Deep Learning based CSI Reconstruction with Limited Feedback for Massive MIMO Systems

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.

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.

Due to the limitation of digital chains' number, the hybrid beamforming structure may be utilized for massive Multiple Input and Multiple Output (MIMO) systems, where the eNodeB may maintain a number of analog Tx beams. To achieve higher spectral efficiency, two analog beam interpolation methods are further proposed to enhance the accuracy of downlink analog beamforming, which based on amplitude or angle interpolation algorithm. In this paper, the typical channel state information(CSI) from user equipment (UE) is used to determine the candidate Transmitting (Tx) beams. Especially, by exploiting the spatial correlation of neighbor candidate beams, the best downlink analog Tx beam for each UE can be generated by interpolating the two strongest Tx beams from the Uplink Control Informat ion (UCI) report. Moreover, we develop a limited CSI feedback to alleviate the feedback load in traditional frequency division duplexing (FDD) Massive-MIMO systems, giving rise to better coverage and spectrum efficiency. Simulation results show that with the help of analog beam interpolation, the downlink spectrum efficiency can be significantly increased.

In Multi-user MIMO (MU-MIMO) downlink system, suitable user selection schemes can improve spatial diversity gain. In most of previous studies, it is always assumed that the base station (BS) knows full channel state information (CSI) of each user, which does not consider the reality. However, there are only limited feedback bits in real system. Besides, user fairness is often ignored in most of current user selection schemes. To discuss the user fairness and limited feedback, in this paper, the user selection scheme with limited feedback bits is proposed. The BS utilizes codebook precoding transmitting strategy with LTE codebook. Furthermore, this paper analyzes the influence of the number of feedback bits and the number of users on user fairness and system sum capacity. Simulation results show that in order to achieve better user fairness, we can use fewer bits for feedback CSI when the number of user is small, and more feedback bits when the number of users is large.

A multihop regenerative relay system is considered. We explore adaptive power allocation (PA) under a total power constraint to minimize the system error probability when the perfect channel state information (CSI) is known at the transmitters. We propose adaptive PA schemes with limited feedback when the transmitters have no perfect knowledge of CSI. Simulation results demonstrate the good performance of the proposed PA scheme with perfect CSI. When no perfect CSI is available at the transmitters, the proposed PA schemes with a small number of feedback bits are able to achieve very close performance to the adaptive PA with perfect CSI.

Channel state information (CSI) is crucial for down link beamnforming in massive multiple-input multiple-output orthogonal frequency division multiplexing (MIMO-OFDM) systems. However, obtaining timely and accurate CSI remains a challenge in frequency division duplex (FDD) systems due to the time-varying channel and feedback delay. This paper proposes a deep learning (DL)-based joint prediction and feedback network for eigenvector-based CSI, termed as PCRANet, which could leverage history CSI to predict and feedback future CSI. On one hand, the PCRANet could fit for different subband numbers of input CSI. On the other hand, to enable flexible input lengths of temporal-domain CSI sequences for various UE speeds, padding and masking operations are employed to preprocess input CSI sequences. Furthermore, a feedback bit control unit (FBCU) module is introduced to realize the variable feedback overhead. Experiments indicate that the proposed PCRANet outperforms separated prediction-feedback schemes, and maintains robust performance across various temporal-frequency sizes of input CSI and adjustable feedback overhead.

Channel state information (CSI) at the base station (BS) is crucial to achieve beamforming and multiplexing gains in multiple-input multiple-output (MIMO) systems. State-of-the-art limited feedback schemes require feedback overhead that scales linearly with the number of BS antennas, which is prohibitive for 5G massive MIMO. This paper proposes novel limited feedback algorithms that lift this burden by exploiting the inherent sparsity in double-directional MIMO channel representation using overcomplete dictionaries. These dictionaries are associated with angle of arrival and angle of departure that specifically account for antenna directivity patterns at both ends of the link. The proposed algorithms achieve satisfactory channel estimation accuracy using a small number of feedback bits, even when the number of transmit antennas at the BS is large-making them ideal for 5G massive MIMO. Judicious simulations reveal that they outperform a number of popular feedback schemes and underscore the importance of using angle dictionaries matching the given antenna directivity patterns, as opposed to uniform dictionaries. The proposed algorithms are lightweight in terms of computation, especially on the user equipment side, making them ideal for actual deployment in 5G systems.

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.

It is very important that the radar sensor network has autonomous capabilities such as self-managing, etc. Quite often, MIMO interference channels are applied to radar sensor networks, and for self-managing purpose, interference management in MIMO interference channels is critical. Interference alignment (IA) has the potential to dramatically improve system throughput by effectively mitigating interference in multi-user networks at high signal-to-noise (SNR). However, the implementation of IA predominantly relays on perfect and global channel state information (CSI) at all transceivers. A large amount of CSI has to be fed back to all transmitters, resulting in a proliferation of feedback bits. Thus, IA with limited feedback has been introduced to reduce the sum feedback overhead. In this paper, by exploiting the advantage of heterogeneous path loss, we first investigate the throughput of IA with limited feedback in interference channels while each user transmits multi-streams simultaneously, then we get the upper bound of sum rate in terms of the transmit power and feedback bits. Moreover, we propose a dynamic feedback scheme via bit allocation to reduce the throughput loss due to limited feedback. Simulation results demonstrate that the dynamic feedback scheme achieves better performance in terms of sum rate.

Massive MIMO has been proposed as a means of achieving huge spectral efficiency in cellular networks. However, its implementation is usually based on linear precoding schemes using Channel State Information (CSI). For this reason, the use of the Time Division Duplex (TDD) mode is typically assumed since it removes the dependence on the number of antennas at the transmitter to acquire CSI. Unfortunately, most of the cellular network operate in the Frequency Division Duplex (FDD) mode in which the acquisition of CSI depends proportionally to the number of transmit antennas. For this reason, the implementation of Massive MIMO is handicapped to few transmit antennas for the current cellular networks. In this paper we study the use of reconfigurable antennas that can switch among a set of preset modes, i.e., radiation patterns, through Blind Interference Alignment schemes. We show that it is a solution to exploit a very large number of antennas at the transmit side, overcoming the bottleneck that today constitutes the acquisition of CSI in Massive MIMO systems operating in FDD mode.

Channel state information (CSI) acquisition is a crucial issue in downlink FDD-based massive multi-input multioutput (MIMO) networks, where the channel reciprocity is not applicable. Thus, users are expected to feedback the bestmatch quantized channels to serving transmitters. Hence, an extensively large size of the feedback overhead is needed, which is linearly scaled at each user with the number of transmit antennas at the base-station (BS). In turn, the uplink (UL) channel capacity may be consumed and the overall performance becomes fundamentally limited by the downlink (DL) channel quantization precision. An alternative CSI acquisition scheme is critically needed. In this paper, we propose a novel FDD massive MIMO system based on a spatial DL channel estimation scheme; it relies on the statistical spatial correlation of the UL and DL channel clusters, given an arbitrary frequency band gap between the UL and DL channels. A transformation matrix is constructed to precode the observed UL channel on the estimated dominant DL angles of departure. The proposed scheme significantly outperforms the recent state-of-the-art techniques, without the cost of user feedback overhead bits and prior knowledge of the channel statistics.

Interference mitigation by applying Joint Processing (JP) in cooperative cellular networks boasts cell-edge performance compared to non-cooperative systems. However, downlink JP requires knowledge of Channel State Information (CSI) at all collaborating Base Stations (BSs), where interference received at the User Equipments (UEs) is markedly affected by the level of CSI quality influenced amongst others by the method of distributing CSI to the BSs. In this contribution, we compare CSI Distribution over the Air (CD-A) with CSI Distribution over the Backhaul (CD-B) considering a Frequency Devision Duplex (FDD) system. For CD-A, CSI is directly fed back from the UEs to the collaborating BSs via the uplink channel (assuming link adaptation is performed according to one master BS to which a UE has been assigned). Adopting CD-B, CSI is foremost transmitted only to the master BS and then forwarded to the other BSs using backhaul connections. This method typically introduces additional latency due to routing issues. In this contribution, we show that CD-A outperforms CD-B in the largest part of the cooperation area, even when pedestrian user velocities and small backhaul latencies are considered.

Pilot-assisted channel training for acquiring the channel state information (CSI) is a very critical functionality in URLLC to ensure reliable data transmission. In this paper, we consider the joint uplink and downlink URLLC transmission in which the source uploads a packet to the BS, and then BS forwards this packet to the destination. CSI in the uplink can be estimated at BS by sending a training pilot from source while for the downlink, an extra codebook-based channel quantization is required at destination to measure the estimated channel and feedback the limited-precision results to BS for beamforming before data transmission. This could bring about channel estimation errors, quantization errors and decoding errors, respectively caused by limited training symbols, limited feedback overheads and finite blocklength. In view of this, we for the first time investigate the transmission performance of URLLC under this circumstance and design a performance optimization framework to maximize the transmission reliability. Particularly, the impact of channel estimation accuracy on the transmission reliability of URLLC is studied based on the finite blocklength information theory. An analytical closed-form expression of transmission error probability under imperfect CSI is also derived with respect to key parameters, like the training pilot length, feedback bits and data transmission durations. By jointly optimizing those parameters, we could derive the maximum transmission reliability. Simulation results are provided to validate the effectiveness of our proposed performance optimization framework.

This paper proposes an efficient two-stage beamforming and scheduling algorithm for the limited-feedback cooperative multi-point (CoMP) systems. The system includes multiple base-stations cooperatively transmitting data to a pool of users, which share a rate-limited feedback channel for sending back the channel state information (CSI). The feedback mechanism is divided into two stages that are used separately for scheduling and beamforming. In the first stage, the users report their best channel gain from all the base-station antennas and the basestations schedule the best user for each of their antennas. The scheduled users are then polled in the second stage to feedback their quantized channel vectors. The paper proposes an analytical framework to derive the bit allocation between the two feedback stages and the bit allocation for quantizing each user's CSI. For a total number of feedback bits B, it is shown that the number of bits assigned to the second feedback stage should scale as log B. Furthermore, in quantizing channel vectors from different base-stations, each user should allocate its feedback budget in proportion to the logarithm of the corresponding channel gains. These bit allocation are then used to show that the overall system performance scales double-logarithmically with B and logarithmically with the transmit SNR. The paper further presents several numerical results to show that, in comparison with other beamforming-scheduling algorithms in the literature, the proposed scheme provides a consistent improvement in downlink sum rate and network utility. Such improvements, in particular, are achieved in spite of a significant reduction in the beamforming-scheduling computational complexity, which makes the proposed scheme an attractive solution for practical system implementations.

In multiuser massive MIMO systems, it is not clear whether users should feed back the channel or the precoder when they can exchange the channel state information (CSI). This paper compares the precoder feedback scheme with the channel feedback scheme. It is found that when there are sufficient number of bits for CSI exchange, the precoder feedback scheme can reduce the interference leakage to 1 /(K − 1) of the channel feedback scheme, where K is the number of users. Moreover, the interference leakage under the precoder feedback scheme decreases faster than the channel feedback scheme when the number of feedback bits increases.