Super-Resolution Channel Estimation for MmWave Massive MIMO With Hybrid Precoding

Hu, Chen; Dai, Linglong; Mir, Talha; Gao, Zhen; Fang, Jun

doi:10.1109/tvt.2018.2842724

Cited by 152 publications

(134 citation statements)

References 9 publications

Supporting

Mentioning

133

Contrasting

Order By: Relevance

“…Figure 8 illustrates a comparison of TLS, VS, and unshaped SOMP-based methods when their common parameters are set to N t = 16, N r = 8, N f = 8, N fft = 128 while their training numbers and number of path varies. 19 In this value, of course, the performance is slightly better than the low-complexity TLS method. This superiority is more noticeable as the number of training and path increases.…”

Section: Tls Methodsmentioning

confidence: 92%

“…This procedure continues for M F beamforming matrices. On the other hand, the surrogate method is dependent on the gradient algorithm which 19], which will provide a compromise between the complexity and the estimation performance. On the other hand, the surrogate method is dependent on the gradient algorithm which 19], which will provide a compromise between the complexity and the estimation performance.…”

Section: Tls Methodsmentioning

confidence: 99%

“…The sub-optimum solution of Equation 19 is achieved by a two-stage generalized eigenvalue problem, but due to the space limitation, the procedure for finding the shaping vectors is brought in. The sub-optimum solution of Equation 19 is achieved by a two-stage generalized eigenvalue problem, but due to the space limitation, the procedure for finding the shaping vectors is brought in.…”

Section: Vector Shaping Approach (Vs) With Smv Modelmentioning

confidence: 99%

“…8 Deployment of the orthogonal frequency division multiplexing modulation with a reasonable cyclic prefix contributes to a narrowband channel in each subcarrier along with inter-symbol interference cancelation. For instance, Fang et al and Hu et al 18,19 considered a surrogate function instead of the l 1 optimization in the CS method and introduced a gradient descent method to estimate the parameters of the channel with a high resolution. In other words, the physical frequency-selective MIMO channel in mmW can be accurately modeled by a few multipath which comprises the complex gain, angle of departure at the transmitter, angle of arrival at the receiver, and delay spread in each path.…”

Section: Introductionmentioning

confidence: 99%

“…19] and calculation of other computationally expensive com-ponent ["ζ opt [k]",19] at each iteration. In brief, those complexities could be reduced by the introduction of a threshold ["γ th ¼ Cσ 2…”

mentioning

confidence: 99%

See 4 more Smart Citations

The off‐grid frequency selective millimeter wave channel estimation

Dastgahian

Tehrani

Khoshbin

2018

Int J Communication

View full text Add to dashboard Cite

Broadband millimeter wave (mmW) systems are a promising pioneer of cellular communication for next generation which is utilizing the hybrid baseband/ analog beamforming structures along with the miniature massive antenna arrays at both sides of the communication link. mmW channel with an available unlicensed spread spectrum is frequency selective because the signal bandwidth can be larger than the coherence bandwidth. Due to the sparse nature of mmW channel, extracting compressive sensing model of the system is preferable. In fact, exploiting the sparse structure will lead to the reduction of the computational complexity, because there is a reduction in the channel training length compared with the conventional methods such as least square estimation. Most of the prior works have considered on-grid quantized departure/arrival angles in the input/output antennas to obtain a sparse virtual channel model. However, the sparse angles in the physical channel model are continuous where this continuity indicates a mismatch between the physical angles and the on-grid angles. Such a mismatch will contribute to unwanted components in the virtual channel model. Given these extra components, the conventional compressive sensing tools are unable to recover the channel. In this paper, we propose two solutions for overcoming the problem caused by off-grid angle selection. The first is based on the vector shaping, and the second one is based on the sparse total least square concepts. Simulation results demonstrate that the proposed methods both could obtain an adequate channel recovery and are preferable regarding computational complexity concerning the newly developed surrogate method. KEYWORDScompressive sensing (CS), millimeter wave (mmW), off-grid, on-grid quantization, total least square (TLS), vector shaping (VS)

show abstract

Section: Tls Methodsmentioning

confidence: 92%

Section: Tls Methodsmentioning

confidence: 99%

Section: Vector Shaping Approach (Vs) With Smv Modelmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations

The off‐grid frequency selective millimeter wave channel estimation

Dastgahian

Tehrani

Khoshbin

2018

Int J Communication

View full text Add to dashboard Cite

show abstract

Ensemble learning‐based channel estimation and hybrid precoding for millimeter‐wave massive multiple input multiple output system

Reddy

2023

Trans Emerging Tel Tech

View full text Add to dashboard Cite

Millimeter‐Wave (mmWave) massive Multi‐User Multiple Input Multiple Output (MU‐MIMO) utilizes the hybrid analog precoding design helps to minimize the amount of Radio‐Frequency (RF) chains without causing loss of sum‐rate performance. The Channel Estimation (CE) is designed to demand the conditions of channels with high time‐varying features. The existing CE frameworks are extremely complex. The accurate CE is still challenging in mmWave massive MIMO system based on the hardware constraints, a maximum number of antennas, and so on. Various proficient algorithms were combined to yield reliable channel estimates. The mmWave massive MIMO system utilizes hybrid precoding to minimize energy consumption loss and reduce hardware complexity. The fully connected layer in hybrid precoding results in high energy efficiency loss. This proposed work aims to develop two ensemble deep learning models for CE and hybrid precoding of the “mmWave massive MIMO system.” Here, the estimation of the channel is performed by a “deep Convolutional Neural Network (CNN) with a Recurrent Neural Network (RNN).” With the reconstructed channel, the hybrid precoding phase is performed by deep CNN with AdaBoost. The performance of CE and hybrid precoding is improved with Self Adaptive Searched Galactic Swarm Optimization (SAS‐GSO) by attaining the maximum spectral efficiency and also minimizing the Normalized Mean‐Squared Error (NMSE). Hence, the simulation result is shown that the proposed model achieves enhanced performance. Throughout the analysis, at the 20th iteration, the SAS‐GSO‐ensemble approach gives a lower NMSE rate, which is 31%, 25.9%, 20%, 19.6%, 19.3%, 18.6%, and 18.3% superior to OMP‐Ensemble, AMP‐Ensemble, DLCS‐Ensemble, DGMP‐Ensemble, QALS‐Ensemble, DLQP‐Ensemble, and CSO‐Ensemble, respectively, for hybrid precoding. Therefore, the offered mmWave massive MIMO is achieved better performance regarding diverse error metrics.

show abstract

Effective Utilization of Hybrid Residual Modules in Deep Neural Networks for Super Resolution

Muqeet

Bae

2019

MultiMedia Modeling

View full text Add to dashboard Cite

Super-Resolution Channel Estimation for MmWave Massive MIMO With Hybrid Precoding

Cited by 152 publications

References 9 publications

The off‐grid frequency selective millimeter wave channel estimation

The off‐grid frequency selective millimeter wave channel estimation

Ensemble learning‐based channel estimation and hybrid precoding for millimeter‐wave massive multiple input multiple output system

Effective Utilization of Hybrid Residual Modules in Deep Neural Networks for Super Resolution

Contact Info

Product

Resources

About