|Computers, Materials & Continua |
An Efficient Machine Learning Based Precoding Algorithm for Millimeter-Wave Massive MIMO
1Department of Electrical Engineering, University of Engineering and Technology Peshawar, Pakistan
2Department of Mechanical Engineering, College of Engineering, Taif University, Taif, 21944, Saudi Arabia
3School of Software, Hallym University, Chuncheon, 24252, Korea
*Corresponding Author: Wonjong Noh. Email: email@example.com
Received: 25 July 2021; Accepted: 16 November 2021
Abstract: Millimeter wave communication works in the 30–300 GHz frequency range, and can obtain a very high bandwidth, which greatly improves the transmission rate of the communication system and becomes one of the key technologies of fifth-generation (5G). The smaller wavelength of the millimeter wave makes it possible to assemble a large number of antennas in a small aperture. The resulting array gain can compensate for the path loss of the millimeter wave. Utilizing this feature, the millimeter wave massive multiple-input multiple-output (MIMO) system uses a large antenna array at the base station. It enables the transmission of multiple data streams, making the system have a higher data transmission rate. In the millimeter wave massive MIMO system, the precoding technology uses the state information of the channel to adjust the transmission strategy at the transmitting end, and the receiving end performs equalization, so that users can better obtain the antenna multiplexing gain and improve the system capacity. This paper proposes an efficient algorithm based on machine learning (ML) for effective system performance in mmwave massive MIMO systems. The main idea is to optimize the adaptive connection structure to maximize the received signal power of each user and correlate the RF chain and base station antenna. Simulation results show that, the proposed algorithm effectively improved the system performance in terms of spectral efficiency and complexity as compared with existing algorithms.
Keywords: MIMO; phased array; precoding scheme; machine learning optimization
In the past ten years, the rapid development of various business systems such as the Internet of Things (IoT) and the Internet of Vehicles (IoV), as well as the advancement of wireless equipment manufacturing processes, have promoted the development and deployment of 5G mobile communication systems with high-speed, large connections and low latency. In general, the improvement of spectrum efficiency is achieved through network densification of micro-cell millimeter wave and massive MIMO technology . The current low-frequency spectrum resources can no longer meet the people‚s needs for high-rate data transmission in people‚s lives. The development of wireless communication technology has forced researchers to include the millimeter wave frequency band into the scope of research. The millimeter wave frequency is distributed in the range of 30 to 300 GHz, and its ample bandwidth has become a hot spot in the academic and industrial circles. However, millimeter waves are severely absorbed by the atmosphere and rain in the process of space transmission, and the limited number of propagation paths, resulting in their short effective propagation distance, which is very suitable for micro-cell communication with small coverage and high data transmission rate. It provides sufficient array gain to form a needle beam to reduce interference, which is regarded as one of the key technologies of 5G . When a large number of antennas are deployed at the millimeter wave transmitting end, all-digital precoding needs to be equipped with a dedicated radio frequency chain for each transmitting antenna, and a radio frequency chain with the number of antennas (composed of digital-to-analog converters, mixers, etc.) which has high cost and unacceptable energy consumption. Therefore, the research and design of low-dimensional baseband digital precoders and high-dimensional radio frequency analog precoders instead of all-digital precoders has aroused great interest. In a single-user communication system, there are ways to improve the spectrum efficiency by minimizing the Euclidean distance between hyrbid analog and digital precoding and fully-digital precoding [3‐6], and also through a joint transmitter analog precoder and receiver analog research on the maximized spectral efficiency of the combiner [7–9]. In multi-user communication systems, the hybrid precoding with a fully connected structure has also been studied [10–14]. For example, reference  directly uses the phase of the channel conjugate transpose as an analog precoder, and then uses the zero forcing (ZF) technique to design the baseband digital precoder. Reference  proposed a two-stage hybrid precoding and designed analog precoding in which the phases of all digital precoding obtained by maximum ratio transmission (MRT) and ZF precoding were extracted respectively, and then combined with channel moments. The equivalent matrix obtained by the matrix eliminates inter-user interference through ZF to obtain the baseband digital precoding. However, these hybrid precoding based on the fully connected structure require the use of more RF chains and high-precision phase devices, resulting in expensive hardware costs and power consumption proportional to accuracy, hindering the hybrid precoding structure in the base station and mobile end millimeter wave cellular network systems with strict size and power restrictions in the deployment.
In order to reduce the number and precision of the hardware used, further research on hybrid precoding of partial connection structures has been carried out [17,18]. Reference  first proposed the hybrid precoding of switch and inverter based on machine learning adaptive cross entropy. The authors in [20–22] applied machine learning adaptive cross entropy to the hybrid precoding of lens array switch structure, and further analyzed the impact of important parameters based on the sum rate and energy efficiency of the hybrid precoding of the switch and inverter structure. Reference  proposes a precoding with an adaptive connection structure, which can better achieve beam gain, but still requires a higher-precision phase shifter (at least 6-bit accuracy) to achieve close to the optimal fully-digital precoding and lower precision such as 1-bit quantized adaptive connection structure is achievable and rate performance is severely reduced. In order to solve the problem of poor accessibility a rate performance of the 1-bit quantized phase shifter of the adaptive connection structure, this paper proposes a 1-bit quantized phase shift based on machine learning adaptive cross-entropy hybrid precoding. The adaptive connection structure that obtains the matching relationship between the RF chain and the base station antenna by maximizing the user's received signal power is more flexible than the fixed sub-connection. According to the probability distribution, the analog precoder is randomly generated, and the classic ZF precoding is used to obtain the corresponding digital precoder. The analog precoder is adaptively weighted according to the reach and rate. Then, the probability distribution of the simulated precoding is updated by reducing the cross entropy and adding a constant smoothing parameter, and repeating this way, a hybrid precoder with almost optimal performance and rate is finally obtained. Numerical simulations are performed to evaluate the effectiveness of the proposed scheme.
2 System Model
As shown in Fig. 1, consider the massive MIMO system of multi-user downlink, the base station deploys N antennas, radio frequency chains, and simultaneously serves K non-cooperative users with a single antenna. Generally, the massive MIMO system satisfies , and is set in this article. The analog precoder is composed of a small number of radio frequency (RF) chains, an adaptive connection network and a large number of antennas . It is assumed that the symbols sent to K users are independently and identically distributed, and they are all symbols that obey the zero-mean complex Gaussian distribution, satisfying where P is the total transmitted signal power of the base station .
The signal received by K users can be expressed as
Among them, represents the downlink channel from the base station to all users, represents the channel gain from the base station to the k th user. represents a high-dimensional constant modulus analog precoder that can only adjust the signal phase, stands for a low-dimensional digital precoder that can adjust the signal amplitude and phase. represents dimension vector additive white Gaussian noise AWGN, satisfying , represents the power of additive white Gaussian noise . In order to ensure the total transmit power constraint, after the analog precoding matrix is designed, the corresponding digital precoding matrix should meet the power constraint . For the channel gain vector, in order to better describe the characteristics of limited space selection and limited scattering caused by the extremely high propagation path loss of the millimeter wave channel, and the large number of antenna arrays of the millimeter wave transmitter leads to a large degree of antenna correlation, using the geometric Saleh-Valenzuela channel model, assuming that the transmitting end is a uniform linear array (ULA) antenna, the channel gain of the -th user can be expressed as:
Among them, N is the number of antennas of the base station ULA antenna. represents the number of propagation paths of the -th user, and respectively represent the complex gain and departure azimuth angle of the -th path of the -th user, and represents the response vector of the transmission array with a dimension of , which can be expressed for
Here, is the wavelength of the millimeter wave, and d is the distance between the ULA antenna elements, usually set to half the wavelength [27--28].
3 Proposed Algorithm
The analog precoding of the traditional low-precision phase shifter partial connection structure often fails to achieve the array gain of the millimeter wave large-dimensional antenna. Therefore, this article simulates the precoder adaptive connection and deploy adaptive connection network instead of fixed sub-connection switch and reverse vectorizer (equivalent to a phase shifter with 1-bit quantization). The same as the fixed sub-connection structure, the adaptive connection only requires N number of 1-bit quantized phase shifters, number of radio frequency chains, and no adders. Compared with fully connected high-precision phase shifters, radio frequency chains, and N adders, it greatly reduces the hardware complexity, cost and energy consumption. The adaptive connection network can better match the downlink to improve the user's reach and rate. In order to make better use of adaptive cross-entropy optimization to apply to the adaptive connection structure, it is necessary to match the RF chain and the antenna under specific channel state information, that is, to find the position of the non-zero element in the corresponding analog precoding matrix .
Due to the special structure of the adaptive connection and the normal mode constraint of the elements in the analog precoder, the corresponding analog precoding is a sparse matrix, as shown in Eq. (4), represents the element in the j th row and k th column, and the k th radio frequency chain is connected to the j th base station antenna, and its element . Sum each row of to get vector , and its j th element . Therefore, the adaptive connection structure simulation precoding has two constraints:
Indicates the connection relationship between all radio frequency chains and all antennas of the base station. Assuming that is an integer, divide the N base station antennas into K independent sets, use to represent the set of -th radio frequency chain connected to base station antennas, and to denote the -th radio frequency chain connected to j th base station antenna, and . So Eq. (5) is also equivalent to
The designed analog precoding and digital precoding should maximize the downlink reachability and rate of the K users served
where represents the signal-to-interference and noise ratio of the -th user, which is expressed as
It can be seen that Eq. (7) is a non-convex optimization problem under the constraints of power constraints and adaptive connection structure. To maximize the total reachability and rate, that is, to maximize the SINR of the symbols received by each user, the -th column of the must satisfy
There are only M non-zero elements in , and the first M maximum values of the channel element modulus need to be matched, one element is matched each time and the antenna position is returned, and stored in the set
The constraint means that the j th line of the has not been evaluated or the j th antenna has not been matched.
Performing Eq. (10) once can match the -th radio frequency chain with a base station antenna, that is, get a non-zero element position in the . To ensure the fairness of the radio frequency chain, K radio frequency chains are matched in turn, and M turns are performed. The position of the N non-zero elements of the that meets the constraint (5) can be obtained, or the set of matching relationships between the RF chain and the antenna that meets the constraint (6) .
In the ACN-MLACE algorithm, since the phase shifter is quantized by 1 bit, the non-convex optimization problem of Eq. (7) has been transformed into a combinatorial optimization problem. How to obtain the and with the largest reach and rate? It is necessary to search exhaustively to find N determinations. The combinations of non-zero element values in the position involve extremely high computational complexity. For millimeter-wave massive MIMO systems, N is often large, such as N = 56, and an exhaustive search is required for . Take another path, deploy ML adaptive algorithm based on cross-entropy optimization to intelligently look for the best adaptive connection precoding. In each iteration, dimensional vectors are generated according to a probability distribution. These vectors can be set according to the matching relationship between each radio frequency chain and the base station antenna to obtain Z candidates analog precoding matrices , each candidate analog precoding matrix has a corresponding reach and rate. Select the with the best performance and rate, and update the probability distribution by minimizing cross entropy and adding smoothing parameters. In this way, it will be generated with probability 1 and close to the optimal candidate simulates the probability distribution of non-zero elements in the precoding matrix. The vector composed of non-zero elements in each row of the analog precoder is reconstructed according to the found matching relationship, and is used to represent the probability of the corresponding element in , and the -th element is Bernoulli random variables, that is, the probability of is , and the probability of is .
By initializing the probability distribution parameter to , ( is a vector of all 1 s), according to the probability distribution , generate Z candidate vectors samples, the matching relationship is reconstructed to generate Z candidate analog precoding samples, and then according to the equivalent channel , , the corresponding Z digital precoding . In this article, the classic digital ZF precoding is used to eliminate the inter-user interference to obtain the corresponding -th digital precoding matrix
Put power constraints on it: .
After that, the achievable sum rate is obtained by substituting and into Eqs. (7) and (8), and sort Z and in descending order. In order to adaptively update the next probability , it is necessary to obtain the first and with the best achievable rate, and define the best analog precoding achievable rate average value is T which is expressed as
Here, the -th analog precoding matrix is the first vector obtained according to the matching relationship between the radio frequency chain and the base station antenna, so . The weight of the achievable rate corresponding to the -th analog precoding matrix is . Update the probability of the next time adaptively according to the current probability distribution and weight
There is also , where
Substituting Eq. (14) into Eq. (13), and then Eq. (13) finds the first derivative of its j-th probability element to obtain
Setting Eq. (15) equal to zero, the j-th element of the next probability distribution can be obtained
In order to ensure that the adaptive cross entropy optimization converges to the optimal solution to avoid local convergence, a constant smoothing parameter Ӫ can be further added between the current probability distribution and the next probability distribution.
Here , until the end of the l iteration is reached, the probability distribution for generating the optimal analog precoding will be obtained, and the optimal analog precoding and the optimal number in the l generation sample will be selected precoding , which is the almost optimal adaptive connection hybrid precoding under the corresponding channel state information. The specific algorithm is as follows in Algorithm 1.
4 Simulation Results
This section provides the simulation results and analysis. The proposed machine learning based precoding algorithm is compared with fully digital precoding, hybrid precoding of adaptive connection structure, and the conventional OMP precoding of structure. The combined precoding has the same lower hardware complexity and eliminates the phase shifters and N adders required by the fully connected hybrid structure. Therefore, the sum rate and complexity are used here as a comparison of the performance of different precoding schemes. The simulation parameters are set as follows in Tab. 1.
4.1 Achievable Sum Rate Comparison with Different Number of RF Chains and Data Streams with Fixed Antennas
Fig. 2 compares the achievable sum rate of the proposed algorithm, fully digital, and other algorithms vs. SNR for system configuration when the number of RF chains and number of data streams . As can be seen from Fig. 2 that, the achievable sum rates of all algorithms increase with increasing SNR. Moreover, the proposed algorithm gives close performance with optimal fully digital scheme which indicates its effectiveness over the existing algorithms.
Fig. 3 illustrates the achievable sum rate of the proposed algorithm, fully digital, and other algorithms vs. SNR for system configuration when the number of RF chains and number of data streams . As can be seen from Fig. 3, the achievable sum rates of all algorithms increases with increasing SNR. Moreover, the proposed algorithm gives close performance with optimal fully-digital scheme which indicates its effectiveness over the existing algorithms. Here, the results are closed for all algorithms because the number of RF chains are increased. But the energy consumption drastically increases in the existing algorithms in contrast, which makes them unsuitable for deployment. Also, increasing the number of RF chains increases the computational complexity and hardware structure.
4.2 Achievable Sum Rate Comparison with Different Number of Antennas with Fixed Number of RF Chains and Data Streams
Fig. 4 compares the achievable sum rate of the algorithms versus SNR when the number of transmitter antennas , the number of receiver antennas and number of RF chains and data streams is . As can be seen from Fig. 4, the achievable rate of all algorithms increases with SNR. Moreover, the proposed algorithm gives better performance and shows close sum rate with optimal fully digital precoding. It is also clear from Fig. 4 that, due to increasing number of antennas, the sum rate is about 72 bps/Hz for SNR = 25 dB, whereas the sum rate is 63 bps/Hz for SNR = 25 dB in Figs. 3 and 4, respectively. This proves that the sum rate increases with increasing the number of antennas, which is one of the main features of massive MIMO. Fig. 5 compares the achievable sum rate of the algorithms versus SNR when the number of transmitter antennas , the number of receiver antennas and number of RF chains and data streams is . As can be seen from Fig. 5, the achievable rate of all algorithms increases with SNR. Moreover, the proposed algorithm gives better performance and shows close sum rate with optimal fully digital precoding. It is also clear from Fig. 5 that, due to increasing number of antennas, the sum rate is about 87 bps/Hz for SNR = 25 dB, whereas the sum rate is 63 bps/Hz for SNR = 25 dB in Figs. 3 and 4, respectively. This further proves that the sum rate increases with increasing the number of antennas, which is one of the main features of massive MIMO.
4.3 Complexity Analysis
Fig. 6 compares complexity of the algorithms with increasing number of antennas at the transmitters and , and . As can be seen from Fig. 6, the complexities of all algorithms increase with increasing the number of antennas at the BS. Moreover, the complexity of the proposed algorithm is lower than the complexities of existing algorithms and also closed to the optimal fully digital precoding. This means that the proposed algorithm requires less number of iterations to achieve the same performance as compared with existing algorithms.
This paper proposes an adaptive connection network hybrid precoding with 1-bit quantization, and applies the adaptive algorithm based on machine learning to the adaptive connection structure hybrid precoding, which improves the 1-bit quantization phase shift of the adaptive connection structure. Under the same low hardware complexity, the proposed solution has a higher computational complexity than the switch and inverter hybrid precoding based on the fixed sub-connection of machine learning and the hybrid precoding based on the adaptive connection structure and achievable rate performance. Recently, highly efficient deep learning methods have been applied to hybrid precoding, and precoding with lower computational complexity and better spectral efficiency is worthy of further research.
Acknowledgement: Taif University Researchers Supporting Project Number (TURSP-2020/260), Taif University, Taif, Saudi Arabia.
Funding Statement: The authors received no specific funding for this study.
Conflicts of Interest: The authors declare that they have no conflicts of interest to report regarding the present study.
|This work is licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.|