (slices) within a specific region are served by a set of base stations (BSs) through orthogonal frequency division multiple access (OFDMA). In particular, we develop a joint BS assignment, sub-carrier and power allocation algorithm to maximize the network throughput, while satisfying the minimum required rate of each slice. Under the assumption that each user at each transmission instance can connect to no more than one BS, we introduce the user-association factor (UAF) to represent the joint sub-carrier and BS assignment as the optimization variable vector in the mathematical problem formulation. Sub-carrier reuse is allowed in different cells, but not within one cell. As the proposed optimization problem is inherently non-convex and NP-hard, by applying the successive convex approximation (SCA) and complementary geometric programming (CGP), we develop an efficient two-step iterative approach with low computational complexity to solve the proposed problem. For a given power-allocation, Step 1 derives the optimum userassociation and subsequently, for an obtained user-association, Step 2 find the optimum power-allocation. Simulation results demonstrate that the proposed iterative algorithm outperforms the traditional approach in which each user is assigned to the BS with the largest average value of signal strength, and then, joint sub-carrier and power allocation is obtained for the assigned users of each cell. Especially, for the cell-edge users, simulation results reveal a coverage improvement up to 57% and 71% for uniform and non-uniform users distribution, respectively leading to more reliable transmission and higher spectrum efficiency for VWN.
This paper investigates the benefits of incorporating underlaid full-duplex (FD) device-todevice ... more This paper investigates the benefits of incorporating underlaid full-duplex (FD) device-todevice (D2D) communications into cellular networks. Toward this end, we provide an analytical performance characterization of underlaid D2D cellular networks where D2D users operate in FD mode under the presence of residual self-interference. In considered networks, the base-stations (BSs) are distributed according to a hexagonal grid, while the locations of cellular and D2D users follow Poisson point processes (PPPs). Based on the stochastic-geometry approach, we develop the approximations of key performance metrics including coverage probabilities and achievable sum-rates of both cellular and D2D links, and such approximations involve quickly commutable integrals. Under a special case in which the number of D2D links is sufficiently large, the obtained approximations can be simplified to closed-form expressions, allowing characterize the sum-rate behaviors under the effects of various system parameters. We show that underlaid D2D communications in cellular network can offer a significant spectral efficiency gain as compared to pure cellular transmission. With a sufficiently low self-interference cancellation level, FD D2D can offer substantial spectral efficiency improvement over the half-duplex (HD) counterpart. Finally, the resulting performance metrics are compared with multi-cell networks operating in standard and fractional frequency reuse modes, and observe that frequency reuse provides improved coverage probabilities of both cellular and D2D links, but substantially reduces the D2D sum-rate performance. INDEX TERMS Device-to-device communications, cellular networks, full-duplex, stochastic geometry. III. COVERAGE AND SUM-RATE ANALYSIS OF CELLULAR LINK
In this paper, a multilevel-quantized soft-limiting (SL-MQ) detector for frequency hopping spread... more In this paper, a multilevel-quantized soft-limiting (SL-MQ) detector for frequency hopping spread spectrum multiple access (FH-SSMA) system is proposed and analyzed. Numerical and simulation results in frequency selective Rayleigh fading channels show that as compared to the hard-limiting (HL) detector, the new SL-MQ with Å can improve the system capacity by almost ½¼± at the bit error rate level of ½¼ ¿. Furthermore, the performance of the SL-MQ has low sensitivity to the optimum value of the amplitude threshold so that it can tolerate inaccurate estimate of its optimum in practice.
This paper presents a novel channel estimation technique for the multiuser massive multiple-input... more This paper presents a novel channel estimation technique for the multiuser massive multiple-input multiple-output (MU-mMIMO) systems using angular-based hybrid precoding (AB-HP). The proposed channel estimation technique generates group-wise channel state information (CSI) of user terminal (UT) zones in the service area by deep neural networks (DNN) and fuzzy c-Means (FCM) clustering. The slow time-varying CSI between the base station (BS) and feasible UT locations in the service area is calculated from the geospatial data by offline ray tracing and a DNN-based path estimation model associated with the 1-dimensional convolutional neural network (1D-CNN) and regression tree ensembles. Then, the UT-level CSI of all feasible locations is grouped into clusters by a proposed FCM clustering. Finally, the service area is divided into a number of non-overlapping UT zones. Each UT zone is characterized by a corresponding set of clusters named as UT-group CSI, which is utilized in the analog RF beamformer design of AB-HP to reduce the required large online CSI overhead in the MU-mMIMO systems. Then, the reduced-size online CSI is employed in the baseband (BB) precoder of AB-HP. Simulations are conducted in the indoor scenario at 28 GHz and tested in an AB-HP MU-mMIMO system with a uniform rectangular array (URA) having 16 × 16 = 256 antennas and 22 RF chains. Illustrative results indicate that 91.4% online CSI can be reduced by using the proposed offline channel estimation technique as compared to the conventional online channel sounding. The proposed DNN-based path estimation technique produces same amount of UT-level CSI with runtime reduced by 65.8% as compared to the computationally expensive ray tracing. The imperfection of UT-level CSI introduced by the DNN-based path estimation technique is mitigated by the FCM clustering technique, where the AB-HP using offline UT-group CSI generated by the DNN-based channel estimation model and ray tracing based model for the RF beamformer achieves, respectively, 98.7% and 99.1% sum-rate performance of the fully digital precoding (FDP) technique using full-size online CSI.
EURASIP Journal on Advances in Signal Processing, May 7, 2007
This paper presents a high-performance, low-complexity, quasi-distributed dynamic spectrum manage... more This paper presents a high-performance, low-complexity, quasi-distributed dynamic spectrum management (DSM) algorithm suitable for DSL systems. We analytically demonstrate that the rate degradation of the distributed iterative waterfilling (IW) algorithm in near-far scenarios is caused by the insufficient utilization of all available frequency and power resources due to its nature of noncooperative game theoretic formulation. Inspired by this observation, we propose the selective IW (SIW) algorithm that can considerably alleviate the performance degradation of IW by applying IW selectively to different groups of users over different frequency bands so that all the available resources can be fully utilized. For N users, the proposed SIW algorithm needs at most N times the complexity of the IW algorithm, and is much simpler than the centralized optimal spectrum balancing (OSB), while it can offer a rate performance much better than that of the IW and close to the maximum possible rate region computed by the OSB in realistic near-far DSL scenarios. Furthermore, its predominantly distributed structure makes it suitable for DSL implementation.
This paper focuses on the design of vector perturbation (VP) precoding for coordinated multi-poin... more This paper focuses on the design of vector perturbation (VP) precoding for coordinated multi-point (CoMP) multiuser downlink transmission. Precoding is performed by individual base stations (BSs) in a distributed manner using only the downlink channel coefficients and user data local to a BS. A cascade precoder structure with an outer precoder managing the inter-cell interference (ICI) and an inner precoder performing mean-squared-error (MSE) minimization-based VP to mitigate the intra-cell interference is proposed. Three different outer precoding techniques are considered. In the first technique, the outer precoder is designed to fully eliminate the ICI by trading off the degrees of freedom (DoFs) available through multiple antennas. While the proposed technique outperforms existing conventional-VP based designs, a large portion of DoF is consumed by the ICI elimination. To overcome this issue, in the second technique, interference alignment-based outer precoding that minimizes the total leakage interference is proposed. To further improve the system performance, in the third approach, precoding by joint minimization of total leakage interference plus MSE is performed. Numerical results show that the proposed cascade precoding structure is an efficient way to use the DoF of CoMP multiuser downlink transmission. INDEX TERMS MIMO broadcast channel, vector perturbation, non-linear precoding, multiuser MIMO, coordinated multipoint transmission.
This paper presents a group-orthogonal OFDMA (GO-OFDMA) suitable for broadband access systems in ... more This paper presents a group-orthogonal OFDMA (GO-OFDMA) suitable for broadband access systems in a fast time-varying frequency-selective fading environment when channel knowledge is not available at the transmitter. Subcarrier grouping to achieve the diversity gain of a time-domain Rake receiver is discussed and the closed-form expression of its bit-error-rate (BER) performance over a frequencyselective Rayleigh fading channel is derived. The proposed GO-OFDMA scheme uses a split-and-group structure and a maximum-likelihood (ML) multiuser detection (MUD) to increase the number of supportable active users and to reduce the peak-to-average ratio (PAR). Analytical and simulation results are in an excellent agreement. Performance evaluation indicates that the proposed GO-OFDMA provides a lower PAR and similar BER as compared with the group-orthogonal multi-carrier CDMA (GO-MC-CDMA), and outperforms the random-hopping (RH)-OFDMA and matched-filter based MC-CDMA.
This paper proposes a capacity-approaching, yet simple scheme for multi-input multiple-output (MI... more This paper proposes a capacity-approaching, yet simple scheme for multi-input multiple-output (MIMO) channels. The proposed scheme is based on a concatenation of a mixture of short memory-length convolutional codes or repetition codes and a short, and simple rate-1 linear block code, followed by either 1-dimensional (1-D) anti-Gray or Gray mapping of quadrature phase-shift keying (QPSK) modulation. By interpreting the rate-1 code and the 1-D mapping as a multi-D mapping performed over multiple transmit antennas, the error performance is analyzed in two regions. In the error-floor region, a tight union bound and the corresponding design criterion on the asymptotic performance are derived. The bound provides a useful tool to predict the error performance at relatively low bit error rate (BER) values. Based on the obtained design criterion, an optimal rate-1 code for each 1-D mapping is then constructed to achieve the best asymptotic performance. In the turbo pinch-off region, by using extrinsic information transfer (EXIT) charts, the most suitable mixed codes are selected for both symmetric and asymmetric antenna configurations. It is demonstrated that the simple concatenation scheme can achieve a near-capacity performance over the MIMO channels. Furthermore, its error performance is shown to be comparable to that obtained by using well-designed irregular LDPC and RA codes, and therefore, the proposed scheme significantly outperforms a scheme employing a parallel concatenated turbo code. Simulation results in various cases are provided to verify the analysis.
This paper is concerned with linear precoding designs for multiuser downlink transmissions. We co... more This paper is concerned with linear precoding designs for multiuser downlink transmissions. We consider a multiple-input single-output (MISO) system with multiple single-antenna user equipment (UE) experiencing nonhomogeneous average signal-to-noise ratio (SNR) conditions. The first part of this work examines different precoding schemes with perfect channel state information (CSI) and average SNR at the base-station (eNB). We then propose a weighted minimum mean squared error (WMMSE) precoder, which takes advantage of the non-homogeneous SNR conditions. Given in a closed-form solution, the proposed WMMSE precoder outperforms other well-known linear precoders, such as zero-forcing (ZF), regularized ZF (RZF), while achieving a close performance to the locally optimal iterative WMMSE (IWMMSE) precoder, in terms of the achievable network sum-rate. In the second part of this work, we consider the non-homogeneous multiuser system with limited and quantized channel quality indicator (CQI) and channel direction indicator (CDI) feedbacks. Based on the CQI and CDI feedback models proposed for the Long-Term Evolution Advanced standard, we then propose a robust WMMSE precoder in a closed-form solution which takes into account the quantization errors. Simulation shows a significant improvement in the achievable network sum-rate by the proposed robust WMMSE precoder, compared to non-robust linear precoder designs.
This paper proposes a delay-aware resource provisioning policy for virtualized wireless networks ... more This paper proposes a delay-aware resource provisioning policy for virtualized wireless networks (VWNs) to minimize the total average transmit power while holding the minimum required average rate of each slice and maximum average packet transmission delay for each user. The proposed cross-layer optimization problem is inherently non-convex and has high computational complexity. To develop an efficient solution, we first transform cross-layer dependent constraints into physical layer dependent ones. Afterwards, we apply different convexification techniques based on variable transformations and relaxations, and propose an iterative algorithm to reach the optimal solution. Simulation results illustrate the effects of the required average packet transmission delay and minimum average slice rate on the total transmission power in VWN. Index Terms-Delay-aware resource provisioning, virtualized wireless networks.
Eurasip Journal on Wireless Communications and Networking, 2006
This paper presents a detailed performance analysis of adaptive downlink resource allocation base... more This paper presents a detailed performance analysis of adaptive downlink resource allocation based on users' instantaneous channel responses using power minimization (PM) and bandwidth constrained power minimization (BCPM) strategies. This study shows that, in cellular systems, where interference is a dominant factor, the link outage performance of a resource allocation strategy varies significantly depending on the user channel parameters. In particular, both analytical and simulation results indicate that the PM strategy outperforms BCPM in a mild shadowing environment. However, in severe shadowing conditions, this trend is reversed. This assessment holds true for both flat and frequency-selective fading.
This paper presents a prototype pair of dual-layer electromagnetic bandgap slotted circularly pol... more This paper presents a prototype pair of dual-layer electromagnetic bandgap slotted circularly polarized patch 8x8Tx/8x8Rx orthogonally-polarized antenna arrays designed for in-band full-duplex massive MIMO. The design processes of the antenna array, beamforming feeding network, and beam steering algorithms are discussed. The proposed prototype operates in a bandwidth of 250 MHz (or 7.1% at 3.5 GHz) from 3.35 GHz to 3.6 GHz. Using orthogonal Tx/Rx polarization, an average mutual coupling between any two Tx and Rx antenna elements of-49.2 dB can be achieved. Further improvement in Tx-to-Rx isolation at the beam level can be achieved with the two proposed beamforming schemes. Illustrative results using Tx/Rx 1x4 sub-arrays indicate the formed beams with an average mutual coupling of-65.6 dB between any two Tx and Rx beams (i.e., an improvement of 16.4 dB in Tx/Rx isolation). The sample Tx and Rx farfield beam patterns, radiation null depth and position, and corresponding Tx-to-Rx mutual coupling levels are investigated, which gives insight into the beamforming isolation improvement design. Within a steering range from-50 to 50 degrees, a directivity over 10.7 dB, a 3dB-beamwidth narrower than 33.0 degrees, and a normalized sidelobe level lower than-9.0 dB are achieved. Simulation and measurement results on the mutual coupling, beamforming directivity, beamwidth, and radiation patterns in various scenarios are investigated and found to be in good agreement.
The University of Danang - Journal of Science and Technology, Aug 31, 2016
A thesis submitted to McGill University in partial fulfillment of the requirements for the degree... more A thesis submitted to McGill University in partial fulfillment of the requirements for the degree of Master of Engineering.
A multistage interference cancettation (MIC) technique with Rake diversity (MIC-Rnke) for QPSK as... more A multistage interference cancettation (MIC) technique with Rake diversity (MIC-Rnke) for QPSK asynchronous dwect sequence code division multiple access (DS/CDMA) system over frequency selective multipath Rayleigh fading channels is presented. Unlike the conventiouatMIC, which tries to subtract the lump sum of the multipleaccessinterference (MAI) and the seW-interference(S1),the MIC-Rake attempts to cancel the MAI and the partial S1, and to treat the residual S1 as useful siguat for symbol decision. The Rake combining is employed to collect signat replicas over multiple fading paths, The upper and lower bounds on the bit error probability are derived by using a Gaussian approximation. Furthermore, the effect of the chaunel estimation error is studied, Analysis and simulation show that the MIC-Rake can provide a performance close to the ideal performance of single-user system, and outperforms the conventional MIC even in the presence of channel estimation error.
This paper considers the down-link transmission of an OFDMA-based multi-cell virtualized wireless... more This paper considers the down-link transmission of an OFDMA-based multi-cell virtualized wireless network (VWN) to serve users belonging to different service providers (slices) with the quality-of-service (QoS) requirements in terms of each slice's minimum reserved rate. In order to improve energy efficiency, we formulate a joint base station (BS) assignment, sub-carrier and power allocation problem to minimize the total transmit power of all the BSs subject to the QoS constraints. This problem is inherently a non-convex optimization problem. To tackle its computational complexity, we apply successive convex approximation (SCA) and complementary geometric programming (CGP) to convert the problem into a computationally efficient formulation and propose an iterative algorithm for solving the problem. We introduce a variable called user association factor (UAF) for jointly assigning users to both BSs and sub-carriers. Simulation results illustrate the performance enhancement of the VWN achieved through our formulation for different network scenarios.
A thesis submitted to McGill University in partial fulfillment of the requirements of the degree ... more A thesis submitted to McGill University in partial fulfillment of the requirements of the degree of Master of Engineering.
In this paper, we present a joint channel estimation and synchronization technique for burst-mode... more In this paper, we present a joint channel estimation and synchronization technique for burst-mode OFDM systems. The technique uses the non-linear recursive least squares algorithm (NL-RLS) to jointly estimate the channel impulse response (CIR), the carrier frequency offset (CFO), the sampling clock offset (SCO), and the timing offset, and to correct for each of these in the digital domain and without the use of a delay-locked loop (DLL). The need for pilot symbols in the payload data is eliminated by a decision-directed operation mode. The time-domain NL-RLS algorithm needs a small number of parameters to be estimated and can suppress the effect of decision feedback errors. Simulation results confirm its improved performance and relatively low variance.
An adaptive multistage parallel interference cancellation technique based on a partial interferen... more An adaptive multistage parallel interference cancellation technique based on a partial interference cancellation (IC) approach [1] was proposed in [2] for multipath fading channels. In this paper, the proposed technique is applied to develop a receiver structure in an AWGN environment. Unlike the scheme in [1], the weighting factors in this proposed scheme are derived by minimizing the meansquare error between the received signal and its estimate through an LMS algorithm. Neither training sequence nor pilot signal is needed. The complexity of the proposed adaptive multistage PIC structure is much lower than that of linear multiuser detectors. Simulation results show superior performance of the proposed receiver structure over an AWGN channel and in various conditions.
EURASIP Journal on Advances in Signal Processing, Jun 6, 2014
This paper is concerned with linear precoding designs for multiuser downlink transmissions. We co... more This paper is concerned with linear precoding designs for multiuser downlink transmissions. We consider a multiple-input single-output (MISO) system with multiple single-antenna user equipment (UE) experiencing nonhomogeneous average signal-to-noise ratio (SNR) conditions. The first part of this work examines different precoding schemes with perfect channel state information (CSI) and average SNR at the base-station (eNB). We then propose a weighted minimum mean squared error (WMMSE) precoder, which takes advantage of the non-homogeneous SNR conditions. Given in a closed-form solution, the proposed WMMSE precoder outperforms other well-known linear precoders, such as zero-forcing (ZF), regularized ZF (RZF), while achieving a close performance to the locally optimal iterative WMMSE (IWMMSE) precoder, in terms of the achievable network sum-rate. In the second part of this work, we consider the non-homogeneous multiuser system with limited and quantized channel quality indicator (CQI) and channel direction indicator (CDI) feedbacks. Based on the CQI and CDI feedback models proposed for the Long-Term Evolution Advanced standard, we then propose a robust WMMSE precoder in a closed-form solution which takes into account the quantization errors. Simulation shows a significant improvement in the achievable network sum-rate by the proposed robust WMMSE precoder, compared to non-robust linear precoder designs.
In this paper, we consider a down-link transmission of multicell virtualized wireless networks (V... more In this paper, we consider a down-link transmission of multicell virtualized wireless networks (VWNs) where users of different service providers (slices) within a specific region are served by a set of base stations (BSs) through orthogonal frequency division multiple access (OFDMA). In particular, we develop a joint BS assignment, sub-carrier and power allocation algorithm to maximize the network throughput, while satisfying the minimum required rate of each slice. Under the assumption that each user at each transmission instance can connect to no more than one BS, we introduce the user-association factor (UAF) to represent the joint sub-carrier and BS assignment as the optimization variable vector in the mathematical problem formulation. Sub-carrier reuse is allowed in different cells, but not within one cell. As the proposed optimization problem is inherently non-convex and NP-hard, by applying the successive convex approximation (SCA) and complementary geometric programming (CGP), we develop an efficient two-step iterative approach with low computational complexity to solve the proposed problem. For a given power-allocation, Step 1 derives the optimum userassociation and subsequently, for an obtained user-association, Step 2 find the optimum power-allocation. Simulation results demonstrate that the proposed iterative algorithm outperforms the traditional approach in which each user is assigned to the BS with the largest average value of signal strength, and then, joint sub-carrier and power allocation is obtained for the assigned users of each cell. Especially, for the cell-edge users, simulation results reveal a coverage improvement up to 57% and 71% for uniform and non-uniform users distribution, respectively leading to more reliable transmission and higher spectrum efficiency for VWN.
This paper investigates the benefits of incorporating underlaid full-duplex (FD) device-todevice ... more This paper investigates the benefits of incorporating underlaid full-duplex (FD) device-todevice (D2D) communications into cellular networks. Toward this end, we provide an analytical performance characterization of underlaid D2D cellular networks where D2D users operate in FD mode under the presence of residual self-interference. In considered networks, the base-stations (BSs) are distributed according to a hexagonal grid, while the locations of cellular and D2D users follow Poisson point processes (PPPs). Based on the stochastic-geometry approach, we develop the approximations of key performance metrics including coverage probabilities and achievable sum-rates of both cellular and D2D links, and such approximations involve quickly commutable integrals. Under a special case in which the number of D2D links is sufficiently large, the obtained approximations can be simplified to closed-form expressions, allowing characterize the sum-rate behaviors under the effects of various system parameters. We show that underlaid D2D communications in cellular network can offer a significant spectral efficiency gain as compared to pure cellular transmission. With a sufficiently low self-interference cancellation level, FD D2D can offer substantial spectral efficiency improvement over the half-duplex (HD) counterpart. Finally, the resulting performance metrics are compared with multi-cell networks operating in standard and fractional frequency reuse modes, and observe that frequency reuse provides improved coverage probabilities of both cellular and D2D links, but substantially reduces the D2D sum-rate performance. INDEX TERMS Device-to-device communications, cellular networks, full-duplex, stochastic geometry. III. COVERAGE AND SUM-RATE ANALYSIS OF CELLULAR LINK
In this paper, a multilevel-quantized soft-limiting (SL-MQ) detector for frequency hopping spread... more In this paper, a multilevel-quantized soft-limiting (SL-MQ) detector for frequency hopping spread spectrum multiple access (FH-SSMA) system is proposed and analyzed. Numerical and simulation results in frequency selective Rayleigh fading channels show that as compared to the hard-limiting (HL) detector, the new SL-MQ with Å can improve the system capacity by almost ½¼± at the bit error rate level of ½¼ ¿. Furthermore, the performance of the SL-MQ has low sensitivity to the optimum value of the amplitude threshold so that it can tolerate inaccurate estimate of its optimum in practice.
This paper presents a novel channel estimation technique for the multiuser massive multiple-input... more This paper presents a novel channel estimation technique for the multiuser massive multiple-input multiple-output (MU-mMIMO) systems using angular-based hybrid precoding (AB-HP). The proposed channel estimation technique generates group-wise channel state information (CSI) of user terminal (UT) zones in the service area by deep neural networks (DNN) and fuzzy c-Means (FCM) clustering. The slow time-varying CSI between the base station (BS) and feasible UT locations in the service area is calculated from the geospatial data by offline ray tracing and a DNN-based path estimation model associated with the 1-dimensional convolutional neural network (1D-CNN) and regression tree ensembles. Then, the UT-level CSI of all feasible locations is grouped into clusters by a proposed FCM clustering. Finally, the service area is divided into a number of non-overlapping UT zones. Each UT zone is characterized by a corresponding set of clusters named as UT-group CSI, which is utilized in the analog RF beamformer design of AB-HP to reduce the required large online CSI overhead in the MU-mMIMO systems. Then, the reduced-size online CSI is employed in the baseband (BB) precoder of AB-HP. Simulations are conducted in the indoor scenario at 28 GHz and tested in an AB-HP MU-mMIMO system with a uniform rectangular array (URA) having 16 × 16 = 256 antennas and 22 RF chains. Illustrative results indicate that 91.4% online CSI can be reduced by using the proposed offline channel estimation technique as compared to the conventional online channel sounding. The proposed DNN-based path estimation technique produces same amount of UT-level CSI with runtime reduced by 65.8% as compared to the computationally expensive ray tracing. The imperfection of UT-level CSI introduced by the DNN-based path estimation technique is mitigated by the FCM clustering technique, where the AB-HP using offline UT-group CSI generated by the DNN-based channel estimation model and ray tracing based model for the RF beamformer achieves, respectively, 98.7% and 99.1% sum-rate performance of the fully digital precoding (FDP) technique using full-size online CSI.
EURASIP Journal on Advances in Signal Processing, May 7, 2007
This paper presents a high-performance, low-complexity, quasi-distributed dynamic spectrum manage... more This paper presents a high-performance, low-complexity, quasi-distributed dynamic spectrum management (DSM) algorithm suitable for DSL systems. We analytically demonstrate that the rate degradation of the distributed iterative waterfilling (IW) algorithm in near-far scenarios is caused by the insufficient utilization of all available frequency and power resources due to its nature of noncooperative game theoretic formulation. Inspired by this observation, we propose the selective IW (SIW) algorithm that can considerably alleviate the performance degradation of IW by applying IW selectively to different groups of users over different frequency bands so that all the available resources can be fully utilized. For N users, the proposed SIW algorithm needs at most N times the complexity of the IW algorithm, and is much simpler than the centralized optimal spectrum balancing (OSB), while it can offer a rate performance much better than that of the IW and close to the maximum possible rate region computed by the OSB in realistic near-far DSL scenarios. Furthermore, its predominantly distributed structure makes it suitable for DSL implementation.
This paper focuses on the design of vector perturbation (VP) precoding for coordinated multi-poin... more This paper focuses on the design of vector perturbation (VP) precoding for coordinated multi-point (CoMP) multiuser downlink transmission. Precoding is performed by individual base stations (BSs) in a distributed manner using only the downlink channel coefficients and user data local to a BS. A cascade precoder structure with an outer precoder managing the inter-cell interference (ICI) and an inner precoder performing mean-squared-error (MSE) minimization-based VP to mitigate the intra-cell interference is proposed. Three different outer precoding techniques are considered. In the first technique, the outer precoder is designed to fully eliminate the ICI by trading off the degrees of freedom (DoFs) available through multiple antennas. While the proposed technique outperforms existing conventional-VP based designs, a large portion of DoF is consumed by the ICI elimination. To overcome this issue, in the second technique, interference alignment-based outer precoding that minimizes the total leakage interference is proposed. To further improve the system performance, in the third approach, precoding by joint minimization of total leakage interference plus MSE is performed. Numerical results show that the proposed cascade precoding structure is an efficient way to use the DoF of CoMP multiuser downlink transmission. INDEX TERMS MIMO broadcast channel, vector perturbation, non-linear precoding, multiuser MIMO, coordinated multipoint transmission.
This paper presents a group-orthogonal OFDMA (GO-OFDMA) suitable for broadband access systems in ... more This paper presents a group-orthogonal OFDMA (GO-OFDMA) suitable for broadband access systems in a fast time-varying frequency-selective fading environment when channel knowledge is not available at the transmitter. Subcarrier grouping to achieve the diversity gain of a time-domain Rake receiver is discussed and the closed-form expression of its bit-error-rate (BER) performance over a frequencyselective Rayleigh fading channel is derived. The proposed GO-OFDMA scheme uses a split-and-group structure and a maximum-likelihood (ML) multiuser detection (MUD) to increase the number of supportable active users and to reduce the peak-to-average ratio (PAR). Analytical and simulation results are in an excellent agreement. Performance evaluation indicates that the proposed GO-OFDMA provides a lower PAR and similar BER as compared with the group-orthogonal multi-carrier CDMA (GO-MC-CDMA), and outperforms the random-hopping (RH)-OFDMA and matched-filter based MC-CDMA.
This paper proposes a capacity-approaching, yet simple scheme for multi-input multiple-output (MI... more This paper proposes a capacity-approaching, yet simple scheme for multi-input multiple-output (MIMO) channels. The proposed scheme is based on a concatenation of a mixture of short memory-length convolutional codes or repetition codes and a short, and simple rate-1 linear block code, followed by either 1-dimensional (1-D) anti-Gray or Gray mapping of quadrature phase-shift keying (QPSK) modulation. By interpreting the rate-1 code and the 1-D mapping as a multi-D mapping performed over multiple transmit antennas, the error performance is analyzed in two regions. In the error-floor region, a tight union bound and the corresponding design criterion on the asymptotic performance are derived. The bound provides a useful tool to predict the error performance at relatively low bit error rate (BER) values. Based on the obtained design criterion, an optimal rate-1 code for each 1-D mapping is then constructed to achieve the best asymptotic performance. In the turbo pinch-off region, by using extrinsic information transfer (EXIT) charts, the most suitable mixed codes are selected for both symmetric and asymmetric antenna configurations. It is demonstrated that the simple concatenation scheme can achieve a near-capacity performance over the MIMO channels. Furthermore, its error performance is shown to be comparable to that obtained by using well-designed irregular LDPC and RA codes, and therefore, the proposed scheme significantly outperforms a scheme employing a parallel concatenated turbo code. Simulation results in various cases are provided to verify the analysis.
This paper is concerned with linear precoding designs for multiuser downlink transmissions. We co... more This paper is concerned with linear precoding designs for multiuser downlink transmissions. We consider a multiple-input single-output (MISO) system with multiple single-antenna user equipment (UE) experiencing nonhomogeneous average signal-to-noise ratio (SNR) conditions. The first part of this work examines different precoding schemes with perfect channel state information (CSI) and average SNR at the base-station (eNB). We then propose a weighted minimum mean squared error (WMMSE) precoder, which takes advantage of the non-homogeneous SNR conditions. Given in a closed-form solution, the proposed WMMSE precoder outperforms other well-known linear precoders, such as zero-forcing (ZF), regularized ZF (RZF), while achieving a close performance to the locally optimal iterative WMMSE (IWMMSE) precoder, in terms of the achievable network sum-rate. In the second part of this work, we consider the non-homogeneous multiuser system with limited and quantized channel quality indicator (CQI) and channel direction indicator (CDI) feedbacks. Based on the CQI and CDI feedback models proposed for the Long-Term Evolution Advanced standard, we then propose a robust WMMSE precoder in a closed-form solution which takes into account the quantization errors. Simulation shows a significant improvement in the achievable network sum-rate by the proposed robust WMMSE precoder, compared to non-robust linear precoder designs.
This paper proposes a delay-aware resource provisioning policy for virtualized wireless networks ... more This paper proposes a delay-aware resource provisioning policy for virtualized wireless networks (VWNs) to minimize the total average transmit power while holding the minimum required average rate of each slice and maximum average packet transmission delay for each user. The proposed cross-layer optimization problem is inherently non-convex and has high computational complexity. To develop an efficient solution, we first transform cross-layer dependent constraints into physical layer dependent ones. Afterwards, we apply different convexification techniques based on variable transformations and relaxations, and propose an iterative algorithm to reach the optimal solution. Simulation results illustrate the effects of the required average packet transmission delay and minimum average slice rate on the total transmission power in VWN. Index Terms-Delay-aware resource provisioning, virtualized wireless networks.
Eurasip Journal on Wireless Communications and Networking, 2006
This paper presents a detailed performance analysis of adaptive downlink resource allocation base... more This paper presents a detailed performance analysis of adaptive downlink resource allocation based on users' instantaneous channel responses using power minimization (PM) and bandwidth constrained power minimization (BCPM) strategies. This study shows that, in cellular systems, where interference is a dominant factor, the link outage performance of a resource allocation strategy varies significantly depending on the user channel parameters. In particular, both analytical and simulation results indicate that the PM strategy outperforms BCPM in a mild shadowing environment. However, in severe shadowing conditions, this trend is reversed. This assessment holds true for both flat and frequency-selective fading.
Uploads
Papers by Tho Le-Ngoc