A DNN-based 5G MIMO System Adopting a Mix of Tactics

A DNN-based 5G MIMO System Adopting a Mix of Tactics | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article A DNN-based 5G MIMO System Adopting a Mix of Tactics Md. Ashraful Islam, Md. Firoz Ahmed, Md. Matiqul Islam This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5302008/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 12 You are reading this latest preprint version Abstract The emergence of 5G technology necessitates advanced communication systems that can handle high data rates while providing improved reliability. However, current methods often struggle with issues related to complexity and efficiency. This paper addresses these challenges by developing a Multiple Input Multiple Output (MIMO) system that integrates deep neural network (DNN) demappers, Low-Density Parity-Check (LDPC) coding, and polar coding, creating a mix of tactics aimed at boosting performance in flat fading channel environments. The main objectives of this study are to improve the bit error rate (BER) and spectral efficiency across different orders of Quadrature Amplitude Modulation (QAM)—specifically, 4-QAM, 16-QAM, and 256-QAM. The proposed system employs a Linear Minimum Mean Square Error (LMMSE) equalizer to ensure effective signal detection. Simulation results indicate that the LDPC coding technique significantly outperforms polar coding across all evaluated QAM modulation orders, highlighting its effectiveness in enhancing system performance. An analysis of BER and spectral efficiency demonstrates considerable reliability and data throughput improvements, making the MIMO system particularly well-suited for the demanding requirements of 5G applications. The findings suggest that integrating advanced coding strategies with neural demapping can effectively tackle the complexities introduced by the 5G environment. This research establishes a foundation for future studies and the development of robust communication systems capable of supporting the next generation of wireless technology, thereby enabling a wide range of applications, from enhanced mobile broadband to extensive IoT connectivity. LDPC code Polar code DNN demapper 5G MIMO BER Spectral efficiency Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 1. Introduction The rapid evolution of 5G technology has introduced a transformative phase in wireless communication, necessitating innovative solutions to meet the increasing demands for higher data rates, enhanced reliability, and improved spectral efficiency [ 1 ]. With applications spanning enhanced mobile broadband (eMBB), massive machine-type communications (mMTC), and ultra-reliable low-latency communications (URLLC), 5G systems are challenged to manage complex performance requirements while promoting user connectivity [ 2 ]. A critical component in addressing these challenges is the implementation of Multiple Input Multiple Output (MIMO) systems, which utilize multiple antenna elements to enhance data throughput and link reliability substantially [ 3 ]. MIMO technology operates on spatial diversity and multiplexing, allowing a single wireless channel to transmit multiple data streams simultaneously. This capability is vital in high-demand environments, where conventional methods may struggle to maintain efficiency and performance due to increased interference and fading effects. Despite the advantages of MIMO systems, traditional techniques face limitations in terms of computational complexity and adaptability to varying channel conditions, particularly in high-order modulation scenarios [ 4 ]. Recent studies have proposed integrating advanced signal processing methods to overcome these limitations to improve performance. DNN demappers represent a significant advancement in the decoding process for MIMO systems. By leveraging deep learning algorithms, DNN demappers can learn complex mappings from received signals to transmitted symbols, effectively improving BER performance [ 5 ]. This approach has been shown to outperform traditional demapping techniques, especially in high-order Quadrature Amplitude Modulation (QAM) scenarios. Moreover, coding techniques such as LDPC and polar coding have gained prominence for their ability to approach channel capacity. LDPC codes provide excellent error correction capabilities, vital for maintaining performance in noisy environments [ 6 ]. In contrast, polar codes, introduced by [ 7 ], have emerged as a powerful alternative with their capacity-achieving properties, particularly in high-throughput applications. Both coding schemes can be effectively combined with MIMO systems to enhance spectral efficiency and reliability. This paper offers a thorough analysis of a 5G Multiple Input Multiple Output (MIMO) system that integrates deep neural network (DNN) demappers, Low-Density Parity-Check (LDPC) coding, and polar coding techniques, forming a mixed strategy (MS) to enhance overall performance while overcoming the limitations of traditional MIMO methods. This study aims to assess the effectiveness of this MS in augmenting bit error rate (BER) and spectral efficiency across various orders of Quadrature Amplitude Modulation (QAM). The findings demonstrate substantial improvements in both BER and spectral efficiency, underscoring the potential of this MS approach for future wireless communication systems. 2. Literature Review This section focuses on 5G MIMO systems and highlights the integration of cutting-edge signal processing techniques, particularly deep neural networks (DNNs), Low-Density Parity-Check (LDPC) coding, and polar coding. These advancements are designed to enhance performance metrics, including bit error rate (BER) and spectral efficiency, across various orders of Quadrature Amplitude Modulation (QAM). DNNs have been adopted for decoding processes in Massive-MIMO systems, leading to significant improvements in both decoding performance and spectral efficiency [ 8 ]. A comprehensive DNN-based decoder for polar-LDPC codes achieves performance levels comparable to traditional decoding methods while optimizing resource utilization [ 9 ]. Furthermore, LDPC codes implemented in large-scale MIMO configurations demonstrate a decrease in performance gaps as the number of antennas increases [ 10 ]. The use of CNNs for channel estimation in polar-coded MIMO-OFDM systems results in notable reductions in BER and mean square error, thereby enhancing the overall reliability of the system [ 11 ]. Recent studies have introduced innovative techniques such as a multi-input deep learning-based joint pilot decontamination and symbol detection mechanism tailored for 5G massive multiple-input multiple-output (MAMIMO) systems [ 12 ]. Additionally, a method utilizing matrix inversion for data detection in Massive MIMO employing DNNs has been proposed [ 13 ]. A deep learning-based algorithm for multi-edge type decoding of 5G NR LDPC codes has also been presented [ 14 ]. The decoding of short 5G LDPC codes via DNNs has shown promising results [ 15 ]. Furthermore, a CNN-based strategy aimed at enhancing the decoding performance of 5G LDPC codes has been suggested [ 16 ]. Studies have also explored the BER performance of joint polar-coded signals in MIMO systems utilizing millimeter-wave technology [ 17 ], along with the performance evaluation of 5G New Radio LDPC codes across various log-normal fading channel scenarios [ 18 ]. The comparison of BER performance between Low-Density Parity-Check codes and polar codes has been conducted [ 19 ], as well as an analysis of error correction capabilities of LDPC and polar codes in the context of 5G Machine Type Communications [ 20 ]. Proposals for 5G systems utilizing Low-Density Parity Check-based channel coding for enhanced mobile broadband applications have been made [ 21 ], alongside evaluations of LDPC and polar coding schemes within the framework of 5G technology, particularly for Massive Machine Type Communication [ 22 ]. Lastly, the efficacy of LDPC decoding algorithms in 5G channel modeling for MIMO-OFDM systems under spatial correlation influences has been investigated [ 23 ], and the performance of 5G LDPC and polar decoding in spatially correlated MIMO-OFDMA systems has been analyzed [ 23 ]. Insights into the performance of polar codes for next-generation 5G technology have also been discussed [ 24 ]. 3. Methods 3.1 Mix of Tactics A "mix of tactics" refers to the strategic blend of various advanced techniques aimed at optimizing system performance. In this scenario, this combination includes the integration of deep neural network (DNN) demappers, Low-Density Parity-Check (LDPC) coding, and polar coding. Each of these techniques offers distinct advantages: DNN demappers leverage machine learning algorithms to significantly improve signal detection accuracy, particularly in challenging and noisy environments [ 25 ]. LDPC coding serves as a robust error-correcting technique, enhancing reliability by efficiently correcting transmission errors. Polar coding is another effective error-correcting method, recognized for its strong performance even at low signal-to-noise ratios [ 26 ]. By uniting these approaches, the system gains improved error correction, enhanced signal detection, and overall communication reliability, resulting in a more resilient and efficient solution for demanding applications like 5G technology[ 27 ]. 3.2 Deep Neural Network (DNN) Demappers Deep neural network (DNN) demappers represent a revolutionary advancement in signal detection for 5G wireless systems, particularly for sophisticated modulation schemes. Traditional demapping techniques often encounter significant challenges in environments characterized by noise and interference. In contrast, DNN demappers leverage the power of artificial intelligence (AI) and machine learning (ML) to enhance bit error rate (BER) performance and overall system efficiency [ 28 ]. At the core of DNNs is a structure composed of multiple layers of neurons, which are capable of learning complex mappings between input signal samples and the corresponding output symbols. This capability makes DNNs exceptionally effective in demapping tasks that typically hinder conventional methods [ 29 ]. Their proficiency in handling high-dimensional data is particularly advantageous for next-generation modulation schemes, such as 64-QAM and 256-QAM, allowing them to adapt seamlessly to varying channel conditions [ 30 ]. Research has demonstrated that DNN demappers significantly outperform traditional methods, achieving substantial reductions in BER even in high-noise environments [ 31 ]. Additionally, they contribute to improved spectral efficiency, thereby supporting advanced modulation formats that enhance system throughput [ 32 ]. Nonetheless, challenges remain, including issues related to computational complexity, the risk of overfitting, and the interpretability of the models [ 33 ]. Ongoing advancements in hardware and research into explainable AI are actively addressing these concerns. In conclusion, DNN demappers are set to play a pivotal role in advancing BER, spectral efficiency, and overall performance in future 5G communication systems, marking a significant leap forward in the evolution of wireless technology. 3.3 Low-Density Parity-Check (LDPC) coding Low-Density Parity-Check (LDPC) coding represents a sophisticated approach to error correction utilized in modern communication systems, aiming to enhance data reliability by accurately detecting and correcting errors during transmission [ 6 ]. Characterized by a sparse parity-check matrix—where the majority of the elements are zeros—LDPC codes minimize computational complexity while providing robust error correction capabilities [ 34 ]. The fundamental principle behind LDPC coding involves the incorporation of redundant information into transmitted data. This redundancy enables the receiver to identify and rectify errors, even in the presence of significant noise or interference [ 35 ]. A common method for decoding LDPC codes is the use of iterative algorithms, notably the belief propagation algorithm, which systematically checks for and resolves inconsistencies between the transmitted and received bits [ 36 ]. LDPC codes are notably efficient, approaching the theoretical limits of channel capacity as outlined by Shannon [ 37 ]. This characteristic makes them particularly suitable for high-demand applications, including 5G technology, Wi-Fi networks, and satellite communications. By achieving a favorable balance between complexity and performance, LDPC codes ensure low bit error rates (BER) while maximizing data throughput. The key advantages of LDPC codes include: High Error Correction Capability : They perform exceptionally well in noisy channel environments. Scalability : LDPC codes can be applied across a range of block lengths and code rates, making them versatile. Efficiency : They are designed to approach optimal performance limits, ensuring reliable communication even under challenging conditions. Due to their robustness and efficiency, LDPC coding has become a staple in contemporary communication systems, playing a vital role in enhancing data transmission capabilities in advanced networks like 5G [ 38 ]. 3.4 Polar Coding Polar coding is a groundbreaking error correction technique that enhances reliable data transmission by polarizing communication channels into distinct groups of highly reliable and unreliable subchannels. Introduced by Erdal Arıkan in 2009 [ 39 ], polar coding is notable for being the first coding scheme to rigorously demonstrate its capability to achieve Shannon capacity—the theoretical maximum data rate for error-free transmission—through a straightforward encoding and decoding process [ 39 ]. The core principle of polar coding is channel polarization, where a communication channel is decomposed into virtual subchannels. As the code length increases, certain subchannels become highly reliable while others turn unreliable. The strategic approach involves transmitting vital information bits through these reliable subchannels, while assigning fixed or random (frozen) bits to the unreliable ones, thus optimizing the error correction process. Polar codes utilize a recursive structure based on XOR operations, which transforms the original message into a longer codeword that is spread across the polarized subchannels. The decoding process typically employs the Successive Cancellation (SC) algorithm, as well as its enhanced variant, the Successive Cancellation List (SCL) decoding. SCL decoding further mitigates errors by assessing multiple decoding paths and selecting the most optimal one [ 40 ]. Notably, polar codes are capacity-achieving, rendering them highly effective for noisy communication channels. Their computational efficiency in both encoding and decoding, coupled with their adaptability across various block lengths and code rates, makes them ideal for real-time applications. In the context of 5G networks, polar codes are particularly advantageous for control channels, given their strong performance with short block lengths and low error rates. They are often employed alongside other coding techniques, such as Low-Density Parity-Check (LDPC) codes, for data channels within 5G systems [ 41 ]. In conclusion, polar coding signifies a remarkable advancement in the field of channel coding, harmonizing theoretical sophistication with practical applicability for next-generation wireless communications. 3.5 Linear Minimum Mean Square Error (LMMSE) equalizer The Linear Minimum Mean Square Error (LMMSE) equalizer is an advanced signal processing method widely utilized in contemporary communication systems to mitigate the adverse effects of noise and interference during the process of signal detection. The primary goal of the LMMSE equalizer is to estimate the transmitted signal while minimizing the mean square error (MSE) between the estimated signal and the actual transmitted signal, all while taking into consideration both channel distortions and noise [ 42 ]. Functioning as a linear filter, the LMMSE equalizer transforms the received signal into a more accurate estimate of the original transmitted signal. This attribute makes the LMMSE equalizer computationally efficient, which is crucial for real-time applications in dynamic communication environments [ 43 ]. By concentrating on minimizing the MSE—defined as the average squared difference between the estimated and actual transmitted signals—the LMMSE equalizer effectively reduces noise and interference, thereby helping to maintain the integrity of the signal. This technique has shown exceptional performance in channels characterized by additive noise, such as Gaussian noise, and in scenarios experiencing distortion or fading, both of which are prevalent in wireless communication systems [ 44 ]. Moreover, the LMMSE equalizer is most effective when the statistical properties of the channel and noise are known or can be accurately estimated, allowing for optimized filtering and signal recovery [ 45 ]. In a communication system, the received signal, denoted as y, can be represented as the sum of the transmitted signal ( x ) affected by the communication channel ( H ) and noise ( n ): y = Hx + n ............(1) Here, H encapsulates the effects of fading and multipath experienced during transmission, while n represents the additive noise, often modeled as Gaussian noise. The goal of the Linear Minimum Mean Square Error (LMMSE) equalizer is to effectively estimate the original transmitted signal ( x ) from the received signal ( y ) by applying a linear filter ( W ): x̂ = Wy ..................(2) Where, x̂ represents the estimated transmitted signal. The LMMSE equalizer aims to minimize the Mean Square Error (MSE) between the estimated signal x̂ and the actual transmitted signal x . This optimal filter W is determined by leveraging the channel matrix ( H ) and the noise covariance matrix ( R n ): W = (H H H + R n ) −1 H H ................. (3) Here, H H denotes the conjugate transpose of H . This formulation allows the LMMSE equalizer to effectively mitigate the effects of noise and interference, thereby enhancing the accuracy of signal recovery in communication systems [ 46 – 48 ]. The LMMSE equalizer proves particularly valuable in scenarios characterized by high noise and interference, such as wireless communication channels. By minimizing the bit error rate (BER) and enhancing overall signal quality, it plays a crucial role in improving communication reliability. This makes it particularly relevant in modern communication systems like Multiple Input Multiple Output (MIMO) and Orthogonal Frequency Division Multiplexing (OFDM) systems, prevalent in 4G and 5G networks. Despite its effectiveness, the LMMSE equalizer necessitates knowledge of the channel and noise characteristics, often requiring estimation in real-world settings, potentially introducing errors. Moreover, the matrix inversion involved in its computation can be computationally demanding, particularly in MIMO systems with numerous antennas. 4. Flat Fading Channel A flat fading channel is a specific type of wireless communication channel characterized by a coherence bandwidth that exceeds the bandwidth of the transmitted signal. In this context, all frequency components of the signal experience identical fading characteristics, resulting in uniform attenuation or amplification across the entire bandwidth. The amplitude of the transmitted signal can fluctuate over time due to variations in the wireless environment, which may include the movement of the transmitter, receiver, or surrounding objects. Since the signal bandwidth is smaller than the channel's coherence bandwidth, there is minimal frequency-dependent fading. This means that the channel impacts all segments of the transmitted signal uniformly, in contrast to frequency-selective fading, where different portions of the signal spectrum experience varying levels of fading. Flat fading typically arises from a single dominant propagation path, such as in line-of-sight communication, or when multipath components are closely spaced in time and exert similar effects on the signal [ 49 ]. In a flat fading channel, the received signal y ( t ) = h ( t ) ⋅ x ( t ) + n ( t )..................(4) Where, x ( t ) is the transmitted signal, h ( t ) is the time-varying fading coefficient (a complex scalar) that represents the attenuation and phase shift introduced by the channel, n ( t ) is the additive noise. The fading coefficient h ( t ) typically follows a Rayleigh or Rician distribution, depending on the presence of a line-of-sight path. In scenarios where no dominant line-of-sight component exists, the fading tends to be Rayleigh-distributed; conversely, if a line-of-sight path is present, the fading follows a Rician distribution [ 50 ]. Flat fading influences both the amplitude and phase of the signal uniformly across its entire bandwidth, which can lead to signal degradation, particularly in high-mobility environments that induce rapid fading. This may result in deep fades, where the signal power experiences significant momentary drops, leading to errors in signal detection and increased bit error rates (BER). To counteract these effects, various techniques—such as diversity, channel coding, and equalization—are employed. Flat fading channels are commonly observed in narrowband communication systems, such as certain cellular or satellite communications, where the transmitted signal bandwidth is small relative to the channel's coherence bandwidth [ 51 ]. In contrast, for wideband systems, like those utilized in broadband wireless communications (e.g., 4G and 5G), frequency-selective fading becomes increasingly significant. 5. Proposed System The block diagram illustrates in Fig. 1 a 5G channel-coded 4×16 MIMO neural demapper-based communication system, showcasing a modern wireless communication design with error correction, modulation, and machine learning enhancements. Binary Source This generates the input data in binary form. 5G Encoder (LDPC and Polar Coding) The binary data is encoded using LDPC (Low-Density Parity-Check) and Polar Codes, both commonly used in 5G for error correction. QAM Mapper The encoded bits are modulated using Quadrature Amplitude Modulation (QAM) to form symbols for transmission. Flat Fading Channel The signal passes through a flat fading channel, which simulates the wireless channel's effects like fading and noise. LMMSE Equalizer A Linear Minimum Mean Square Error (LMMSE) equalizer is used to counteract the distortions caused by the fading channel, improving signal recovery. Neural Demapper A neural network with dense layers demaps the received signal back into bits. This step replaces traditional demapping methods with machine learning for potentially better accuracy. 5G Decoder (LDPC and Polar Coding) The demapped bits are decoded using LDPC and Polar Codes to retrieve the original data. Compute BER The Bit Error Rate (BER) is calculated to evaluate the system's performance by comparing the original and decoded bits. This system integrates MIMO (Multiple Input, Multiple Output) technology with machine learning-based demapping to enhance performance in 5G communication environments. The Table 1 below displays the deep neural network (DNN)-based demapper's training settings. Table 1 Neural demapper training specifications Training Parameters Value Batch size 64 Epochs 30000 Optimizer SGD Loss function Binary cross entropy Modulation M-ary (M = 4,16,256) QAM Channel Flat Fading 6. Simulated Results This section presents a comparative analysis of the performance between a novel LDPC-coded 5G MIMO deep neural network (DNN) demapper communication system and a polar-coded 5G MIMO neural demapper system, evaluated across various QAM modulation orders. The primary focus is on assessing both systems in terms of their bit error rate (BER) while also considering the influence of spectral efficiency resulting from different coding techniques. 6.1 BER analysis The Bit Error Rate (BER) is a crucial metric in the realm of wireless communications, functioning as an essential indicator of the integrity of data transmission. It quantifies the frequency of errors occurring during data transmission, providing insights into the quality of the communication link. An elevated BER signifies a decline in link quality, resulting in increased data corruption, a higher rate of retransmissions, and a reduction in overall system efficiency. In contrast, a low BER is a testament to superior link quality, characterized by fewer errors, enhanced data integrity, and improved system performance. Several factors can influence the BER, including the signal-to-noise ratio (SNR), the modulation techniques utilized, error correction coding methods, and the prevailing channel conditions, such as fading and interference. It is imperative to effectively manage the BER to achieve an optimal balance between data rate and reliability in wireless communication systems. In our simulation study, we explored performance in a flat fading channel with SNR values ranging from − 5 dB to 15 dB. The analysis focused on a proposed 5G MIMO communication system that employs advanced techniques such as polar coding and Low-Density Parity-Check (LDPC) coding, alongside various Quadrature Amplitude Modulation (QAM) techniques (4-QAM, 16-QAM, and 256-QAM). The performance evaluation was conducted on a MIMO system equipped with 4×16 transmitting and receiving antennas. The simulation results are visually represented in Figs. 2 to 4, which showcase the spectra for both polar-coded and LDPC-coded 5G MIMO communication systems. In these figures, the solid blue curve symbolizes the performance of the polar-coded system, while the solid red curve illustrates the LDPC-coded results, both utilizing the 4×16 antenna configuration. In Fig. 2, we present the simulation outcomes for the 5G MIMO communication system utilizing 4-QAM modulation. Notably, at an SNR of -1 dB, the LDPC code achieves a Bit Error Rate (BER) of approximately 0.041382, in stark contrast to the polar code's BER of 0.094177. This finding underscores the superiority of the LDPC code, which provides a substantial improvement of 7.23 dB over the polar code. Overall, the simulation results highlight the enhanced performance of the LDPC-coded 5G MIMO neural demapper-based communication system. Furthermore, Figs. 3 and 4 further validate that the LDPC-coded MIMO communication system consistently outperforms its polar-coded counterpart, showcasing significant performance enhancements. These findings reaffirm the importance of selecting appropriate coding strategies and modulation techniques in advancing the reliability and efficiency of wireless communication systems. 6.2 Spectral efficiency Spectral efficiency is a critical metric that quantifies the effectiveness of a wireless communication system in utilizing its designated bandwidth. Typically measured in bits per second per Hertz (bps/Hz), spectral efficiency reflects the volume of data that can be transmitted over a specific frequency spectrum. Its significance cannot be overstated, as it has a direct correlation with the capacity and performance of wireless networks. In an era marked by an ever-increasing demand for wireless services, high spectral efficiency enables a greater number of users to connect and enjoy higher data rates within the same frequency band. This efficiency is vital for effective resource allocation, minimizing latency, and enhancing overall network reliability. Conversely, low spectral efficiency indicates a failure to optimally use available bandwidth, resulting in slower data rates, reduced user capacity, and wasted resources. Such systems may struggle to cope with the surging demand for mobile data, leading to congestion and diminished service quality. On the other hand, high spectral efficiency signifies the optimal utilization of the available spectrum, facilitating increased data transmission rates and accommodating more simultaneous users. These systems are better positioned to deliver high-quality services and support advanced applications, such as video streaming and online gaming, ultimately enhancing the user experience. Thus, striving for higher spectral efficiency is crucial for satisfying the relentless demand for reliable wireless connectivity. The visual data represented in Figs. 5 to 7 further elucidate the relationship between Signal-to-Noise Ratio (SNR) in dB and spectral efficiency values. In Fig. 5, it is apparent that Low-Density Parity-Check (LDPC) codes achieve greater spectral efficiency within the SNR range of -3 dB to 0 dB when utilizing 4QAM modulation. Notably, there is a plateau in spectral efficiency for both coding techniques beyond 0 dB SNR. Transitioning to Fig. 6, LDPC codes again demonstrate higher spectral efficiency from SNR − 3 dB to 4 dB with 16QAM modulation, with no significant changes beyond this threshold for either coding technique. Similarly, Fig. 7 illustrates the trend with 256 QAM modulation, where LDPC codes exhibit higher spectral efficiency up to 10 dB, before leveling off. This analysis underscores the enhanced performance of LDPC-coded 5G MIMO communication systems, particularly those employing 4x16 transmitting and receiving antennas. In conclusion, spectral efficiency remains a fundamental concept in wireless communications, playing a pivotal role in the effectiveness and capacity of communication systems. The pursuit of higher spectral efficiency is imperative for addressing the growing demands for wireless connectivity in our increasingly digital world. Based on the simulation outcomes, it can be concluded that the LDPC-encoded 5G MIMO neural demapper communication system, utilizing a configuration of 4×16 transmitting and receiving antennas, demonstrates superior performance compared to the polar-encoded 5G MIMO neural demapper system. This is evident in both Bit Error Rate (BER) measurements and spectral analysis. 7. Discussion The simulation results for the 5G Multiple Input Multiple Output (MIMO) communication system employing 4-QAM modulation indicate that Low-Density Parity-Check (LDPC) codes significantly outperform polar codes, particularly in terms of Bit Error Rate (BER). At a Signal-to-Noise Ratio (SNR) of -1 dB, LDPC codes achieve a BER of approximately 0.041382, while polar codes yield a BER of 0.094177, demonstrating an impressive 7.23 dB improvement. This superior performance of LDPC codes remains consistent under various conditions, as illustrated in Figs. 3 and 4. These findings highlight the advantages of LDPC coding in bolstering the reliability and efficiency of 5G MIMO systems. Moreover, spectral efficiency—a vital metric in wireless communications—also sees enhancement through the use of LDPC coding, as evidenced in Figs. 5 to 7. LDPC codes exhibit higher spectral efficiency across different modulation schemes (4-QAM, 16-QAM, and 256-QAM) within specific SNR ranges. For instance, with 4-QAM modulation, LDPC codes surpass polar codes between − 3 dB and 0 dB SNR, and this advantage continues with higher modulation schemes up to 10 dB SNR. Such results underscore the effectiveness of LDPC codes for systems that demand high spectral efficiency, particularly in 5G networks utilizing 4×16 MIMO configurations. However, despite the clear benefits, LDPC decoding comes with increased computational complexity when compared to polar codes, potentially leading to higher system latency and energy consumption in large-scale 5G implementations. Additionally, as SNR levels rise, both coding techniques tend to reach a performance plateau, diminishing the relative advantages of LDPC codes. It's also important to note that the simulations are based on ideal conditions, which may not fully represent the real-world challenges posed by interference and hardware limitations. In conclusion, while LDPC coding demonstrates considerable performance enhancements in 5G MIMO systems, particularly at low to moderate SNR levels, the associated complexity and practical application challenges must be thoroughly evaluated. 8. Conclusion This study highlights the effectiveness of incorporating advanced communication techniques, particularly a MIMO system that employs deep neural network (DNN) demappers alongside LDPC and polar coding, in tackling the challenges associated with 5G technology. By concentrating on the enhancement of bit error rate (BER) and spectral efficiency across various orders of Quadrature Amplitude Modulation (QAM), our proposed system demonstrates substantial performance improvements, with LDPC coding notably surpassing polar coding. The findings reveal significant advancements in both reliability and data throughput, positioning the system as highly suitable for the rigorous demands of 5G applications. This research sets a solid foundation for future innovations in robust communication systems capable of supporting a wide array of uses, ranging from improved mobile broadband to extensive IoT connectivity, ultimately contributing to the evolution of next-generation wireless technology. Declarations Conflict of Interests Declaration The author has stated that they do not possess any potential conflicts of interest. Disclosure of Funding This research did not receive any dedicated financial support from any funding agency. Code availability The code is available from the corresponding author by request. Author Contribution Author 1 contributed to generating ideas and implementing them. Author 2 assisted in evaluating results and analyzing findings. Author 3 helped with the literature review, article writing, proofreading, and verifying the article's content. Data Availability The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.In our study, data sharing is not restricted, as we used randomly generated data to train our deep neural network (DNN) demappers. Since no real participant data was involved, there are no privacy concerns, allowing us to share the data and results openly. Figures 2 through 7 display results from the trained DNN based on different parameter values, and Figure 1 illustrates our proposed model. References Zhang J, et al. A Review of 5G Wireless Communication Systems. IEEE Access. 2020; 8: 27042–27053. Andrews J,et al.What Will 5G Be? IEEE Journal on Selected Areas in Communications. 2014; 32(6): 1065–1082. Foschini GJ, et al.On Limits of Wireless Communications in a Fading Environment When Using Multiple Antennas. Wireless Personal Communications. 1998; 6(3): 311–335. Marzetta TL,et al.Noncooperative Cellular Wireless with Unlimited Numbers of Base Station Antennas. IEEE Transactions on Wireless Communications.2016; 9(11): 3590–3600. O'Sheaet al.Deep learning based MIMO communications.2017; arXiv preprint arXiv:1707.07980. Gallageret al.Low-density parity-check codes. IRE Transactions on information theory. 1960; 8(1):21–28. Arikan E,et al. Channel Polarization: A Method for Constructing Capacity-Achieving Codes for Symmetric Binary-Input Memoryless Channels. IEEE Transactions on Information Theory. 2019; 55(7): 3051–3073. Shashanket al. Advanced Decoding Methods for Massive-MIMO Systems Employing Deep Learning. 2024. doi: 10.1109/icaect60202.2024.10468822 . Wanget al. A Unified Deep Learning Based Polar-LDPC Decoder for 5G Communication Systems. 2018. doi: 10.1109/WCSP.2018.8555891 . Vietet al. Performance of Deep Learning LDPC Coded Communications in Large Scale MIMO Channels. 2019.doi: 10.1109/NICS48868.2019.9023820 . MeenalakshmiM, et al.Enhancing channel estimation accuracy in polar-coded MIMO-OFDM systems via CNN with 5G channel models.2023. doi: 10.1016/j.aeue.2023.155016 . Victor C, et al. Multi-input fully CNN for joint pilot decontamination and symbol detection in 5G massive MIMO. Iet Communications. 2023; 17:1899–1906. doi: 10.1049/cmu2.12670 . Heydar A, et al.Massive MIMO Data Detection Using Matrix Inversion Method and Deep Neural Network. 2023; 165–169. doi: 10.1109/apwimob59963.2023.10365630 . Tianyuet al.A Deep Learning based Multi-edge-type decoding algorithm for 5G NR LDPC codes. 2023; 224–228. doi: 10.1109/IWCMC58020.2023.10182649 . Kirillet al.Deep Neural Network Based Decoding of Short 5G LDPC Codes. 2021. doi: 10.1109/REDUNDANCY52534.2021.9606471 Sivaramaet al.CNN-Based Approach for Enhancing 5G LDPC Code Decoding Performance. IEEE Access. 2024;1–1. doi: 10.1109/access.2024.3420106 . Hala F, et al.Ber performance of joint polar coded signal and mimo using millimeter wave. 2021; 4(1):10–18. doi: 10.31987/IJICT.4.1.124 . Mohammed HA, et al.Performance Evaluation of 5g New Radio Ldpc Over Different Scenarios of Lognormal Fading Channel. 2024. doi: 10.2139/ssrn.4718427 . Putri SW, et al.Bit Error Rate Performance Comparison of Low-Density Parity-Check Code and Polar Code. 2023; 1–6. doi: 10.1109/ict60153.2023.10374061 . Salima B, et al.On error correction performance of LDPC and Polar codes for the 5G Machine Type Communications. 2021; 1–4. doi: 10.1109/IEMTRONICS52119.2021.9422665 . Joyet al.5G Systems with Low Density Parity Check based Chanel Coding for Enhanced Mobile Broadband Scheme. 2020;2(1):42–49. doi: 10.36548/JSWS.2020.1.005 . Guntur KP,et al.Evaluation of LDPC Code and Polar Code Coding Scheme in 5G Technology – Massive Machine Type Communication. 2018; 170–174. doi: 10.1109/ICITEED.2018.8534937 Nguyenet al. The Effectiveness of LDPC Decoding Algorithms in 5G Channel Modelling of MIMO-OFDM System under the Influence of Spatial Correlation. JST: Smart Systems and Devices. 2022;32(2):6–13.doi: 10.51316/jst.158.ssad.2022.32.2.2 . Pooja M, et al. Performance Analysis of Polar codes for Next Generation 5G Technology. 2022;1–4. doi: 10.1109/incet54531.2022.9824746 . Aizazul H et al.Addressing tactic volatility in self-adaptive systems using evolved recurrent neural networks and uncertainty reduction tactics. 2022. doi: 10.1145/3512290.3528745 . Supreeth BR,et al.Autonomous Heterogeneous Network for Mixed Strategy Analysis in Defence Attack Game Model Based on 5G in Machine Learning Environment. 5th International Conference on Contemporary Computing and Informatics (IC3I). 2022;2029–2033. doi: 10.1109/ic3i56241.2022.10072526 . Mebratu A. Selection of Architectural Patterns based on Tactics. 2022;13–18. doi: 10.1109/ICT4DA56482.2022.9971369 Osseiran A, et al. 5G: The Next Generation of Mobile Communication. IEEE Communications Magazine. 2014;52(2):38–45. Bertocco J, et al.Deep Learning for Wireless Communications: A Review. IEEE Communications Surveys & Tutorials. 2019;21(3):1446–1468. Zhang C, et al. Machine Learning for Signal Detection in 5G Wireless Networks. IEEE Journal on Selected Areas in Communications.2020;38(2):245–258 Wang Z, et al. Performance Analysis of Deep Neural Network Demappers for 5G Wireless Systems. IEEE Transactions on Wireless Communications. 2019;18(12):5753–5767. Li Y, et al.Spectral Efficiency Improvement for 5G Systems Using Deep Learning Techniques. Journal of Communications and Networks. 2021;23(1):24–35. Yuan J, et al.Challenges and Opportunities in Explainable Artificial Intelligence for Wireless Communications. IEEE Wireless Communications. 2022;29(1):56–62. Richardson T, et al.The Capacity of Low-Density Parity-Check Codes Under Message-Passing Decoding. IEEE Transactions on Information Theory. 2001;47(2):599–618. MacKay DJC. Information Theory, Inference, and Learning Algorithms.Cambridge University Press. 2003. Fossorier M. Decoding Algorithms for Non-Binary Low-Density Parity-Check Codes. IEEE Transactions on Information Theory. 1995;41(2):224–234. Shannon CE. A Mathematical Theory of Communication. Bell System Technical Journal. 1948;27(3): 379–423. Chung SY, et al.On the Design of LDPC Codes." IEEE Transactions on Information Theory. 2001;47(7):2460–2470. Arıkan E. Channel Polarization: A Method for Constructing Capacity-Achieving Codes for Symmetric Binary-Input Memoryless Channels. IEEE Transactions on Information Theory. 2009;55(7):3051–3073. Tal I, et al.List Decoding of Polar Codes. IEEE Transactions on Information Theory. 2013;59(11):6218–6228. Chen X, et al. Polar Codes for 5G Wireless Communication Systems. IEEE Access. 2019;7: 103627–103641. Haykin S. Communication Systems. John Wiley & Sons. 2008. Proakis JG, et al.Digital Communications. McGraw-Hill. 2007. Van Trees HL. Detection, Estimation, and Modulation Theory, Part I. Wiley-Interscience. 2002. Garg VK. Wireless Communication and Networking. Morgan Kaufmann. 2010. HaykinS. Adaptive filter theory. Pearson Education. 2001. Haykin S. Adaptive Filter Theory. 4th ed. Prentice Hall. 2002. Van de Beek JJ,et al.On the Search for the Optimal Linear Equalizer. IEEE Transactions on Signal Processing. 1997;45(11): 2850–2861. Rappaport TS. Wireless Communications: Principles and Practice. 2nd ed. Prentice Hall. 2002. Goldsmith A. Wireless Communications. Cambridge University Press. 2005. Proakis JG et al.Digital Communications. 5th ed. McGraw-Hill. 2008. Additional Declarations No competing interests reported. Supplementary Files SupplementaryFile.docx Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 09 Dec, 2024 Reviews received at journal 29 Nov, 2024 Reviews received at journal 23 Nov, 2024 Reviewers agreed at journal 20 Nov, 2024 Reviewers agreed at journal 18 Nov, 2024 Reviewers agreed at journal 18 Nov, 2024 Reviewers agreed at journal 18 Nov, 2024 Reviewers agreed at journal 18 Nov, 2024 Reviewers invited by journal 18 Nov, 2024 Editor assigned by journal 06 Nov, 2024 Submission checks completed at journal 05 Nov, 2024 First submitted to journal 21 Oct, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5302008","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":378833646,"identity":"e4369c0c-a026-4fb4-a290-ac0bcb669094","order_by":0,"name":"Md. Ashraful Islam","email":"","orcid":"","institution":"University of Rajshahi","correspondingAuthor":false,"prefix":"","firstName":"Md.","middleName":"Ashraful","lastName":"Islam","suffix":""},{"id":378833647,"identity":"162bcf10-9a1b-437f-b9c1-39ede9d6477d","order_by":1,"name":"Md. Firoz Ahmed","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3klEQVRIie3PsQrCMBCA4cjBZTnpGkH0FU4KRRD0VSpCpwo+gKhQcNRX6eQcCPYZXIvg4CQUxEHEVhyc2roJ5h+OEO6DRAib7QeTUTFZCATQ+Um1KwmZN3Ek+gWhaqLfh9aW+HVRTWQzTW+zYYcNZefDvE9Cmn1cSkC6rHjismnuBmGSP4yC4FBGRoComPU4zokbYk4UeaWECuKzXsaGTm74qEs0a78VERyn63oEeiue9LaAHkw3irDqL+QkjfR2H3bRMccsvC46jjRJKfkM1WvWXS+CyzfbNpvN9j89AfCCOsocjWfxAAAAAElFTkSuQmCC","orcid":"","institution":"University of Rajshahi","correspondingAuthor":true,"prefix":"","firstName":"Md.","middleName":"Firoz","lastName":"Ahmed","suffix":""},{"id":378833648,"identity":"929edf90-c3b3-4bc5-9bb7-792a4e1f3cf6","order_by":2,"name":"Md. Matiqul Islam","email":"","orcid":"","institution":"University of Rajshahi","correspondingAuthor":false,"prefix":"","firstName":"Md.","middleName":"Matiqul","lastName":"Islam","suffix":""}],"badges":[],"createdAt":"2024-10-21 07:23:23","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5302008/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5302008/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":69228782,"identity":"f09b9804-0ec2-4708-afb9-fbec7e3d8057","added_by":"auto","created_at":"2024-11-18 08:29:00","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":70235,"visible":true,"origin":"","legend":"\u003cp\u003eBlock diagram of the 5G channel-coded 4×16 MIMO neural demapper-based communication system\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-5302008/v1/f94ca3414c09b37b8ebc7ef3.png"},{"id":69226902,"identity":"a3aed4ec-93a3-48a1-8f10-9872bc699d04","added_by":"auto","created_at":"2024-11-18 08:13:00","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":103040,"visible":true,"origin":"","legend":"\u003cp\u003eBER performance under 4-QAM modulation\u003c/p\u003e","description":"","filename":"image2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5302008/v1/6c247db632486b7945bc4d56.jpeg"},{"id":69227099,"identity":"b62bfad6-9bf9-45a5-9b9d-22a8680da00b","added_by":"auto","created_at":"2024-11-18 08:21:00","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":125390,"visible":true,"origin":"","legend":"\u003cp\u003eBER performance under 16-QAM modulation.\u003c/p\u003e","description":"","filename":"image3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5302008/v1/e2acabc8d7f216f18eea5dfc.jpeg"},{"id":69226903,"identity":"14396b11-bbdb-40f3-a4a8-3de142100a29","added_by":"auto","created_at":"2024-11-18 08:13:00","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":114557,"visible":true,"origin":"","legend":"\u003cp\u003eBER performance under 256-QAM modulation.\u003c/p\u003e","description":"","filename":"image4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5302008/v1/f1b99e670d12eb5acc39006b.jpeg"},{"id":69226904,"identity":"9c0afcd3-f197-4a72-988f-2b55792dd3c4","added_by":"auto","created_at":"2024-11-18 08:13:00","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":25894,"visible":true,"origin":"","legend":"\u003cp\u003eEffect of Spectral Efficiency under 4-QAM modulation.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-5302008/v1/3ffbb0d2159aa0dd75ac20f7.png"},{"id":69226905,"identity":"6735f8bf-0ee3-4f74-bf02-f84635c9676e","added_by":"auto","created_at":"2024-11-18 08:13:00","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":21497,"visible":true,"origin":"","legend":"\u003cp\u003eEffect of Spectral Efficiency under 16-QAM modulation.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-5302008/v1/e791f6dab6ecdf49016fda29.png"},{"id":69226907,"identity":"39ac1006-b219-4a1b-bc8f-48cf420ad227","added_by":"auto","created_at":"2024-11-18 08:13:00","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":22739,"visible":true,"origin":"","legend":"\u003cp\u003eEffect of Spectral Efficiency under 256-QAM modulation.\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-5302008/v1/0b9ea35da4184e598191c935.png"},{"id":69228783,"identity":"aa236d06-aa1f-4fdc-a164-0790adb68c35","added_by":"auto","created_at":"2024-11-18 08:29:05","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":984938,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5302008/v1/eed6c826-d27e-4a93-af56-ffb55bc7599c.pdf"},{"id":69226909,"identity":"36949d41-486d-419c-9804-104f35ad6c9e","added_by":"auto","created_at":"2024-11-18 08:13:00","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":21074,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFile.docx","url":"https://assets-eu.researchsquare.com/files/rs-5302008/v1/919c5044b618286c6b8a15e8.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"A DNN-based 5G MIMO System Adopting a Mix of Tactics","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eThe rapid evolution of 5G technology has introduced a transformative phase in wireless communication, necessitating innovative solutions to meet the increasing demands for higher data rates, enhanced reliability, and improved spectral efficiency [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. With applications spanning enhanced mobile broadband (eMBB), massive machine-type communications (mMTC), and ultra-reliable low-latency communications (URLLC), 5G systems are challenged to manage complex performance requirements while promoting user connectivity [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. A critical component in addressing these challenges is the implementation of Multiple Input Multiple Output (MIMO) systems, which utilize multiple antenna elements to enhance data throughput and link reliability substantially [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. MIMO technology operates on spatial diversity and multiplexing, allowing a single wireless channel to transmit multiple data streams simultaneously. This capability is vital in high-demand environments, where conventional methods may struggle to maintain efficiency and performance due to increased interference and fading effects. Despite the advantages of MIMO systems, traditional techniques face limitations in terms of computational complexity and adaptability to varying channel conditions, particularly in high-order modulation scenarios [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eRecent studies have proposed integrating advanced signal processing methods to overcome these limitations to improve performance. DNN demappers represent a significant advancement in the decoding process for MIMO systems. By leveraging deep learning algorithms, DNN demappers can learn complex mappings from received signals to transmitted symbols, effectively improving BER performance [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. This approach has been shown to outperform traditional demapping techniques, especially in high-order Quadrature Amplitude Modulation (QAM) scenarios. Moreover, coding techniques such as LDPC and polar coding have gained prominence for their ability to approach channel capacity. LDPC codes provide excellent error correction capabilities, vital for maintaining performance in noisy environments [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. In contrast, polar codes, introduced by [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e], have emerged as a powerful alternative with their capacity-achieving properties, particularly in high-throughput applications. Both coding schemes can be effectively combined with MIMO systems to enhance spectral efficiency and reliability.\u003c/p\u003e \u003cp\u003eThis paper offers a thorough analysis of a 5G Multiple Input Multiple Output (MIMO) system that integrates deep neural network (DNN) demappers, Low-Density Parity-Check (LDPC) coding, and polar coding techniques, forming a mixed strategy (MS) to enhance overall performance while overcoming the limitations of traditional MIMO methods. This study aims to assess the effectiveness of this MS in augmenting bit error rate (BER) and spectral efficiency across various orders of Quadrature Amplitude Modulation (QAM). The findings demonstrate substantial improvements in both BER and spectral efficiency, underscoring the potential of this MS approach for future wireless communication systems.\u003c/p\u003e"},{"header":"2. Literature Review","content":"\u003cp\u003eThis section focuses on 5G MIMO systems and highlights the integration of cutting-edge signal processing techniques, particularly deep neural networks (DNNs), Low-Density Parity-Check (LDPC) coding, and polar coding. These advancements are designed to enhance performance metrics, including bit error rate (BER) and spectral efficiency, across various orders of Quadrature Amplitude Modulation (QAM). DNNs have been adopted for decoding processes in Massive-MIMO systems, leading to significant improvements in both decoding performance and spectral efficiency [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. A comprehensive DNN-based decoder for polar-LDPC codes achieves performance levels comparable to traditional decoding methods while optimizing resource utilization [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Furthermore, LDPC codes implemented in large-scale MIMO configurations demonstrate a decrease in performance gaps as the number of antennas increases [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. The use of CNNs for channel estimation in polar-coded MIMO-OFDM systems results in notable reductions in BER and mean square error, thereby enhancing the overall reliability of the system [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eRecent studies have introduced innovative techniques such as a multi-input deep learning-based joint pilot decontamination and symbol detection mechanism tailored for 5G massive multiple-input multiple-output (MAMIMO) systems [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Additionally, a method utilizing matrix inversion for data detection in Massive MIMO employing DNNs has been proposed [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. A deep learning-based algorithm for multi-edge type decoding of 5G NR LDPC codes has also been presented [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. The decoding of short 5G LDPC codes via DNNs has shown promising results [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Furthermore, a CNN-based strategy aimed at enhancing the decoding performance of 5G LDPC codes has been suggested [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Studies have also explored the BER performance of joint polar-coded signals in MIMO systems utilizing millimeter-wave technology [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], along with the performance evaluation of 5G New Radio LDPC codes across various log-normal fading channel scenarios [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. The comparison of BER performance between Low-Density Parity-Check codes and polar codes has been conducted [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e], as well as an analysis of error correction capabilities of LDPC and polar codes in the context of 5G Machine Type Communications [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. Proposals for 5G systems utilizing Low-Density Parity Check-based channel coding for enhanced mobile broadband applications have been made [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e], alongside evaluations of LDPC and polar coding schemes within the framework of 5G technology, particularly for Massive Machine Type Communication [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Lastly, the efficacy of LDPC decoding algorithms in 5G channel modeling for MIMO-OFDM systems under spatial correlation influences has been investigated [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e], and the performance of 5G LDPC and polar decoding in spatially correlated MIMO-OFDMA systems has been analyzed [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. Insights into the performance of polar codes for next-generation 5G technology have also been discussed [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e].\u003c/p\u003e"},{"header":"3. Methods","content":"\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Mix of Tactics\u003c/h2\u003e \u003cp\u003eA \"mix of tactics\" refers to the strategic blend of various advanced techniques aimed at optimizing system performance. In this scenario, this combination includes the integration of deep neural network (DNN) demappers, Low-Density Parity-Check (LDPC) coding, and polar coding. Each of these techniques offers distinct advantages:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eDNN demappers leverage machine learning algorithms to significantly improve signal detection accuracy, particularly in challenging and noisy environments [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e].\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eLDPC coding serves as a robust error-correcting technique, enhancing reliability by efficiently correcting transmission errors.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003ePolar coding is another effective error-correcting method, recognized for its strong performance even at low signal-to-noise ratios [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e].\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eBy uniting these approaches, the system gains improved error correction, enhanced signal detection, and overall communication reliability, resulting in a more resilient and efficient solution for demanding applications like 5G technology[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Deep Neural Network (DNN) Demappers\u003c/h2\u003e \u003cp\u003eDeep neural network (DNN) demappers represent a revolutionary advancement in signal detection for 5G wireless systems, particularly for sophisticated modulation schemes. Traditional demapping techniques often encounter significant challenges in environments characterized by noise and interference. In contrast, DNN demappers leverage the power of artificial intelligence (AI) and machine learning (ML) to enhance bit error rate (BER) performance and overall system efficiency [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. At the core of DNNs is a structure composed of multiple layers of neurons, which are capable of learning complex mappings between input signal samples and the corresponding output symbols. This capability makes DNNs exceptionally effective in demapping tasks that typically hinder conventional methods [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. Their proficiency in handling high-dimensional data is particularly advantageous for next-generation modulation schemes, such as 64-QAM and 256-QAM, allowing them to adapt seamlessly to varying channel conditions [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. Research has demonstrated that DNN demappers significantly outperform traditional methods, achieving substantial reductions in BER even in high-noise environments [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. Additionally, they contribute to improved spectral efficiency, thereby supporting advanced modulation formats that enhance system throughput [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. Nonetheless, challenges remain, including issues related to computational complexity, the risk of overfitting, and the interpretability of the models [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. Ongoing advancements in hardware and research into explainable AI are actively addressing these concerns. In conclusion, DNN demappers are set to play a pivotal role in advancing BER, spectral efficiency, and overall performance in future 5G communication systems, marking a significant leap forward in the evolution of wireless technology.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Low-Density Parity-Check (LDPC) coding\u003c/h2\u003e \u003cp\u003eLow-Density Parity-Check (LDPC) coding represents a sophisticated approach to error correction utilized in modern communication systems, aiming to enhance data reliability by accurately detecting and correcting errors during transmission [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Characterized by a sparse parity-check matrix\u0026mdash;where the majority of the elements are zeros\u0026mdash;LDPC codes minimize computational complexity while providing robust error correction capabilities [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. The fundamental principle behind LDPC coding involves the incorporation of redundant information into transmitted data. This redundancy enables the receiver to identify and rectify errors, even in the presence of significant noise or interference [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. A common method for decoding LDPC codes is the use of iterative algorithms, notably the belief propagation algorithm, which systematically checks for and resolves inconsistencies between the transmitted and received bits [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. LDPC codes are notably efficient, approaching the theoretical limits of channel capacity as outlined by Shannon [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. This characteristic makes them particularly suitable for high-demand applications, including 5G technology, Wi-Fi networks, and satellite communications. By achieving a favorable balance between complexity and performance, LDPC codes ensure low bit error rates (BER) while maximizing data throughput. The key advantages of LDPC codes include:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eHigh Error Correction Capability\u003c/b\u003e: They perform exceptionally well in noisy channel environments.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eScalability\u003c/b\u003e: LDPC codes can be applied across a range of block lengths and code rates, making them versatile.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eEfficiency\u003c/b\u003e: They are designed to approach optimal performance limits, ensuring reliable communication even under challenging conditions.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eDue to their robustness and efficiency, LDPC coding has become a staple in contemporary communication systems, playing a vital role in enhancing data transmission capabilities in advanced networks like 5G [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Polar Coding\u003c/h2\u003e \u003cp\u003ePolar coding is a groundbreaking error correction technique that enhances reliable data transmission by polarizing communication channels into distinct groups of highly reliable and unreliable subchannels. Introduced by Erdal Arıkan in 2009 [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e], polar coding is notable for being the first coding scheme to rigorously demonstrate its capability to achieve Shannon capacity\u0026mdash;the theoretical maximum data rate for error-free transmission\u0026mdash;through a straightforward encoding and decoding process [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. The core principle of polar coding is channel polarization, where a communication channel is decomposed into virtual subchannels. As the code length increases, certain subchannels become highly reliable while others turn unreliable. The strategic approach involves transmitting vital information bits through these reliable subchannels, while assigning fixed or random (frozen) bits to the unreliable ones, thus optimizing the error correction process. Polar codes utilize a recursive structure based on XOR operations, which transforms the original message into a longer codeword that is spread across the polarized subchannels. The decoding process typically employs the Successive Cancellation (SC) algorithm, as well as its enhanced variant, the Successive Cancellation List (SCL) decoding. SCL decoding further mitigates errors by assessing multiple decoding paths and selecting the most optimal one [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]. Notably, polar codes are capacity-achieving, rendering them highly effective for noisy communication channels. Their computational efficiency in both encoding and decoding, coupled with their adaptability across various block lengths and code rates, makes them ideal for real-time applications. In the context of 5G networks, polar codes are particularly advantageous for control channels, given their strong performance with short block lengths and low error rates. They are often employed alongside other coding techniques, such as Low-Density Parity-Check (LDPC) codes, for data channels within 5G systems [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. In conclusion, polar coding signifies a remarkable advancement in the field of channel coding, harmonizing theoretical sophistication with practical applicability for next-generation wireless communications.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e3.5 \u003cb\u003eLinear Minimum Mean Square Error (LMMSE) equalizer\u003c/b\u003e\u003c/h2\u003e \u003cp\u003eThe Linear Minimum Mean Square Error (LMMSE) equalizer is an advanced signal processing method widely utilized in contemporary communication systems to mitigate the adverse effects of noise and interference during the process of signal detection. The primary goal of the LMMSE equalizer is to estimate the transmitted signal while minimizing the mean square error (MSE) between the estimated signal and the actual transmitted signal, all while taking into consideration both channel distortions and noise [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. Functioning as a linear filter, the LMMSE equalizer transforms the received signal into a more accurate estimate of the original transmitted signal. This attribute makes the LMMSE equalizer computationally efficient, which is crucial for real-time applications in dynamic communication environments [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. By concentrating on minimizing the MSE\u0026mdash;defined as the average squared difference between the estimated and actual transmitted signals\u0026mdash;the LMMSE equalizer effectively reduces noise and interference, thereby helping to maintain the integrity of the signal. This technique has shown exceptional performance in channels characterized by additive noise, such as Gaussian noise, and in scenarios experiencing distortion or fading, both of which are prevalent in wireless communication systems [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. Moreover, the LMMSE equalizer is most effective when the statistical properties of the channel and noise are known or can be accurately estimated, allowing for optimized filtering and signal recovery [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn a communication system, the received signal, denoted as y, can be represented as the sum of the transmitted signal (\u003cem\u003ex\u003c/em\u003e) affected by the communication channel (\u003cem\u003eH\u003c/em\u003e) and noise (\u003cem\u003en\u003c/em\u003e):\u003c/p\u003e \u003cp\u003e \u003cem\u003ey\u003c/em\u003e\u0026thinsp;=\u0026thinsp;\u003cem\u003eHx\u003c/em\u003e\u0026thinsp;+\u0026thinsp;\u003cem\u003en\u003c/em\u003e............(1)\u003c/p\u003e \u003cp\u003eHere, \u003cem\u003eH\u003c/em\u003e encapsulates the effects of fading and multipath experienced during transmission, while \u003cem\u003en\u003c/em\u003e represents the additive noise, often modeled as Gaussian noise.\u003c/p\u003e \u003cp\u003eThe goal of the Linear Minimum Mean Square Error (LMMSE) equalizer is to effectively estimate the original transmitted signal (\u003cem\u003ex\u003c/em\u003e) from the received signal (\u003cem\u003ey\u003c/em\u003e) by applying a linear filter (\u003cem\u003eW\u003c/em\u003e):\u003c/p\u003e \u003cp\u003e \u003cem\u003ex̂\u003c/em\u003e = \u003cem\u003eWy\u003c/em\u003e..................(2)\u003c/p\u003e \u003cp\u003eWhere, \u003cem\u003ex̂\u003c/em\u003e represents the estimated transmitted signal.\u003c/p\u003e \u003cp\u003eThe LMMSE equalizer aims to minimize the Mean Square Error (MSE) between the estimated signal \u003cem\u003ex̂\u003c/em\u003e and the actual transmitted signal \u003cem\u003ex\u003c/em\u003e.\u003c/p\u003e \u003cp\u003eThis optimal filter \u003cem\u003eW\u003c/em\u003e is determined by leveraging the channel matrix (\u003cem\u003eH\u003c/em\u003e) and the noise covariance matrix (\u003cem\u003eR\u003c/em\u003e\u003csub\u003e\u003cem\u003en\u003c/em\u003e\u003c/sub\u003e):\u003c/p\u003e \u003cp\u003e \u003cem\u003eW\u003c/em\u003e = (H\u003csup\u003eH\u003c/sup\u003eH + \u003cem\u003eR\u003c/em\u003e\u003csub\u003e\u003cem\u003en\u003c/em\u003e\u003c/sub\u003e)\u003csup\u003e\u0026minus;1\u003c/sup\u003e\u003cem\u003eH\u003c/em\u003e\u003csup\u003e\u003cem\u003eH\u003c/em\u003e\u003c/sup\u003e.................\u003cem\u003e(3)\u003c/em\u003e\u003c/p\u003e \u003cp\u003eHere, \u003cem\u003eH\u003c/em\u003e\u003csup\u003e\u003cem\u003eH\u003c/em\u003e\u003c/sup\u003e denotes the conjugate transpose of \u003cem\u003eH\u003c/em\u003e. This formulation allows the LMMSE equalizer to effectively mitigate the effects of noise and interference, thereby enhancing the accuracy of signal recovery in communication systems [\u003cspan additionalcitationids=\"CR47\" citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe LMMSE equalizer proves particularly valuable in scenarios characterized by high noise and interference, such as wireless communication channels. By minimizing the bit error rate (BER) and enhancing overall signal quality, it plays a crucial role in improving communication reliability. This makes it particularly relevant in modern communication systems like Multiple Input Multiple Output (MIMO) and Orthogonal Frequency Division Multiplexing (OFDM) systems, prevalent in 4G and 5G networks. Despite its effectiveness, the LMMSE equalizer necessitates knowledge of the channel and noise characteristics, often requiring estimation in real-world settings, potentially introducing errors. Moreover, the matrix inversion involved in its computation can be computationally demanding, particularly in MIMO systems with numerous antennas.\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Flat Fading Channel","content":"\u003cp\u003eA flat fading channel is a specific type of wireless communication channel characterized by a coherence bandwidth that exceeds the bandwidth of the transmitted signal. In this context, all frequency components of the signal experience identical fading characteristics, resulting in uniform attenuation or amplification across the entire bandwidth. The amplitude of the transmitted signal can fluctuate over time due to variations in the wireless environment, which may include the movement of the transmitter, receiver, or surrounding objects. Since the signal bandwidth is smaller than the channel's coherence bandwidth, there is minimal frequency-dependent fading. This means that the channel impacts all segments of the transmitted signal uniformly, in contrast to frequency-selective fading, where different portions of the signal spectrum experience varying levels of fading. Flat fading typically arises from a single dominant propagation path, such as in line-of-sight communication, or when multipath components are closely spaced in time and exert similar effects on the signal [\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn a flat fading channel, the received signal\u003c/p\u003e \u003cp\u003e \u003cem\u003ey\u003c/em\u003e (\u003cem\u003et\u003c/em\u003e)\u0026thinsp;=\u0026thinsp;\u003cem\u003eh\u003c/em\u003e (\u003cem\u003et\u003c/em\u003e) \u0026sdot;\u003cem\u003ex\u003c/em\u003e (\u003cem\u003et\u003c/em\u003e)\u0026thinsp;+\u0026thinsp;\u003cem\u003en\u003c/em\u003e (\u003cem\u003et\u003c/em\u003e)..................(4)\u003c/p\u003e \u003cp\u003eWhere, \u003cem\u003ex\u003c/em\u003e(\u003cem\u003et\u003c/em\u003e) is the transmitted signal, \u003cem\u003eh\u003c/em\u003e(\u003cem\u003et\u003c/em\u003e) is the time-varying fading coefficient (a complex scalar) that represents the attenuation and phase shift introduced by the channel, \u003cem\u003en\u003c/em\u003e(\u003cem\u003et\u003c/em\u003e) is the additive noise.\u003c/p\u003e \u003cp\u003eThe fading coefficient \u003cem\u003eh\u003c/em\u003e (\u003cem\u003et\u003c/em\u003e) typically follows a Rayleigh or Rician distribution, depending on the presence of a line-of-sight path. In scenarios where no dominant line-of-sight component exists, the fading tends to be Rayleigh-distributed; conversely, if a line-of-sight path is present, the fading follows a Rician distribution [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e]. Flat fading influences both the amplitude and phase of the signal uniformly across its entire bandwidth, which can lead to signal degradation, particularly in high-mobility environments that induce rapid fading. This may result in deep fades, where the signal power experiences significant momentary drops, leading to errors in signal detection and increased bit error rates (BER). To counteract these effects, various techniques\u0026mdash;such as diversity, channel coding, and equalization\u0026mdash;are employed. Flat fading channels are commonly observed in narrowband communication systems, such as certain cellular or satellite communications, where the transmitted signal bandwidth is small relative to the channel's coherence bandwidth [\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e]. In contrast, for wideband systems, like those utilized in broadband wireless communications (e.g., 4G and 5G), frequency-selective fading becomes increasingly significant.\u003c/p\u003e"},{"header":"5. Proposed System","content":"\u003cp\u003eThe block diagram illustrates in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea 5G channel-coded 4\u0026times;16 MIMO neural demapper-based communication system, showcasing a modern wireless communication design with error correction, modulation, and machine learning enhancements.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eBinary Source\u003c/strong\u003e \u003cp\u003eThis generates the input data in binary form.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003e5G Encoder (LDPC and Polar Coding)\u003c/strong\u003e \u003cp\u003eThe binary data is encoded using LDPC (Low-Density Parity-Check) and Polar Codes, both commonly used in 5G for error correction.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eQAM Mapper\u003c/strong\u003e \u003cp\u003eThe encoded bits are modulated using Quadrature Amplitude Modulation (QAM) to form symbols for transmission.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eFlat Fading Channel\u003c/strong\u003e \u003cp\u003eThe signal passes through a flat fading channel, which simulates the wireless channel's effects like fading and noise.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eLMMSE Equalizer\u003c/strong\u003e \u003cp\u003eA Linear Minimum Mean Square Error (LMMSE) equalizer is used to counteract the distortions caused by the fading channel, improving signal recovery.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eNeural Demapper\u003c/strong\u003e \u003cp\u003eA neural network with dense layers demaps the received signal back into bits. This step replaces traditional demapping methods with machine learning for potentially better accuracy.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003e5G Decoder (LDPC and Polar Coding)\u003c/strong\u003e \u003cp\u003eThe demapped bits are decoded using LDPC and Polar Codes to retrieve the original data.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eCompute BER\u003c/strong\u003e \u003cp\u003eThe Bit Error Rate (BER) is calculated to evaluate the system's performance by comparing the original and decoded bits.\u003c/p\u003e \u003c/p\u003e \u003cp\u003eThis system integrates MIMO (Multiple Input, Multiple Output) technology with machine learning-based demapping to enhance performance in 5G communication environments. The Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e below displays the deep neural network (DNN)-based demapper's training settings.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eNeural demapper training specifications\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTraining Parameters\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eValue\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBatch size\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e64\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEpochs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOptimizer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSGD\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLoss function\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBinary cross entropy\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModulation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eM-ary (M\u0026thinsp;=\u0026thinsp;4,16,256) QAM\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eChannel\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFlat Fading\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e"},{"header":"6. Simulated Results","content":"\u003cp\u003eThis section presents a comparative analysis of the performance between a novel LDPC-coded 5G MIMO deep neural network (DNN) demapper communication system and a polar-coded 5G MIMO neural demapper system, evaluated across various QAM modulation orders. The primary focus is on assessing both systems in terms of their bit error rate (BER) while also considering the influence of spectral efficiency resulting from different coding techniques.\u003c/p\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e6.1 BER analysis\u003c/h2\u003e \u003cp\u003eThe Bit Error Rate (BER) is a crucial metric in the realm of wireless communications, functioning as an essential indicator of the integrity of data transmission. It quantifies the frequency of errors occurring during data transmission, providing insights into the quality of the communication link. An elevated BER signifies a decline in link quality, resulting in increased data corruption, a higher rate of retransmissions, and a reduction in overall system efficiency. In contrast, a low BER is a testament to superior link quality, characterized by fewer errors, enhanced data integrity, and improved system performance. Several factors can influence the BER, including the signal-to-noise ratio (SNR), the modulation techniques utilized, error correction coding methods, and the prevailing channel conditions, such as fading and interference. It is imperative to effectively manage the BER to achieve an optimal balance between data rate and reliability in wireless communication systems.\u003c/p\u003e \u003cp\u003eIn our simulation study, we explored performance in a flat fading channel with SNR values ranging from \u0026minus;\u0026thinsp;5 dB to 15 dB. The analysis focused on a proposed 5G MIMO communication system that employs advanced techniques such as polar coding and Low-Density Parity-Check (LDPC) coding, alongside various Quadrature Amplitude Modulation (QAM) techniques (4-QAM, 16-QAM, and 256-QAM). The performance evaluation was conducted on a MIMO system equipped with 4\u0026times;16 transmitting and receiving antennas. The simulation results are visually represented in Figs.\u0026nbsp;2 to 4, which showcase the spectra for both polar-coded and LDPC-coded 5G MIMO communication systems. In these figures, the solid blue curve symbolizes the performance of the polar-coded system, while the solid red curve illustrates the LDPC-coded results, both utilizing the 4\u0026times;16 antenna configuration.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn Fig.\u0026nbsp;2, we present the simulation outcomes for the 5G MIMO communication system utilizing 4-QAM modulation. Notably, at an SNR of -1 dB, the LDPC code achieves a Bit Error Rate (BER) of approximately 0.041382, in stark contrast to the polar code's BER of 0.094177. This finding underscores the superiority of the LDPC code, which provides a substantial improvement of 7.23 dB over the polar code. Overall, the simulation results highlight the enhanced performance of the LDPC-coded 5G MIMO neural demapper-based communication system. Furthermore, Figs.\u0026nbsp;3 and 4 further validate that the LDPC-coded MIMO communication system consistently outperforms its polar-coded counterpart, showcasing significant performance enhancements. These findings reaffirm the importance of selecting appropriate coding strategies and modulation techniques in advancing the reliability and efficiency of wireless communication systems.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e6.2 Spectral efficiency\u003c/h2\u003e \u003cp\u003eSpectral efficiency is a critical metric that quantifies the effectiveness of a wireless communication system in utilizing its designated bandwidth. Typically measured in bits per second per Hertz (bps/Hz), spectral efficiency reflects the volume of data that can be transmitted over a specific frequency spectrum. Its significance cannot be overstated, as it has a direct correlation with the capacity and performance of wireless networks. In an era marked by an ever-increasing demand for wireless services, high spectral efficiency enables a greater number of users to connect and enjoy higher data rates within the same frequency band. This efficiency is vital for effective resource allocation, minimizing latency, and enhancing overall network reliability. Conversely, low spectral efficiency indicates a failure to optimally use available bandwidth, resulting in slower data rates, reduced user capacity, and wasted resources. Such systems may struggle to cope with the surging demand for mobile data, leading to congestion and diminished service quality. On the other hand, high spectral efficiency signifies the optimal utilization of the available spectrum, facilitating increased data transmission rates and accommodating more simultaneous users. These systems are better positioned to deliver high-quality services and support advanced applications, such as video streaming and online gaming, ultimately enhancing the user experience. Thus, striving for higher spectral efficiency is crucial for satisfying the relentless demand for reliable wireless connectivity.\u003c/p\u003e \u003cp\u003eThe visual data represented in Figs.\u0026nbsp;5 to 7 further elucidate the relationship between Signal-to-Noise Ratio (SNR) in dB and spectral efficiency values.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn Fig.\u0026nbsp;5, it is apparent that Low-Density Parity-Check (LDPC) codes achieve greater spectral efficiency within the SNR range of -3 dB to 0 dB when utilizing 4QAM modulation. Notably, there is a plateau in spectral efficiency for both coding techniques beyond 0 dB SNR. Transitioning to Fig.\u0026nbsp;6, LDPC codes again demonstrate higher spectral efficiency from SNR \u0026minus;\u0026thinsp;3 dB to 4 dB with 16QAM modulation, with no significant changes beyond this threshold for either coding technique.\u003c/p\u003e \u003cp\u003eSimilarly, Fig.\u0026nbsp;7 illustrates the trend with 256 QAM modulation, where LDPC codes exhibit higher spectral efficiency up to 10 dB, before leveling off. This analysis underscores the enhanced performance of LDPC-coded 5G MIMO communication systems, particularly those employing 4x16 transmitting and receiving antennas. In conclusion, spectral efficiency remains a fundamental concept in wireless communications, playing a pivotal role in the effectiveness and capacity of communication systems. The pursuit of higher spectral efficiency is imperative for addressing the growing demands for wireless connectivity in our increasingly digital world.\u003c/p\u003e \u003cp\u003eBased on the simulation outcomes, it can be concluded that the LDPC-encoded 5G MIMO neural demapper communication system, utilizing a configuration of 4\u0026times;16 transmitting and receiving antennas, demonstrates superior performance compared to the polar-encoded 5G MIMO neural demapper system. This is evident in both Bit Error Rate (BER) measurements and spectral analysis.\u003c/p\u003e \u003c/div\u003e"},{"header":"7. Discussion","content":"\u003cp\u003eThe simulation results for the 5G Multiple Input Multiple Output (MIMO) communication system employing 4-QAM modulation indicate that Low-Density Parity-Check (LDPC) codes significantly outperform polar codes, particularly in terms of Bit Error Rate (BER). At a Signal-to-Noise Ratio (SNR) of -1 dB, LDPC codes achieve a BER of approximately 0.041382, while polar codes yield a BER of 0.094177, demonstrating an impressive 7.23 dB improvement. This superior performance of LDPC codes remains consistent under various conditions, as illustrated in Figs.\u0026nbsp;3 and 4. These findings highlight the advantages of LDPC coding in bolstering the reliability and efficiency of 5G MIMO systems. Moreover, spectral efficiency\u0026mdash;a vital metric in wireless communications\u0026mdash;also sees enhancement through the use of LDPC coding, as evidenced in Figs.\u0026nbsp;5 to 7. LDPC codes exhibit higher spectral efficiency across different modulation schemes (4-QAM, 16-QAM, and 256-QAM) within specific SNR ranges. For instance, with 4-QAM modulation, LDPC codes surpass polar codes between \u0026minus;\u0026thinsp;3 dB and 0 dB SNR, and this advantage continues with higher modulation schemes up to 10 dB SNR. Such results underscore the effectiveness of LDPC codes for systems that demand high spectral efficiency, particularly in 5G networks utilizing 4\u0026times;16 MIMO configurations. However, despite the clear benefits, LDPC decoding comes with increased computational complexity when compared to polar codes, potentially leading to higher system latency and energy consumption in large-scale 5G implementations. Additionally, as SNR levels rise, both coding techniques tend to reach a performance plateau, diminishing the relative advantages of LDPC codes. It's also important to note that the simulations are based on ideal conditions, which may not fully represent the real-world challenges posed by interference and hardware limitations. In conclusion, while LDPC coding demonstrates considerable performance enhancements in 5G MIMO systems, particularly at low to moderate SNR levels, the associated complexity and practical application challenges must be thoroughly evaluated.\u003c/p\u003e"},{"header":"8. Conclusion","content":"\u003cp\u003eThis study highlights the effectiveness of incorporating advanced communication techniques, particularly a MIMO system that employs deep neural network (DNN) demappers alongside LDPC and polar coding, in tackling the challenges associated with 5G technology. By concentrating on the enhancement of bit error rate (BER) and spectral efficiency across various orders of Quadrature Amplitude Modulation (QAM), our proposed system demonstrates substantial performance improvements, with LDPC coding notably surpassing polar coding. The findings reveal significant advancements in both reliability and data throughput, positioning the system as highly suitable for the rigorous demands of 5G applications. This research sets a solid foundation for future innovations in robust communication systems capable of supporting a wide array of uses, ranging from improved mobile broadband to extensive IoT connectivity, ultimately contributing to the evolution of next-generation wireless technology.\u003c/p\u003e "},{"header":"Declarations","content":"\u003ch2\u003e\u003cstrong\u003eConflict of Interests Declaration\u003c/strong\u003e\u003c/h2\u003e\n\u003cp\u003eThe author has stated that they do not possess any potential conflicts of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDisclosure of Funding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research did not receive any dedicated financial support from any funding agency.\u003c/p\u003e\n\n\u003ch2\u003e\u003cstrong\u003eCode availability\u003c/strong\u003e\u003c/h2\u003e\n\u003cp\u003eThe code is available from the corresponding author by request.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eAuthor 1 contributed to generating ideas and implementing them. Author 2 assisted in evaluating results and analyzing findings. Author 3 helped with the literature review, article writing, proofreading, and verifying the article's content.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.In our study, data sharing is not restricted, as we used randomly generated data to train our deep neural network (DNN) demappers. Since no real participant data was involved, there are no privacy concerns, allowing us to share the data and results openly. Figures 2 through 7 display results from the trained DNN based on different parameter values, and Figure 1 illustrates our proposed model.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eZhang J, et al. A Review of 5G Wireless Communication Systems. IEEE Access. 2020; 8: 27042\u0026ndash;27053.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAndrews J,et al.What Will 5G Be? IEEE Journal on Selected Areas in Communications. 2014; 32(6): 1065\u0026ndash;1082.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFoschini GJ, et al.On Limits of Wireless Communications in a Fading Environment When Using Multiple Antennas. Wireless Personal Communications. 1998; 6(3): 311\u0026ndash;335.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarzetta TL,et al.Noncooperative Cellular Wireless with Unlimited Numbers of Base Station Antennas. IEEE Transactions on Wireless Communications.2016; 9(11): 3590\u0026ndash;3600.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eO'Sheaet al.Deep learning based MIMO communications.2017; arXiv preprint arXiv:1707.07980.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGallageret al.Low-density parity-check codes. IRE Transactions on information theory. 1960; 8(1):21\u0026ndash;28.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArikan E,et al. Channel Polarization: A Method for Constructing Capacity-Achieving Codes for Symmetric Binary-Input Memoryless Channels. IEEE Transactions on Information Theory. 2019; 55(7): 3051\u0026ndash;3073.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShashanket al. Advanced Decoding Methods for Massive-MIMO Systems Employing Deep Learning. 2024. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1109/icaect60202.2024.10468822\u003c/span\u003e\u003cspan address=\"10.1109/icaect60202.2024.10468822\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWanget al. A Unified Deep Learning Based Polar-LDPC Decoder for 5G Communication Systems. 2018. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1109/WCSP.2018.8555891\u003c/span\u003e\u003cspan address=\"10.1109/WCSP.2018.8555891\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVietet al. Performance of Deep Learning LDPC Coded Communications in Large Scale MIMO Channels. 2019.doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1109/NICS48868.2019.9023820\u003c/span\u003e\u003cspan address=\"10.1109/NICS48868.2019.9023820\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeenalakshmiM, et al.Enhancing channel estimation accuracy in polar-coded MIMO-OFDM systems via CNN with 5G channel models.2023. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.aeue.2023.155016\u003c/span\u003e\u003cspan address=\"10.1016/j.aeue.2023.155016\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVictor C, et al. Multi-input fully CNN for joint pilot decontamination and symbol detection in 5G massive MIMO. Iet Communications. 2023; 17:1899\u0026ndash;1906. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1049/cmu2.12670\u003c/span\u003e\u003cspan address=\"10.1049/cmu2.12670\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHeydar A, et al.Massive MIMO Data Detection Using Matrix Inversion Method and Deep Neural Network. 2023; 165\u0026ndash;169. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1109/apwimob59963.2023.10365630\u003c/span\u003e\u003cspan address=\"10.1109/apwimob59963.2023.10365630\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTianyuet al.A Deep Learning based Multi-edge-type decoding algorithm for 5G NR LDPC codes. 2023; 224\u0026ndash;228. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1109/IWCMC58020.2023.10182649\u003c/span\u003e\u003cspan address=\"10.1109/IWCMC58020.2023.10182649\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKirillet al.Deep Neural Network Based Decoding of Short 5G LDPC Codes. 2021. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1109/REDUNDANCY52534.2021.9606471\u003c/span\u003e\u003cspan address=\"10.1109/REDUNDANCY52534.2021.9606471\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSivaramaet al.CNN-Based Approach for Enhancing 5G LDPC Code Decoding Performance. IEEE Access. 2024;1\u0026ndash;1. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1109/access.2024.3420106\u003c/span\u003e\u003cspan address=\"10.1109/access.2024.3420106\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHala F, et al.Ber performance of joint polar coded signal and mimo using millimeter wave. 2021; 4(1):10\u0026ndash;18. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.31987/IJICT.4.1.124\u003c/span\u003e\u003cspan address=\"10.31987/IJICT.4.1.124\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMohammed HA, et al.Performance Evaluation of 5g New Radio Ldpc Over Different Scenarios of Lognormal Fading Channel. 2024. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2139/ssrn.4718427\u003c/span\u003e\u003cspan address=\"10.2139/ssrn.4718427\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePutri SW, et al.Bit Error Rate Performance Comparison of Low-Density Parity-Check Code and Polar Code. 2023; 1\u0026ndash;6. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1109/ict60153.2023.10374061\u003c/span\u003e\u003cspan address=\"10.1109/ict60153.2023.10374061\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSalima B, et al.On error correction performance of LDPC and Polar codes for the 5G Machine Type Communications. 2021; 1\u0026ndash;4. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1109/IEMTRONICS52119.2021.9422665\u003c/span\u003e\u003cspan address=\"10.1109/IEMTRONICS52119.2021.9422665\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJoyet al.5G Systems with Low Density Parity Check based Chanel Coding for Enhanced Mobile Broadband Scheme. 2020;2(1):42\u0026ndash;49. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.36548/JSWS.2020.1.005\u003c/span\u003e\u003cspan address=\"10.36548/JSWS.2020.1.005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuntur KP,et al.Evaluation of LDPC Code and Polar Code Coding Scheme in 5G Technology \u0026ndash; Massive Machine Type Communication. 2018; 170\u0026ndash;174. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1109/ICITEED.2018.8534937\u003c/span\u003e\u003cspan address=\"10.1109/ICITEED.2018.8534937\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNguyenet al. The Effectiveness of LDPC Decoding Algorithms in 5G Channel Modelling of MIMO-OFDM System under the Influence of Spatial Correlation. JST: Smart Systems and Devices. 2022;32(2):6\u0026ndash;13.doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.51316/jst.158.ssad.2022.32.2.2\u003c/span\u003e\u003cspan address=\"10.51316/jst.158.ssad.2022.32.2.2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePooja M, et al. Performance Analysis of Polar codes for Next Generation 5G Technology. 2022;1\u0026ndash;4. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1109/incet54531.2022.9824746\u003c/span\u003e\u003cspan address=\"10.1109/incet54531.2022.9824746\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAizazul H et al.Addressing tactic volatility in self-adaptive systems using evolved recurrent neural networks and uncertainty reduction tactics. 2022. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1145/3512290.3528745\u003c/span\u003e\u003cspan address=\"10.1145/3512290.3528745\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSupreeth BR,et al.Autonomous Heterogeneous Network for Mixed Strategy Analysis in Defence Attack Game Model Based on 5G in Machine Learning Environment. 5th International Conference on Contemporary Computing and Informatics (IC3I). 2022;2029\u0026ndash;2033. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1109/ic3i56241.2022.10072526\u003c/span\u003e\u003cspan address=\"10.1109/ic3i56241.2022.10072526\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMebratu A. Selection of Architectural Patterns based on Tactics. 2022;13\u0026ndash;18. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1109/ICT4DA56482.2022.9971369\u003c/span\u003e\u003cspan address=\"10.1109/ICT4DA56482.2022.9971369\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOsseiran A, et al. 5G: The Next Generation of Mobile Communication. IEEE Communications Magazine. 2014;52(2):38\u0026ndash;45.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBertocco J, et al.Deep Learning for Wireless Communications: A Review. IEEE Communications Surveys \u0026amp; Tutorials. 2019;21(3):1446\u0026ndash;1468.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang C, et al. Machine Learning for Signal Detection in 5G Wireless Networks. IEEE Journal on Selected Areas in Communications.2020;38(2):245\u0026ndash;258\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang Z, et al. Performance Analysis of Deep Neural Network Demappers for 5G Wireless Systems. IEEE Transactions on Wireless Communications. 2019;18(12):5753\u0026ndash;5767.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Y, et al.Spectral Efficiency Improvement for 5G Systems Using Deep Learning Techniques. Journal of Communications and Networks. 2021;23(1):24\u0026ndash;35.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYuan J, et al.Challenges and Opportunities in Explainable Artificial Intelligence for Wireless Communications. IEEE Wireless Communications. 2022;29(1):56\u0026ndash;62.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRichardson T, et al.The Capacity of Low-Density Parity-Check Codes Under Message-Passing Decoding. IEEE Transactions on Information Theory. 2001;47(2):599\u0026ndash;618.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMacKay DJC. Information Theory, Inference, and Learning Algorithms.Cambridge University Press. 2003.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFossorier M. Decoding Algorithms for Non-Binary Low-Density Parity-Check Codes. IEEE Transactions on Information Theory. 1995;41(2):224\u0026ndash;234.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShannon CE. A Mathematical Theory of Communication. Bell System Technical Journal. 1948;27(3): 379\u0026ndash;423.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChung SY, et al.On the Design of LDPC Codes.\" IEEE Transactions on Information Theory. 2001;47(7):2460\u0026ndash;2470.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArıkan E. Channel Polarization: A Method for Constructing Capacity-Achieving Codes for Symmetric Binary-Input Memoryless Channels. IEEE Transactions on Information Theory. 2009;55(7):3051\u0026ndash;3073.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTal I, et al.List Decoding of Polar Codes. IEEE Transactions on Information Theory. 2013;59(11):6218\u0026ndash;6228.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen X, et al. Polar Codes for 5G Wireless Communication Systems. IEEE Access. 2019;7: 103627\u0026ndash;103641.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHaykin S. Communication Systems. John Wiley \u0026amp; Sons. 2008.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eProakis JG, et al.Digital Communications. McGraw-Hill. 2007.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVan Trees HL. Detection, Estimation, and Modulation Theory, Part I. Wiley-Interscience. 2002.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGarg VK. Wireless Communication and Networking. Morgan Kaufmann. 2010.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHaykinS. Adaptive filter theory. Pearson Education. 2001.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHaykin S. Adaptive Filter Theory. 4th ed. Prentice Hall. 2002.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVan de Beek JJ,et al.On the Search for the Optimal Linear Equalizer. IEEE Transactions on Signal Processing. 1997;45(11): 2850\u0026ndash;2861.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRappaport TS. Wireless Communications: Principles and Practice. 2nd ed. Prentice Hall. 2002.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoldsmith A. Wireless Communications. Cambridge University Press. 2005.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eProakis JG et al.Digital Communications. 5th ed. McGraw-Hill. 2008.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"discover-electronics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Discover Electronics](https://www.springer.com/journal/44291)","snPcode":"44291","submissionUrl":"https://submission.nature.com/new-submission/44291","title":"Discover Electronics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"LDPC code, Polar code, DNN demapper, 5G MIMO, BER, Spectral efficiency","lastPublishedDoi":"10.21203/rs.3.rs-5302008/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5302008/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe emergence of 5G technology necessitates advanced communication systems that can handle high data rates while providing improved reliability. However, current methods often struggle with issues related to complexity and efficiency. This paper addresses these challenges by developing a Multiple Input Multiple Output (MIMO) system that integrates deep neural network (DNN) demappers, Low-Density Parity-Check (LDPC) coding, and polar coding, creating a mix of tactics aimed at boosting performance in flat fading channel environments. The main objectives of this study are to improve the bit error rate (BER) and spectral efficiency across different orders of Quadrature Amplitude Modulation (QAM)\u0026mdash;specifically, 4-QAM, 16-QAM, and 256-QAM. The proposed system employs a Linear Minimum Mean Square Error (LMMSE) equalizer to ensure effective signal detection. Simulation results indicate that the LDPC coding technique significantly outperforms polar coding across all evaluated QAM modulation orders, highlighting its effectiveness in enhancing system performance. An analysis of BER and spectral efficiency demonstrates considerable reliability and data throughput improvements, making the MIMO system particularly well-suited for the demanding requirements of 5G applications. The findings suggest that integrating advanced coding strategies with neural demapping can effectively tackle the complexities introduced by the 5G environment. This research establishes a foundation for future studies and the development of robust communication systems capable of supporting the next generation of wireless technology, thereby enabling a wide range of applications, from enhanced mobile broadband to extensive IoT connectivity.\u003c/p\u003e","manuscriptTitle":"A DNN-based 5G MIMO System Adopting a Mix of Tactics","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-11-18 08:12:55","doi":"10.21203/rs.3.rs-5302008/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-12-09T10:15:38+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-11-29T12:11:11+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-11-23T10:27:22+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"252367108194025013936915677257749902102","date":"2024-11-20T10:25:29+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"31778217564329030476011656407803433343","date":"2024-11-19T03:34:46+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"49513747893954845077667555485982933473","date":"2024-11-19T02:16:26+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"186860760317017772791117808446281111409","date":"2024-11-18T05:59:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"172120841752934650367917661038005347265","date":"2024-11-18T05:57:49+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-11-18T05:52:15+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-11-06T19:10:04+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-11-05T07:06:56+00:00","index":"","fulltext":""},{"type":"submitted","content":"Discover Electronics","date":"2024-10-21T07:11:35+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"discover-electronics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Discover Electronics](https://www.springer.com/journal/44291)","snPcode":"44291","submissionUrl":"https://submission.nature.com/new-submission/44291","title":"Discover Electronics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"85fdcf01-d567-45c0-b2ea-b7fa0bb0d105","owner":[],"postedDate":"November 18th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-03-17T13:08:26+00:00","versionOfRecord":[],"versionCreatedAt":"2024-11-18 08:12:55","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5302008","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5302008","identity":"rs-5302008","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}