APCNet: A multi-scale pooling enhanced all-domain joint CSI feedback network for massive MIMO systems

Recently the combination of massive multiple-input multiple-output (MIMO) and spatial modulation (SM), has been considered as a promising concept for uplink transmission, in which each user equipment (UE) uses SM for uplink transmission and base station (BS) is equipped with massive antennas. In this paper, we evaluate a massive single-carrier (SC) SM-MIMO system with frequency domain equalization, where SC transmission is combined with SM (SC-SM) to combat the negative impact of broadband frequency-selective fading, and frequency domain equalization is utilized to mitigate the intersymbol-interference with a low complexity. With frequency domain processing, a framework is proposed to analyze the achievable uplink spectral efficiency (SE) of single-cell massive SC-SM MIMO systems. Based on this framework, the closed-form SE lower bound of frequency domain maximum ratio combining is derived, and both the derivation of framework and SE lower bound are much more complicated than those of systems with time domain combining. Monte Carlo simulations verify the tightness of proposed SE lower bound, and show that massive SC-SM MIMO systems can outperform the SE of conventional single transmit antenna (TA) massive MIMO systems. The systems can even have a better SE performance than massive MIMO systems with spatial multiplexing UEs in a low signal-to-noise ratio. Finally, the SE gain is found to be mainly dependent on the specific number of UE's TAs, which facilitates an SE maximization via optimizing the number of TAs.

One of focuses in green communication studies is the energy efficiency (EE) of massive multiple-input multiple-output (MIMO) systems. Although the massive MIMO technology can improve the spectral efficiency (SE) of cellular networks by configuring a large number of antennas at base stations (BSs), the energy consumption of radio frequency (RF) chains increases dramatically. The increment of energy consumption is caused by the increase of RF chain number to match the antenna number in massive MIMO communication systems. To overcome this problem, a generalized spatial modulation (GSM) solution is presented to simultaneously reduce the number of RF chains and maintain the SE of massive MIMO communication systems. A EE model is proposed to estimate the transmission and computation power of massive MIMO communication systems with GSM. Simulation results demonstrate that the EE of massive MIMO communication systems with GSM outperforms the massive MIMO communication systems without GSM. Besides, the computation power consumed by massive MIMO communication systems with GSM is effectively reduced.

High data rate transfers, high-definition streaming, high-speed internet, and the expanding of the infrastructure such as the ultra-broadband communication systems in wireless communication have become a demand to be considered in improving quality of service and increase the capacity supporting gigabytes bitrate. Massive Multiple-Input Multiple-Output (MIMO) systems technology is evolving from MIMO systems and becoming a high demand for fifth-generation (5G) communication systems and keep expanding further. In the near future, massive MIMO systems could be the main wireless systems of communications technology and can be considered as a key technology to the system in daily lives. The arrangement of the huge number of antenna elements at the base station (BS) for uplink and downlink to support the MIMO systems in increasing its capacity is called a Massive MIMO system, which refers to the vast provisioning of antenna elements at base stations over the number of the single antenna of user equipment. Massive MIMO depends on spatial multiplexing and diversity gain in serving users with simple processing signal of uplink and downlink at the BS. There are challenges in massive MIMO system even though it contains numerous number of antennas, such as channel estimation need to be accurate, precoding at the BS, and signal detection which is related to the first two items. On the other hand, in supporting wideband cellular communication systems and enabling low latency communications and multi-gigabit data rates, the Millimeter-wave (mmWave) technology has been utilized. Also, it is widely influenced the potential of the fifth-generation (5G) New Radio (NR) standard. This study was specifically review and compare on a few designs and methodologies on massive MIMO antenna communication systems. There are three limitations of those antennas were identified to be used for future improvement and to be proposed in designing the massive MIMO antenna systems. A few suggestions to improve the weaknesses and to overcome the challenges have been proposed for future considerations.

This paper first investigates an M-by-2 massive multiple-input multiple-output (MIMO) system that transmits a single stream is investigated. For this system, we propose a space-time line code (STLC), which is a transmitting and combining (at a receiver) scheme that achieves full spatial diversity. For the STLC, two consecutive (time) information symbols are weighted as per channel gains (space), combined at each transmit antenna, and transmitted through the M transmit antennas for two consecutive symbol times. With two receive antennas, the STLC receiver simply combines the signals received in the two symbol times and achieves a diversity order of 2M (full diversity). We show that the proposed STLC asymptotically achieves the maximum (optimal) received signal-to-noise ratio as M increases with significantly reduced computational complexity compared with the optimal scheme. Because the proposed STLC receiver requires no or partial channel state information, it avoids the issue of massive MIMO channel estimation. Furthermore, the rigorous performance evaluation under spatially correlated and uncertain channel conditions reveals that the proposed STLC achieves comparable or better performance than the existing schemes, and the results verify that the proposed STLC scheme is a potential candidate for M-by-2 massive MIMO systems. Next, the transmit antenna allocation algorithms are devised for a multiuser STLC system. Each user achieves full diversity order from the corresponding MIMO channels after the antenna allocation. The signal-tointerference-plus-noise ratio (SINR) of each user is analyzed considering the multiuser interference and channel uncertainty, and its lower bound is derived. Using the SINR lower bound, greedy algorithms that allocate the transmit antennas are devised. Rigorous simulation demonstrates that multiuser STLC with the proposed antenna allocation is robust against channel uncertainty and can improve the average SINR, improving the quality of experience. Furthermore, it is observed that the proposed STLC with antenna allocation method achieves the best performance if M is sufficiently large. The results in this paper show that the STLC can be a potential candidate for an M-by-2 (multiuser) massive MIMO systems.

In this paper, we investigate the suitability of single carrier frequency domain processing (SC-FDP) as the downlink transmission scheme in a massive multiple-input multiple-output (MIMO) system. By deriving the sum-rate of the SC-FDP massive MIMO system theoretically, we show that this method obtains a sum-rate similar to that of orthogonal frequency division multiplexing (OFDM) massive MIMO. We also derive the theoretical sum-rate of both SC-FDP and OFDM in a non-synchronized massive MIMO scenario and show that the rate of the former is significantly larger than that of the latter. Moreover, we theoretically analyze the sum-rate of both systems in the presence of power amplifier non-linearity. All the sum-rates are derived for both zero forcing and matched filter precoding schemes. The results show that the effect of power amplifier non-linearity on the sum-rate of both systems is similar when the number of users is large. We also compare SC-FDP with OFDM from the peak to average power ratio (PAPR) and complexity viewpoints. Although the PAPR of SC-FDP signals is lower than that of OFDM signals, for MIMO systems, the difference between their PAPR decreases as we increase the number of users. Thus, both techniques can have similar PAPR in massive MIMO systems. The overall complexity of SC-FDP and OFDM is similar. Due to the mentioned facts, SC-FDP can be considered as a promising transmission scheme for the downlink of the massive MIMO systems in the presence of carrier frequency offset and power amplifier non-linearities.

This paper concerns with the cascaded channel estimation of massive multiple-input multiple-output (MIMO) relay system. In order to meet the increasing demands for high-speed wireless communication networks, massive MIMO has been recognized as one of the key technologies for the future fifth generation (5G) cellular networks. It is an advanced MIMO technique consists of a very large number of antennas at the base station and serves a smaller number of single-antenna users simultaneously. Basically, the idea of massive MIMO technique is to harvest all the advantages of conventional MIMO system in a much larger scale. To reap the benefits of massive MIMO in practice, an accurate estimation of the channel state information (CSI) is needed. In this paper, the relaying technique has been incorporated with massive MIMO system in order to increase system throughput and improve the coverage in cell-edge users. A relay node is placed in between transmitter and receiver to reduce the path loss and improve the spectral efficiency of massive MIMO system. Cascaded channel estimation technique for massive MIMO relay system is developed in this paper. The mean squared error (MSE) of the cascaded channel estimation for massive MIMO relay sytem is optimized to obtain accurate CSI.

In this paper, we study the uplink achievable rate by a massive multiple-input multiple-output (MIMO) system in which the base station is equipped with a large number of 1-bit analog-to-digital converters (ADCs). We propose a linear minimum mean-squared error (MMSE)-based detector that accounts for the non-linearity effects of the 1-bit quantization as well as for channel estimation error. An analytical framework that derives the achievable rate of the MMSE-based detector in a massive MIMO configuration under the assumption of 1-bit quantized ADCs and channel estimation error is presented. We compare the achievable rates of a massive MIMO system using a 1-bit ADC and a linear detector against a conventional MIMO system with higher-order modulation and near maximum likelihood (ML) detection. We show that in the low signal-to-noise ratio (SNR) regime with channel estimation error, the quantized massive MIMO system can outperform the conventional large MIMO system; however for high SNR, the conventional MIMO system with a near ML detector can outperform the quantized massive MIMO system.

Massive multiple-input multiple-output (MIMO) is a profitable technique to greatly boost spectral efficiency, which has been embraced by the fifth-generation (5G) and sixth-generation (6G) mobile communication systems. By exploiting appropriate downlink precoding algorithms, base stations (BSs) equipped with a large number of antennas are able to provide service to multiple users as well as several cells at the same time and frequency. However, the mutual coupling effect due to the compact antenna array gives misleading results in massive MIMO communication systems. In this paper, we focus on the mutual coupling effect for massive MIMO systems with maximal ratio transmission (MRT), zero-forcing (ZF), regularize ZF (RZF), and minimum mean square error (MMSE) precoding to solve the mutual coupling problem. Additionally, we construct the closed-form expressions of the spectral efficiency (SE) to evaluate the effect of mutual coupling on system performance. Simulation results validate the effectiveness of the proposed mutual coupling effect assessment method and demonstrate the significant impacts of mutual coupling on massive MIMO system performance.

The computation power of baseband units (BBUs) is a major source of power consumption in millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) systems with a large number of users due to complex signal processing. The effective reduction of computation power is critical for improving system energy efficiency. In this paper, the principle of reducing the computation power of BBUs is first investigated in mmWave massive MIMO systems with a hybrid precoding structure. A recursive constraint in decomposing the baseband precoding matrix is derived for reducing the computation power of hybrid precoding systems. Furthermore, the optimal number of sub-matrices minimizing the maximum error in decomposing the baseband precoding matrix is obtained. Based on the proposed principle, consisting of the recursive constraint and the optimal number of sub-matrices, a fast Monte Carlo baseband precoding (FMCBP) algorithm is developed to reduce the computation power of BBUs and improve system energy efficiency. Simulation results show that the total transmission rate and energy efficiency of mmWave systems are coupled with the computation power of BBUs, based on the FMCBP algorithm. Moreover, the FMCBP algorithm maximally improves the energy efficiency of multi-user mmWave massive MIMO communication systems by 124 percent, compared with the conventional equivalent zero-forcing algorithm.

Unleashing the full potential of massive MIMO in FDD mode by reducing the\noverhead of CSI feedback has recently garnered attention. Numerous deep\nlearning for massive MIMO CSI feedback approaches have demonstrated their\nefficiency and potential. However, most existing methods improve accuracy at\nthe cost of computational complexity and the accuracy decreases significantly\nas the CSI compression rate increases. This paper presents a novel neural\nnetwork CLNet tailored for CSI feedback problem based on the intrinsic\nproperties of CSI. CLNet proposes a forge complex-valued input layer to process\nsignals and utilizes attention mechanism to enhance the performance of the\nnetwork. The experiment result shows that CLNet outperforms the\nstate-of-the-art method by average accuracy improvement of 5.41\\% in both\noutdoor and indoor scenarios with average 24.1\\% less computational overhead.\nCodes for deep learning-based CSI feedback CLNet are available at GitHub.\n

Downlink beamforming design for mobile users in massive MIMO system

Aiming at the problem of high complexity and low feedback accuracy of existing channel state information (CSI) feedback algorithms for frequency-division duplexing (FDD) massive multiple-input multiple-output (MIMO) systems, this paper proposes a CSI compression feedback algorithm based on deep learning (DL), which is suitable for single-user and multi-user scenarios in massive MIMO systems. This algorithm considers the spatial correlation of massive MIMO channel and uses bidirectional long short-term memory (Bi-LSTM) and bidirectional convolutional long short-term memory (Bi-ConvLSTM) network to decompress and recover the CSI for single-user and multi-user, respectively. The proposed DL-based CSI feedback network is trained offline by massive MIMO channel data and could learn the structural characteristics of the massive MIMO channel by fully exploiting the channel information in the training samples. The simulation results show that compared with several classical CSI compression feedback algorithms, the proposed CSI compression feedback algorithm has lower computational complexity, higher feedback accuracy, and better system performance in massive MIMO systems.

As a combination of spatial modulation (SM) system and massive multiple-input multiple-output (MIMO) system, massive SM aided MIMO (SM-MIMO) system is recently proposed. In broadband scenarios, single-carrier (SC) modulation is introduced to massive SM-MIMO system, thus massive SC-SM MIMO system is proposed for uplink multi-user transmission over frequency-selective fading channels. In this study, the uplink spectral efficiency (SE) of multi-cell massive SC-SM MIMO system is analysed, meanwhile the pilot contamination effect is taken into consideration. A tight SE lower bound of multi-cell massive SC-SM MIMO system is proposed with maximum ratio (MR) combining, which also takes into account the imperfect channel estimation, transmit antenna correlation and path loss. The tightness of the authors' proposed closed-form SE lower bound is shown via simulation results. The optimal pilot reuse factor can be determined with different system configurations by simulations, and the pilot reuse factor that is larger than one is more suitable for less TAs and user equipments.

Detection in multiple-input-multiple-output (MIMO) wireless communication systems is a crucial procedure in receivers since the multiple access transmission schemes generate interference due to the simultaneous transmission along with the several antennas, unlike single-input-single-output (SISO) transmission schemes. Precoding is a technique in MIMO systems used to mitigate the effects of the channel over the received signal. Hence, it is possible to adjust continuously the transmitted information to reverse the effect of the wireless channel at the receiver side. In this work, linear sub-optimal detectors and precoders for massive MIMO (M-MIMO) systems are implemented, analyzed, and compared in terms of performance-complexity trade-off. It is also being considered numerical results in both channel scenarios: a) receiver and transmitter have perfect channel state information (CSI); b) complex channel coefficients are estimated with different levels of inaccuracy. Monte-Carlo simulations (MCS) reveal that linear zero-forcing (ZF) and minimum mean squared error (MMSE) massive MIMO detectors result in a certain robustness against multi-user interference when operating under low and medium system loading, L = K/M, thanks to the favourable propagation phenomenon arising in massive MIMO systems.

Recently, the massive Multiple-Input Multiple-Output (MIMO) system has been integrated with machine learning approaches to realise automatic channel state information detection. These methods need high computational complexity due to the lack of optimal training sequences and the failure to control the sparsity assembly based on massive MIMO. Moreover, it does not consider optimised training sequences based on a compressive sensing approach in the presence of contamination. Thus, this paper proposed an enhanced deep learning model with an optimised pilot training sequence for sparse channel estimation in a 5 G massive MIMO system. Initially, the system and sparse block channel impulse response model will be constructed by considering the time-domain synchronous Generalized Frequency Division Multiplexing (GFDM) system and additive white Gaussian noise (AWGN). In the proposed work, a seagull optimisation algorithm is developed to design the training pilot sequence depending on the coherence properties of the sensing matrix, which is used to recover the channel impulse response. Then, the sparse massive MIMO channel is evaluated by proposing a new deep channel estimator with an enhanced stacked auto encoder (D-ESAE). The proposed channel MSE is 10−6, the bit error rate is 0.04, and NMSE is 10−9 at 40 dB SNR through Matlab simulation.