DL-based Adaptive Joint CSI Prediction and Feedback for Massive MIMO-OFDM Systems

In MIMO-OFDM communications, the channel state information (CSI) of the forward link is needed for precoder design at the transmitter to achieve the maximal performance gains. In frequency division duplex (FDD) systems, the CSI estimated at the receivers needs to be fed back to the transmitters. Because of the limited network resources, there will be CSI feedback errors due to quantization, delay and clustering (where one CSI feedback is used to represent a cluster of adjacent subcarriers for feedback reduction). Consequently, the system performance degrades and the gains expected from using MIMO diminish. To mitigate this performance degradation, an adaptive codebook-based CSI prediction and interpolation scheme is proposed for multiuser MIMO-OFDM systems. In this scheme, geodesic CSI prediction is employed at the receiver to mitigate the feedback delay and geodesic CSI interpolation is performed at the transmitter to mitigate the cluster feedback effect. Since the performance gain assumed by the CSI prediction and interpolation is limited by the low resolution CSI quantization, a simplified adaptive codebook is developed to support the CSI prediction and interpolation. Simulation results show that the proposed scheme is effective in mitigating the performance loss due to quantization, feedback delay and cluster feedback.

Multicell cooperative processing (MCP) can overcome the detrimental effects of inter-cell interference (ICI). However, since mobile terminals (MTs) need to estimate and feed back channel state information (CSI) related to several cooperative base stations (BSs), reductions in CSI feedback overhead and delay are more demanding in MCP enabled networks. In this paper, a novel overlaid-pilot based CSI feedback scheme for multicell cooperative networks is proposed. MTs directly overlay the received downlink pilot data of all the cooperative BSs on the uplink pilot data and then broadcast it to the cooperative BSs. Each cooperative BS can recover multiple CSIs from the overlaid pilot by a presented overlaid-pilot based channel estimation algorithm. Because the overlaid pilot is fed back on the uplink pilot channel, no extra uplink channel is occupied and the downlink CSI feedback process is with zero-overhead regardless of the number of cooperative BSs. In addition, as it just needs one broadcast to feed back multiple CSIs to BSs, the feedback delay is decreased significantly. Numerical simulations show that the performance of the proposed CSI feedback scheme outperforms the traditional feedback scheme.

Accurate channel state information (CSI) at base station (BS) is the key to achieve full multiplexing and diversity gain in traditional multiple-input multiple-output (MIMO) systems. In frequency-division-duplex (FDD) systems, it is obtained through downlink training and uplink feedback. However, the training and feedback overhead in massive MIMO systems is unacceptable for the large number of antennas equipped at BS. To reduce the CSI feedback overhead and improve CSI quantization accuracy, we propose a limited feedback scheme for spatio-temporal correlated massive MIMO channel. The proposed scheme jointly performs Karhunen-Loève transform (KLT) and polar-cap quantization on CSI. For the compactly deployed antennas at BS, there exists spatial correlation in massive MIMO channel. When the user equipment (UE) moves slowly, adjacent channels feature temporal correlation. The high dimensional and spatial correlated CSI is compressed through KLT. Then the compressed CSI is normalized to be isotropic to perform polar-cap quantization. Simulation results show that the proposed scheme is capable of improving the sum-rate performance with limited feedback bits for massive MIMO systems.

Accurate downlink channel state information (CSI) is required to be fed back to the base station (BS) in frequency division duplexing (FDD) massive multiple-input multiple-output (MIMO) systems in order to achieve maximum antenna diversity and multiplexing. However, downlink CSI feedback overhead scales with the number of transceiver antennas, a major hurdle for practical deployment of FDD massive MIMO systems. To solve this problem, we propose a compressive sampled CSI feedback method based on deep learning (SampleDL). In SampleDL, the massive MIMO channel matrix is sampled uniformly in time/frequency dimension before being fed into neural networks (NNs), which will reduce the computational resource/time at user equipment (UE) as well as enhance the CSI recovery accuracy at the BS. Both theoretical analysis and normalized mean square errors (NMSE) results confirm the advantages of the proposed method in terms of time complexity and recovery accuracy. Besides, a suitable CSI feedback period is explored by link level simulations, which aims to further reduce the overhead of CSI feedback without degrading the communication quality.

Cell-free system where a group of base stations (BSs) cooperatively serves users has received much attention as a promising technology for the future wireless systems. In order to maximize the cooperation gain in the cell-free systems, acquisition of downlink channel state information (CSI) at the BSs is crucial. While this task is relatively easy for the time division duplexing (TDD) systems due to the channel reciprocity, it is not easy for the frequency division duplexing (FDD) systems due to the CSI feedback overhead. This issue is even more pronounced in the cell-free systems since the user needs to feed back the CSIs of multiple BSs. In this paper, we propose a novel feedback reduction technique for the FDD-based cell-free systems. Key feature of the proposed technique is to choose a few dominating paths and then feed back the path gain information (PGI) of the chosen paths. By exploiting the property that the angles of departure (AoDs) are quite similar in the uplink and downlink channels (this property is referred to as angle reciprocity), the BSs obtain the AoDs directly from the uplink pilot signal. From the extensive simulations, we observe that the proposed technique can achieve more than 60% reduction in feedback overhead over the conventional CSI feedback scheme.

The work of cooperative perception realized by mmWave Vehicle to everything (V2X) [3] has pointed out that vehicles require much network bandwidth to exchange sensor information for safe automated driving; e.g., they require higher than 1Gbps V2V bandwidth to ensure safe velocity of 70 km/h. On the other hand, some research works have shown that channel state information (CSI) feedback is able to mitigate interference for 5G frequency division duplexing (FDD) mmWave massive MIMO systems. Existing approaches either do not consider the characteristics of FDD mmWave V2X systems or assume all channels are ideal, i.e., uncorrelated and identical. To achieve higher throughput and low overhead for FDD mmWave V2X systems, there are two major issues: (1) how to acquire an accurate CSI under the environment of non-identical channels, and (2) how to effectively use such information to improve overall system throughput. Our idea is to jointly consider the characteristics of the angle of arrival (AoA) method and Khler manifolds to design of codebook feedback approach for mmWave V2X massive systems. That is, to get an accurate CSI, we leverage the measurement method with the idea of the AoA approach. With this AoA measurement, we propose a novel Khler manifolds codebook feedback approach and determine an optimal bit allotment over each mmWave V2X link for achieving better interference reduction as well as lower overhead for V2X CSI exchange. In addition, our analytic results validate that proposed AoA Khler manifolds codebook feedback scheme is more adequate than the traditional CSI feedback with regard to interference reduction. Simulations show that the proposed AoA Khler manifolds codebook feedback scheme can significantly improve the vehicle rate and reduce the overhead for V2V CSI exchange.

Massive multiple-input multiple-output (MIMO) systems are a main enabler of the excessive throughput requirements in 5G and future generation wireless networks as they can serve many users simultaneously with high spectral and energy efficiency. To achieve this massive MIMO systems require accurate and timely channel state information (CSI), which is acquired by a training process that involves pilot transmission, CSI estimation, and feedback. This training process incurs a training overhead, which scales with the number of antennas, users, and subcarriers. Reducing the training overhead in massive MIMO systems has been a major topic of research since the emergence of the concept. Recently, deep learning (DL)-based approaches have been proposed and shown to provide significant reduction in the CSI acquisition and feedback overhead in massive MIMO systems compared to traditional techniques. In this paper, we present an overview of the state-of-the-art DL architectures and algorithms used for CSI acquisition and feedback, and provide further research directions.

In cellular systems, deploying a large number of antennas at the base station (BS), also called massive multiple-input multiple-output (MIMO), can offer a huge improvement in system throughput. This performance gain achieved through coherent beamforming highly depends on having accurate channel state information (CSI) at the BS. Initially, massive MIMO was considered promising only for time division duplexing (TDD) systems because the downlink training overhead in frequency division duplexing (FDD) systems could be large. To reduce the overhead in FDD systems, we propose efficient eigenspace training and precoding. In the downlink training, the prebeamforming matrix is designed based on the spatial fading correlation to probe the channel and minimize the mean squared error (MSE) in channel estimation. In data transmission, the precoding matrix is designed based on instantaneous CSI to manage the inter-user interference. Our algorithms achieve significant savings in the downlink training because the overhead is associated with the dimension of prebeamforming matrix and no longer limited by the number of BS antennas. Compared with joint spatial division and multiplexing, our algorithms offer much lower MSE and, under some conditions, higher sum rates.

In the multiuser multiple-input multiple-output (MU-MIMO) system, to reduce the influence of channel correlation on system performance, the base station (BS) should select the appropriate subset of user equipments (UEs) according to their channel state information (CSI). Due to a lack of channel reciprocity, the downlink CSI needs to be fed back to the BS in frequency division duplexing (FDD) mode. Some scholars have exploited kinds of deep neural networks (DNNs) for sensing and recovering CSI. However, user selection after all the CSI is reconstructed by DNNs will bring a great time delay. In this paper, we propose a deep learning-based CSI feedback scheme called US-CsiNet. Based on adversarial autoencoder (AAE), US-CsiNet can explicitly cover user schedule information while representing CSI. At the UE side, the encoder of US-CsiNet maps the CSI into codewords of which part are feature information for user schedule. Then the BS applies these partial codewords to separate the UEs into different groups and select active UEs. Finally, the decoder of AAE reconstructs the CSI of these active UEs. US-CsiNet can not only simplify the user selection process but also guarantee the accuracy of CSI reconstruction. The simulation results show that the proposed approach outperforms maximum channel gain (MCG) user selection algorithms and achieves the nearly same performance with semiorthogonal user selection (SUS) which needs full CSI of all users at the BS.

In a massive multiple-input multiple-output (MIMO) system, channel state information (CSI) is essential for the base station (BS) to achieve high performance gain. The user equipment (UE) needs to estimate CSI and then feeds it back to the BS in the frequency division duplexing (FDD) mode. Effective compression of CSI will significantly reduce the cost of channel feedback, and many deep learning (DL) based channel compression schemes have been proposed to achieve this goal. In this paper, rather than viewing CSI as images as in most existing works, we propose a new perspective of viewing CSI as an information sequence and analogize CSI feedback to a machine translation task. Further, we propose a novel sequence to sequence (Seq2Seq) model for CSI feedback composed of only recurrent neural networks and a small-scale fully connected layer, avoiding the convolution and pooling structure commonly used in current DL-based works. The advantage of this scheme is that it fully integrates the physical characteristics of the MIMO channel in the spatial domain into the structure of the neural network and avoids the loss of some prominent unsmooth or discontinuous features caused by the inappropriate convolution and pooling operation. Simulation results show that the proposed Seq2Seq model outperforms other DL-based CSI compression techniques under various communication scenarios.

The demand for higher data traffic is exponentially growing. Since last 2 years or so, new research movements have emerged towards LTE-B (Release 13 and beyond) and 5th generation (5G) networks. As the next generation of cellular networks are going to be an integration of different radio access technologies and denser in term of space and frequency, the problem of interference management becomes more vital and challenging for academic and industrial units. One of the main technical challenges of interference management in the frequency division duplex system is the issue of channel state information (CSI) feedback overhead. Accordingly, this article briefly overviews the key candidate technologies for LTE-B and 5G and as well as to addresses their features and problems from the interference management perspective. Then, the problem of limited feedback in the multi-user MIMO systems is formulated and discussed. Afterward, the performance of distributed antenna systems is investigated under imperfect CSI and different phases of limited feedback technique. It is shown that accuracy of CSI feedback plays a vital role towards the realization of next generation of cellular systems. Finally, the future research directions are addressed in this domain.

Deep learning (DL)-based channel state information (CSI) feedback improves the capacity and energy efficiency of massive multiple-input multiple-output (MIMO) systems in frequency division duplexing mode. However, multiple neural networks with different lengths of feedback overhead are required by time-varying bandwidth resources. The storage space required at the user equipment (UE) and the base station (BS) for these models increases linearly with the number of models. In this paper, we propose a DL-based changeable-rate framework with novel quantization scheme to improve the efficiency and feasibility of CSI feedback systems. This framework can reutilize all the network layers to achieve overhead-changeable CSI feedback to optimize the storage efficiency at the UE and the BS sides. Designed quantizer in this framework can avoid the normalization and gradient problems faced by traditional quantization schemes. Specifically, we propose two DL-based changeable-rate CSI feedback networks CH- CsiNetPro and CH- DualNetSph by introducing a feedback overhead control unit. Then, a pluggable quantization block (PQB) is developed to further improve the encoding efficiency of CSI feedback in an end-to-end way. Compared with existing CSI feedback methods, the proposed framework saves the storage space by about 50% with changeable-rate scheme and improves the encoding efficiency with the quantization module.

Reconfigurable intelligent surfaces (RISs) represent a pioneering technology to realize smart electromagnetic environments by reshaping the wireless channel. Jointly designing the transceiver and RIS relies on the channel state information (CSI), whose feedback has not been investigated in multi-RIS-assisted frequency division duplexing systems. In this study, the limited feedback of the RIS-assisted wireless channel is examined by capitalizing on the ability of the RIS in channel customization. By configuring the phase shifters of the surfaces using statistical CSI, we customize a sparse channel in rich-scattering environments, which significantly reduces the feedback overhead in designing the transceiver and RISs. Since the channel is customized in terms of singular value decomposition (SVD) with full-rank, the optimal SVD transceiver can be approached without a matrix decomposition and feeding back the complete channel parameters. The theoretical spectral efficiency (SE) loss of the proposed transceiver and RIS design is derived by considering the limited CSI quantization. To minimize the SE loss, a bit partitioning algorithm that splits the limited number of bits to quantize the CSI is developed. Extensive numerical results show that the channel customization-based transceiver with reduced CSI can achieve satisfactory performance compared with the optimal transceiver with full CSI. Given the limited number of feedback bits, the bit partitioning algorithm can minimize the SE loss by adaptively allocating bits to quantize the channel parameters.

Chapter 7 - Channel feedback for mmWave massive MIMO

In massive multiple-input multiple-output (MIMO) systems, the channel state information (CSI) feedback enables performance gain in frequency division duplex networks. However, with the increase in the number of antennas, the feedback overhead of CSI will also enhance. To this end, this study addresses the issue of massive MIMO CSI reconstruction using convolutional neural network (CNN), long short-term memory (LSTM) and attention mechanism, and proposes an efficient network architecture (denoted as CNN-LSTM-A). To achieve a compromise between performance and complexity, the proposed method significantly reduces the number of training parameters by utilising a single-stage network rather than a multiple-stage network. Finally, simulation results show that the authors method can reduce the feedback overhead of CSI effectively, and achieves better performance in terms of CSI compression and recovery accuracy compared with existing state-of-the-art methods.