CSI-Based Multi-Antenna and Multi-Point Indoor Positioning Using Probability Fusion

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

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

Due to its high-dimensional data characteristics, the channel state information (CSI) of Wi-Fi signals has become a strong candidate for use in indoor localization. In addition, machine learning techniques can improve the accuracy of indoor localization systems using multiview CSI data received at multiple access points (APs). However, in complex environments, most CSI views collected at APs in non-line-of-sight (NLoS) configurations relative to a transmitter may lose so much useful data information as to become nonsalient. In this paper, we propose a practical machine learning approach named unsupervised view-selective deep learning (UVSDL), in which only the most salient CSI view is selected in an unsupervised manner to be applied in regression for localization. In an experiment in a complex building, our variational deep learning (VDL)-based regression method with the most salient CSI view achieves a localization accuracy of 1.36 m, significantly outperforming the best-known system BiLoc by 25 %.

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

In centralized massive multiple-input multiple-output (MIMO) systems, the channel hardening phenomenon can occur, in which the channel behaves as almost fully deterministic as the number of antennas increases. Nevertheless, in a cell-free massive MIMO system, the channel is less deterministic. In this paper, we propose using instantaneous channel state information (CSI) instead of statistical CSI to obtain the power control coefficient in cell-free massive MIMO. Access points (APs) and user equipment (UE) have sufficient time to obtain instantaneous CSI in a slowly time-varying channel environment. We derive the achievable downlink rate under instantaneous CSI for frequency division duplex (FDD) cell-free massive MIMO systems and apply the results to the power control coefficients. For FDD systems, quantized channel coefficients are proposed to reduce feedback overhead. The simulation results show that the spectral efficiency performance when using instantaneous CSI is approximately three times higher than that achieved using statistical CSI.

Channel state information (CSI) acquisition is one of the major challenges for massive MIMO systems to enable high efficiency MIMO transmissions. Limited feedback schemes, which quantize CSI into limited number of bits and feed these bits back to the base station (BS), are widely used in real-life mobile networks for BSs to obtain downlink CSI. Practical limited feedback schemes usually have large feedback granularity and introduce severe nonlinear quantization error due to the constraint on feedback overhead. It is hard for legacy signal processing technologies to reconstruct CSI accurately for high efficiency downlink transmissions. Different from the legacy signal processing technologies, deep neural networks trained with big data have strong abilities to fit any nonlinear functions and learn the complex signal model behind the big data, which is a useful tool to solve problems that are nonlinear or hard to be described mathematically. Motivated by these observations, we consider a deep learning based approach for BS receivers to reconstruct CSI based on the limited feedback bits from user equipment (UE). The CSI reconstruction is similar with the image super-resolution which reconstructs a high resolution image from a low resolution one based on the structure information learned by the neural network itself. Performance evaluations demonstrate that the proposed method outperforms the legacy algorithms. The improvement on spectrum efficiency is about 25%. It can also reconstruct the CSI with lower density reference signal (RS) than that in current specification, e.g., the RS density can be 1/32 of current one. Therefore, the reference signal and feedback overhead can be reduced with our method. These results show the potential gain of applying deep learning to physical layer signal processing at the receiver side only, which has little air-interface impact and does not require much effort on enhancing network specifications. Therefore, it is feasible to be deployed in current 5th generation mobile networks with the pace of network evolutions, and opens the gate of intelligent physical layer design for future generations of mobile networks.

Radio channel state information (CSI) measured with many receivers is a good resource for localizing a transmitter device with machine learning with a discriminative model. However, CSI localization is nontrivial when the radio map is complicated, such as in building corridors. This article introduces a view-selective deep learning (VSDL) system for indoor localization using CSI of WiFi. The multiview training with CSI obtained from multiple groups of access points (APs) generates latent features on a supervised variational deep network. This information is then applied to an additional network for dominant view classification to minimize the regression loss of localization. As noninformative latent features from multiple views are rejected, we can achieve a localization accuracy of 1.28 m, which outperforms by 30% the best known accuracy in practical applications in a complex building environment. To the best of our knowledge, this is the first approach to apply variational inference and to construct a practical system for radio localization. Furthermore, our work investigates a methodology for supervised learning with multiview data where informative and noninformative views coexist.

Cell-free massive MIMO (CF-mMIMO) provides wireless connectivity for a large number of user equipments (UEs) using access points (APs) distributed across a wide area with high spectral efficiency (SE). The energy efficiency (EE) of the uplink is determined by (i) the transmit power control (TPC) algorithms, (ii) the numbers, configurations, and locations of the APs and the UEs, and (iii) the propagation channels between the APs and the UEs. This paper investigates all three aspects, based on extensive (~30,000 possible AP locations and 128 possible UE locations) channel measurement data at 3.5 GHz. We compare three different TPC algorithms, namely maximization of transmit power (max-power), maximization of minimum SE (max-min SE), and maximization of minimum EE (max-min EE) while guaranteeing a target SE. We also compare various antenna arrangements including fully-distributed and semi-distributed systems, where APs can be located on a regular grid or randomly, and the UEs can be placed in clusters or far apart. Overall, we show that the max-min EE TPC is highly effective in improving the uplink EE, especially when no UE within a set of served UEs is in a bad channel condition and when the BS antennas are fully-distributed.

This paper investigates the performance of downlink simultaneous transmitting and reflecting reconfigurable intelligent surface (STAR-RIS)-assisted cell-free (CF) massive multiple-input multiple-output (mMIMO) systems, where user equipments (UEs) are located on both sides of the RIS. We account for correlated Rayleigh fading and multiple antennas per access point (AP), while the maximum ratio (MR) beamforming is applied for the design of the active beamforming in terms of instantaneous channel state information (CSI). Firstly, we rely on an aggregated channel estimation approach that reduces the overhead required for channel estimation while providing sufficient information for data processing. We obtain the normalized mean square error (NMSE) of the channel estimate per AP, and design the passive beamforming (PB) of the surface based on the long-time statistical CSI. Next, we derive the received signal in the asymptotic regime of numbers of APs and surface elements. Then, we obtain a closed-form expression of the downlink achievable rate for arbitrary numbers of APs and STAR-RIS elements under statistical CSI. Finally, based on the derived expressions, the numerical results show the feasibility and the advantages of deploying a STAR-RIS into conventional CF mMIMO systems. In particular, we theoretically analyze the properties of STAR-RIS-assisted CF mMIMO systems and reveal explicit insights in terms of the impact of channel correlation, the number of surface elements, and the pilot contamination on the achievable rate.

Implementing intelligent reflecting surfaces (IRSs), in high frequency based beyond 5G networks, has become a necessity to overcome the harsh blockage issues that exist in these bands. IRSs can supply user equipment (UEs) with multi alternative virtual line of sight (LOS) links, hence enhancing the spectral efficiency (SE) of the system. As a result of deploying multi IRSs as communication assistants, the step of IRSs-UEs association is required to optimally assign each UE to its best IRS; consideration of the interference between different links is needed, to maximize the system performance. However, this process will be a time and power consuming problem, if conventional schemes, which exhaustively search all possible association patterns to find the optimum one for communication, is adapted. Although iterative search based schemes can reduce this complexity, they still need feedback signaling in real time. Hence, they will be inefficient in terms of power consumption and delay. Moreover, optimal placement of the multi-IRSs in the network, to enlarge the system performance, is still an open issue and needs to be studied. Consequently, in this paper, to handle the IRSs-UEs association problem, we propose a neural network (NN) based scheme using a multi-IRSs aided multi input multi output (MIMO) system. In this system, the estimated angles of arrival (AoAs) of UEs are used as input features for the NN, which is trained to associate each UE to its best IRS based on this information; then, within each IRS, passive beamforming is performed. Adapting this NN in online mode guarantees obtaining better performance while relaxing the complexity of association and increasing response time, giving a performance comparable to the exhaustive and iterative search based schemes. The proposed NN based scheme determines the association pattern without searching or feedback signals. Moreover, the proposed approach maintains the system SE nearly similar to the optimum performance obtained by the conventional scheme. Secondly, a criterion is suggested for optimal deployment of multi IRSs in the network, depending on maximizing the average summation UEs signal-to-interference-plus-noise ratio (SINR). Numerical results prove that this strategy outperforms a reference one, which aims to guarantee certain performance by maximizing minimum UE SINR. In contrast the proposed strategy achieves better system and per UE spectral efficiency.

Reconfigurable intelligent surface (RIS) is thought to be a potential key technique for future wireless communications due to its ability for manipulating the electromagnetic environment smartly. This paper focuses on the rank and capacity analysis when a RIS is introduced into a multiple-input multiple-output (MIMO) system. By establishing a system model for this communication system, various simulations are conducted for identifying the characteristics of the channel. The simulations of different distance between the access point (AP) and user equipment (UE) show that the condition number of the channel is worsen when the distance increases and the role of the RIS in rank improvement is weaken. The spatial distributions of the reciprocal condition number of the channel which corresponds to the RIS locations are obtained and rank-deficient zones are found in different AP, UE and RIS configurations, which depicts the spatial characteristics of the RIS-assisted MIMO channel. The simulations also indicate that the condition number of the channel not only varies with the RIS location, but is also affected by the antenna array size and orientation of AP and UE. In addition, when the AP has a larger amount of antennas than UE, it is advantageous to place the RIS near the UE rather than the AP to achieve better channel condition. Modulation and coding schemes are applied in the simulations for comparison and capacity improvement is witnessed. Beneficial suggestions for RIS deployment in the MIMO system are concluded according to the simulation results.

Reconfigurable intelligent surface (RIS) partitioning offers a strategic solution for serving multiple users equipment (UEs) simultaneously in blockage-prone wireless environments, leveraging multiple-input multiple-output (MIMO) and non-orthogonal multiple access (NOMA) technologies within the millimeter-wave (mmWave) spectrum. However, deploying RIS partitioning (RISP) in MIMONOMA is hindered by the exacerbated computational complexity involved in acquiring accurate channel state information (CSI). This paper proposes a novel learning framework based on multi-agent deep reinforcement learning (MA-DRL) that maximizes the sum rate of the RISP-aided MIMO-NOMA system without requiring UEs’ CSI. The proposed framework jointly optimizes RIS phase shifts, number of RIS partitions, UEs beamformers, and NOMA power allocation (PA), transforming the challenge into a high-dimensional combinatorial optimization problem. The MA-DRL algorithm integrates double deep Q networks (DDQN) and deep deterministic policy gradient (DDPG) agents, where each UE acts as a DDQN agent optimizing its beamformer, while the RIS serves as a DDPG agent handling partitioning and power control. An experimental testbed is developed to gather real-world data for training and evaluation. Results show that the MA-DRL algorithm closely approaches optimal performance, trailing the exhaustive search by only 8%, while reducing complexity by 95% and improving the sum rate by an average of 18% compared to traditional full RIS setups.

The feasibility of implementing non-orthogonal multiple access (NOMA) in massive multiple-input multiple-output (MIMO) downlink with distributed antenna arrays is investigated. The achievable downlink rates are derived for two channel state information (CSI) cases, namely (i) estimated uplink CSI at the access points (APs) via user uplink pilots and statistical CSI at the users, and (ii) estimated uplink CSI at the APs via uplink user pilots and estimated downlink CSI at the users via downlink pilots beamformed by the APs. The trade-off between the number of served user nodes and their achievable rates is investigated. The adverse effects of imperfect successive interference cancellation (SIC) and detection uncertainty on the achievable downlink rates are quantified. We show that the achievable downlink rate of NOMA can be inferior to that of the orthogonal multiple-access (OMA) counterpart in the low-user regime. This is a direct consequence of residual interference caused by intra-cluster pilot contamination and error propagation from imperfect SIC. However, the proposed massive MIMO NOMA downlink outperforms the OMA counterpart in the highuser regime in terms of the achievable downlink rate. Moreover, the impact of downlink pilots on the sum rate is also investigated.

The ultra-dense network (UDN) is one of the most promising technologies in the fifth generation (5G) to address the network system capacity issue. However, it is a new challenge that the user equipment (UE) secure access UDN composed of the access points (APs) which characterized with autonomy, temporary, and dynamic. In 5G UDN, the APs are independent and equal. The UDN can be regarded as a decentralized access network. Compared with the traditional base station, the AP has a smaller coverage. There has an issue that the interaction between the UE and APs will be more frequent when UE moves. However, the existing 4G authentication and key agreement algorithm cannot adapt to this fast and frequent authentication requirement. If the UE moves smoothly in a trusted APs group (APG) without frequent authentication, this problem will be solved very well. In order to achieve this goal, we propose a security authentication scheme of 5G UDN based on the block chaining technologies. In this paper, an APG-PBFT algorithm based on the block chaining technology with Byzantine fault tolerance (PBFT) consensus algorithm is proposed. In the algorithm, the consensus mechanism will be optimized and a new reverse screening method will be embedded. In our solution, a trusted chain APG can be generated with APs by APG-PBFT algorithm, and the authentication results can be shared in the APG using the blockchain message propagation mechanism. The principle of fast authentication with APG-PBFT algorithm is present in this paper. The scheme can reduce the authentication frequency when UE moves among the APs and improve the access efficiency. Finally, we analyze the performance of APG-PBFT algorithm and compare it with the traditional PBFT algorithm. The simulation results show that the APG-PBFT algorithm can improve the APG generation efficiency and reduce the authentication frequency of UE, which will be valuably applied to the UDN environment.

Non-Orthogonal Multiple Access (NOMA) has been recommended as a feasible multiple access technique that satisfies the constraints of Fifth-generation (5G) communications systems. The upcoming Internet of Things (IoT) based life-enhancing application demands dense connection, maximal throughput, and reduction in latency. The research aims to analyze the Downlink performance of Power Domain NOMA (PD-NOMA) system in terms of Bit Error Rate (BER) under various channel conditions namely AWGN, Rayleigh, Rician and Nakagami-m with 5 User Equipments (UEs) in both Transmitting and Receiving end. The scenario is simulated using MATLAB Version 2021a. As PD-NOMA system allocates UE based on the optimal power of their Channel State Information, the significant enhancement in UE performance is accomplished. From the simulation results, it is found that both the near and far UEs experiences lesser BER under AWGN Channel. However, the same UEs under Rayleigh experiences higher BER with an increase of 40%. The far UEs result with higher BER when compared the near UE as the distance between BS and UE is increased.