Internet of Things (IoT) devices and edge-platform robots require privacy-aware path planning under strict compute and bandwidth constraints, where continual on-device learning from real-world data is desirable but costly. We propose and evaluateSplit Federated Learning (SFL)–enabledDeep Q-Networks (DQN), termedSFL-DQN, which partitions the Q-network into a lightweight on-device encoder and a shared server-side action–value head with online/target stabilization. Devices keep raw observations local and exchange only intermediate feature representations and their gradients; periodicFederated Averaging (FedAvg)realigns client encoders to mitigate representation drift under heterogeneous, non-independent and identically distributed (non-IID) experience. In grid-world navigation withonemoving obstacle, SFL-DQN matches a non-federated, non-split DQN baseline in episodic return and task success and attainsBreadth-First Search (BFS)–equivalent shortest-path optimality on held-out maps. WithKclients, each device uses only 1/Kof the single-device interaction budget (e.g., 1,000 vs. 4,000 episodes whenK=4), reducing theclient-sidetraining burden by offloading the value-learning head to the server. Split–federated training introduces feature-level communication overhead and does not necessarily reduce system-wide energy under all network conditions; instead, SFL-DQN enables a practical edge–cloud learning loop without raw-data aggregation.

Millimeter-wave (mmWave) massive multiple-input multiple-output (MIMO) systems have been considered a promising solution to provide high-quality data services. To achieve full beamforming gain, the acquisition of accurate channel information is crucial for the success of massive MIMO systems. In recent years, numerous approaches have been suggested to acquire the downlink channel state information (CSI). However, a significant mismatch between the estimated channel and the actual channel for data transmission causes outdated CSI, leading to a severe degradation in spectral efficiency. In this paper, we propose a channel estimation technique for mmWave massive MIMO systems that obtains multipath components of the downlink channel from the previous channel sequence. To be specific, the proposed technique learns the spatio-temporal correlation be-tween multipath components by exploiting a Transformer-based framework. From the numerical results, we demonstrate that the proposed technique outperforms the conventional channel acquisition techniques in terms of normalized mean square error (NMSE).

As a key technology to meet the ever-increasing data rate demand in beyond 5G and 6G communications, millimeter-wave (mmWave) massive multiple-input multiple-output (MIMO) systems have gained much attention recently. To make the most of mmWave massive MIMO systems, acquisition of accurate channel state information (CSI) at the base station (BS) is crucial. However, this task is by no means easy due to the CSI feedback overhead induced by the large number of antennas. In this paper, we propose a parametric CSI feedback technique for mmWave massive MIMO systems. Key idea of the proposed technique is to compress the mmWave MIMO channel matrix into a few geometric channel parameters (e.g., angles, delays, and path gains). Due to the limited scattering of mmWave signal, the number of channel parameters is much smaller than the number of antennas, thereby reducing the CSI feedback overhead significantly. Moreover, by exploiting the deep learning (DL) technique for the channel parameter extraction and the MIMO channel reconstruction, we can effectively suppress the channel quantization error. From the numerical results, we demonstrate that the proposed technique outperforms the conventional CSI feedback techniques in terms of normalized mean square error (NMSE) and bit error rate (BER).

Faster-Than-Nyquist (FTN) is a technique that can improve the spectral efficiency of communication systems by making better use of available spectrum resources at the cost of inter-symbol interference (ISI) and inter-carrier interference (ICI). In this paper, we propose a hybrid signaling scheme for a practical application of FTN for MIMO transmission. We propose a new slot structure optimized for the hybrid signaling supporting both FTN signaling and orthogonal frequency division multiplexing (OFDM) signaling. Specifically, in the proposed slot structure, data transmission is based on the FTN signaling and the pilot transmission is based on the OFDM signaling. Numerical results confirm that the proposed signaling scheme has clear benefit over the systems employing only OFDM or FTN signaling.

Fifth generation wireless standards require much lower latency than what current wireless systems can guarantee. The main challenge in fulfilling these requirements is the development of short packet transmission, in contrast to most of the current standards, which use a long data packet structure. Since the available training resources are limited by the packet size, reliable channel and interference covariance estimation with reduced training overhead are crucial to any system using short data packets. In this paper, we propose an efficient receiver that exploits useful information available in the data transmission period to enhance the reliability of the short packet transmission. In the proposed method, the receive filter (i.e., the sample covariance matrix) is estimated using the received samples from the data transmission without using an interference training period. A channel estimation algorithm to use the most reliable data symbols as virtual pilots is employed to improve quality of the channel estimate. Simulation results verify that the proposed receiver algorithms enhance the reception quality of the short packet transmission.

5G wireless standards require a much lower latency than what current wireless systems can guarantee. The main challenge to fulfill this requirement is the capability to support short packet transmission, in contrast to most of the current standards which use a long data packet structure. In this paper, we propose an efficient receiver technique that exploits information obtained during the data transmission period to improve the reception quality of the short packet transmission. Two key ingredients of the proposed method are 1) estimation of the receiver filter using the received samples in the data transmission period, not in the interference training period, and 2) soft decision- directed channel estimation that uses the data symbols for re-estimation of the channels. Numerical results confirm the effectiveness of the proposed receiver algorithms.

Acquiring reliable channel state information (CSI) at the basestation is crucial to fully exploit the advantages of massive multiple-input multiple-output (MIMO) systems. However, giving the basestation knowledge of the downlink channel is an important and challenging problem in frequency division duplexing (FDD) systems when the number of basestation antennas is large. In this paper, we propose an eigenvalue decomposition based feedback compression (EFC) framework that transforms the channel information for pursuing further reduction in feedback overhead. The main idea of EFC is to use the compressed channels for generating short-term CSI by exploiting the dominant eigendirections of long-term channel-statistical information. By using the proposed codebook, each user terminal sends the compressed channel information as short-term feedback and channel-statistical information as long-term feedback. Numerical results demonstrate that the proposed method achieves significant feedback overhead reduction over the conventional beamforming techniques under the same target sum rate requirement.

LTE-Advanced system has been deployed with 2 and 4 transmission antennas (Tx) while the specification supports up to 8Tx. Due to deployment space, antenna dimension and complexity, operators have not been interested in the deployment of 8Tx systems. Recently, three dimensional (3D) beamforming using 2D active antenna array has attracted significant attention in the wireless industry. By incorporating 2D active array into LTE-A systems, the system offers freedom in controlling radiation on elevation and horizontal dimension. In addition, 2D array antenna increases the number of antennas without exceeding form-factor where the conventional antennas are deployed. When the number of antennas increases in the form of 2D arrangement, spatial separation can be realized simultaneously in horizontal and elevation domain and vertical beam-steering can increase SINR of UEs in high floors. In this paper, we study the system operations and implementations for supporting 3D beamforming with 8Tx antennas. In our schemes, by reusing the conventional CSI feedback framework, the system can operate 2D active array without harming the backward compatibility. Evaluation results show that 3D beamforming provides capacity boosting over the conventional 2D beamforming systems while keeping same antenna structure.

Recent works on massive multiple-input multiple-output (MIMO) have shown that a potential breakthrough in capacity gains can be achieved by deploying a very large number of antennas at the base station. In order to achieve the performance that massive MIMO systems promise, accurate transmit-side channel state information (CSI) should be available at the base station. While transmit-side CSI can be obtained by employing channel reciprocity in time division duplexing (TDD) systems, explicit feedback of CSI from the user terminal to the base station is needed for frequency division duplexing (FDD) systems. In this paper, we propose an antenna grouping based feedback reduction technique for FDD-based massive MIMO systems. The proposed algorithm, dubbed antenna group beamforming (AGB), maps multiple correlated antenna elements to a single representative value using predesigned patterns. The proposed method modifies the feedback packet by introducing the concept of a header to select a suitable group pattern and a payload to quantize the reduced dimension channel vector. Simulation results show that the proposed method achieves significant feedback overhead reduction over conventional approach performing the vector quantization of whole channel vector under the same target sum rate requirement.

Accurate transmit-side channel state information (CSI) is essential for achieving optimal performance of large-scale multiple-input multiple-output (MIMO) systems. However, due to high dimensionality of the massive MIMO systems, large overhead of the uplink CSI feedback is a big concern for frequency division duplexing (FDD) systems. In this paper, we propose antenna group scheduling (AGS) algorithm that combines antenna selection and user scheduling for pursuing further reduction in feedback overhead. The AGS method selects the antenna group with the maximum channel gain to reduce the dimension of channel vector and schedules users in the group maximizing the sum rate. Simulation results demonstrate that the proposed technique achieves significant feedback overhead reduction over the conventional beamforming techniques (33%~50%) under the same target sum rate requirement.

Recent works on massive multiple-input multiple-output (MIMO) have shown that a potential breakthrough in capacity gains can be achieved by deploying a very large number of antennas at the basestation. Although transmit-side channel state information (CSI) can be obtained by employing channel reciprocity in time division duplexing (TDD) systems, explicit feedback of CSI from the user to the basestation is required for frequency division duplexing (FDD) systems. In this paper, we propose an antenna grouping based feedback reduction technique for FDD-based massive MIMO systems. The proposed algorithm, dubbed antenna group beamforming (AGB), groups antenna elements using pre-designed patterns. The proposed method introduces the concept of using a header of overall feedback resources to select a suitable group pattern and the payload to quantize the effective channel vector. Simulation results show that the proposed method achieves significant feedback overhead reduction over conventional approach.

Large-scale multiple-input multiple-output(MIMO) systems with a large number of antennas at the basestation have drawn considerable interest because of potential ability to achieve high spectral efficiencies. In order to achieve optimal performance of large-scale MIMO systems, the basestation needs to know channel state information (CSI) perfectly. In terms of CSI acquisition, the basestation estimates the downlink channel through channel reciprocity in time division duplexing (TDD) or requires CSI feedback through the uplink in frequency division duplexing (FDD). Due to the large number of transmit antennas at the basestation, uplink CSI feedback would be a major hurdle in developing FDD large-scale MIMO systems. In this paper, we propose an efficient feedback compression technique for FDD large-scale MIMO systems. The proposed method reduces a dimension of vector quantization by grouping high correlated antenna elements. In fact, the proposed method invests a small portion of feedback resources to generate a grouped channel vector and the rest to quantize the grouped channel vector. Simulation results demonstrate that the proposed method achieves significant feedback overhead reduction over conventional methods.

Recent results show that non-parametric linear receiver equipped with multiple receive antennas is an effective solution for ad hoc network under the imperfect channel state information at receiver (CSIR). In this paper, we propose an interference aware node activation strategy that efficiently controls the data transmission and back off based on the measured interference power. Our numerical results show that if optimization over the channel estimation threshold is provided, then the effective data rate of the proposed scheme is higher than that of the conventional channel estimation even under the system delay occurred.

In this paper, we present a novel soft-input soft-output detector for large-scale MIMO systems that offers substantial complexity reduction over the existing detectors and performance close to the full dimensional symbol detector. The proposed algorithm, termed soft-input soft-output successive group (SSG) detector, detects a subset of symbols successively with an aid of the preprocessing designed to suppress the inter-group interference. In fact, the proposed preprocessor mitigates the effect of interfering symbol groups using a priori information of the undetected groups and a posteriori information of the detected groups. Simulation results performed on large-scale multi-input multi-output (MIMO) systems demonstrate that the proposed SSG detector achieves significant complexity reduction over the conventional approaches with negligible performance loss.

Recently, a variety of low complexity soft-input soft-output detection algorithms have been introduced for iterative detection and decoding (IDD) systems. However, it is still challenging to implement soft-input soft-output detector detector at feasible complexity for massive MIMO systems due to the heavy burden in computational complexity. In this paper, we present a novel soft-input soft-output detector for massive MIMO systems that offers substantial complexity reduction over the existing detectors performance close to the full dimensional symbol detector. The proposed soft-input soft-output successive group (SSG) detector detects a subset of symbols successively with an aid of the deliberately designed preprocessing to suppress the inter-group interference. Simulation results performed on 8×8 MIMO flat fading channels demonstrate that the proposed SSG detector achieves significant complexity reduction over the conventional approaches with negligible performance loss.

Recent works on ad hoc network study have shown that achievable throughput can be made to scale linearly with the number of receive antennas even if the transmitter has only a single antenna. In this paper, we propose a non-parametric linear minimum mean square error (MMSE) receiver for achieving further gain in performance when the channel state information at receiver (CSIR) of interferers is imperfect. The key feature to make our approach effective is to exploit the autocorrelation of the received signal. In fact, by incorporating the desired channel information on top of the observations including interference and noise only, the proposed method achieves large fraction of the optimal MMSE transmission capacity without transmission rate loss. Simulation results on the realistic ad hoc network system show that the proposed non-parametric linear MMSE receiver brings substantial performance gain over existing multiple receive antenna algorithms.

Recent works on multiuser MIMO study have shown that the linear growth of capacity in single user MIMO system can be translated to the multiuser MIMO scenario as well. In this paper, we propose a method pursuing performance gain of vector perturbation in multiuser downlink systems. Instead of employing the maximum number of mobile users for communication, we use small part of them as virtual users for improving reliability of users participating communication. By controlling parameters of the virtual users including information and perturbation vector, we obtain considerable improvement in the effective SNR. Simulation results on the realistic multiuser MISO and MIMO downlink systems show that the proposed method brings substantial performance gain over the standard vector perturbation with marginal overhead in computations.

Block diagonalization (BD) algorithm is a generalization of the channel inversion that converts multiuser multi-input multi-output (MIMO) broadcast channel into single-user MIMO channel without inter-user interference. In this paper, we combine the BD technique with a minimum mean square error vector precoding (MMSE-VP) for achieving further gain in performance with minimal computational overhead. Two key ingredients to make our approach effective are the QR decomposition based block diagonalization and joint optimization of transmitter and receiver parameters in the MMSE sense. In fact, by optimizing precoded signal vector and perturbation vector in the transmitter and receiver jointly, we pursue an optimal balance between the residual interference mitigation and the noise enhancement suppression. From the sum rate analysis as well as the bit error rate simulations (both uncoded and coded cases) in realistic multiuser MIMO downlink, we show that the proposed BD-MVP brings substantial performance gain over existing multiuser MIMO algorithms.

Block Diagonalization (BD) algorithm is a generalization of channel inversion scheme in multiuser multi-input multi-output (MIMO) broadcast channels. Although the BD algorithm is effective in removing interuser interference, it has a drawback that additional information should be delivered to the receiver. Recent work on BD with vector perturbation (VP) avoids the need for additional information and hence reduces receiver complexity. In this paper, we propose a method achieving further gain in the BD for multiuser MIMO downlink. By combining the BD and minimum mean square error vector precoding (MMSEVP), we pursue the balance between the interference suppression and the noise enhancement control, resulting in considerable improvement in the effective SINR. In fact, simulation results on the realistic multiuser downlink scenario show that the proposed method brings substantial performance gain over existing multiuser MIMO algorithms (BD, BD-VP and BD-WF).

In this paper, we put forth an approach pursuing performance gain of vector perturbation in multiuser MIMO downlink. Instead of achieving maximal sum capacity by employing maximum number of mobile users for communication, we sacrifice small part of them for improving the performance of ones participating communication. By controlling parameters of these sacrificing users, we achieve considerable improvement in the effective SNR, which can be translated to the large gain in bit error performance. From simulation results on 4×4 and 10×10 multiuser MIMO systems, we show that the proposed method brings substantial performance gain over the standard vector perturbation.