Adapting pretrained artificial intelligence and machine learning (AI/ML) models for on-device inference across diverse environments is challenging due to the extensive offline data collection and training required. This highlights the necessity for online model training or adaptation, where devices locally process training data and perform adaptation via on-device engines. Such capabilities and technologies are essential for various wireless AI/ML features requiring on-device inference to be considered in the upcoming Third Generation Partnership Project (3GPP) 6G standardization works. For battery-powered devices, model adaptation must be efficiently executed in real-time with well-designed timelines, as environmental conditions can change rapidly and vary significantly across locations. In this paper, we present results from over-the-air (OTA) prototyping experiments using handset devices to demonstrate the practical effectiveness of real-time on-device adaptation. We use AI/ML based beam prediction, as outlined by 3GPP 5G-Advanced standards, as an example. Our OTA results show that on-device model adaptation can be completed within 30 seconds in an unfamiliar environment to achieve benchmark performance, strongly supporting the feasibility of real-time on-device model adaptation in 6G communications. The findings and methodologies discussed in this paper can be extended to benefit various other wireless and non-wireless real-world scenarios.

As the world enters the journey toward the 6th generation (6G) of wireless technology, the promises of ultra-high data rates, unprecedented low latency, and a massive surge in connected devices require crucial exploration of network energy saving (NES) solutions to minimize the carbon footprint and overall energy usage of future cellular networks. On the other hand, network-controlled repeaters (NCRs) have been introduced by 3rd generation partnership project (3GPP) as a cost-effective solution to improve network coverage. However, their impact on network power consumption and energy efficiency has not been thoroughly investigated. This paper studies NES schemes for next-generation 6G networks aided by NCRs and proposes optimal NES strategies aiming at maximizing the overall energy efficiency of the network. Repeaters are shown to allow for power savings at next-generation nodeB (gNB), and offer higher overall energy efficiency (EE) and spectral efficiency (SE), thus providing an energy-efficient and cost-efficient alternative to increase the performance of future 6G networks.

The paradigm of wireless communications is shifting from data transfer to ubiquitous services. Perception of the surrounding environment has recently been highlighted as one of the key technologies enabling new applications, such as autonomous driving, extended reality, and healthcare. This movement gives a rise to an emerging concept of integrated sensing and communication (ISAC), where the distributed communication nodes, such as the base station and user devices, serve as probes for wide-area sensing, and the wideband radio signals are shared between communication and sensing. In recent years, a lot of effort has been directed towards the two main pillars of ISAC, i.e., self-localization of a node (positioning) and detection of device-free objects around the node (sensing). Nevertheless, only few of them have been implemented and brought up for real-world verification. This paper provides our latest efforts on validating the feasibility of wireless positioning and sensing. Through over-the-air measurement and processing in a 5G NR millimeter network with a commercial grade user equipment (UE), we demonstrate that the NR positioning and sensing techniques are mature enough to drive the evolution toward perceptive and cognitive wireless communications.

Integrated sensing and communication have drawn great research attention in recent years. Specifically, 5G mmWave has demonstrated its capabilities not only in high-speed communications but also in perceiving the physical environment. Apart from providing locationing services for user equipment (UE), 5G mmWave can also estimate the position of target objects that does not carry any equipment (i.e., device-free). Existing works of device-free wireless localization often employs a single monostatic radar or a few transceivers in fixed positions. In this work, we examine a cooperative sensing case, where multiple UEs cooperate with the infrastructure of transmit/receive points (TRPs) to jointly locate device-free objects. This new setting brings in new challenges for existing locationing algorithms as the number and the locations of the UEs and sensing targets are all dynamic. Our work proposes a novel procedure that uses visualization methods to jointly represent the information in the mmWave channel impulse responses and the locations of UEs and TRPs. We then introduce an end-to-end deep learning transformer architecture inspired by popular models in the computer vision domain to estimate the target objects’ locations from the visualizations. On a dataset generated using 3D ray-tracing simulations, our system can locate multiple device-free objects with an average error of 0.47 meters within a 20 meterby-40 meter experiment area.

Full-duplex (FD) millimeter-wave radios can increase the spectral efficiency and reduce the latency. In order to take an advantage of the FD technology, a self-interference (SI) needs to be suppressed. In this paper, we investigate the characteristics of SI in the 28 GHz band using over-the-air (OTA) measurements. Using our 28 GHz OTA FD platform, we investigate passive isolation with different antenna layouts and beam steering. We also use an intra-band carrier aggregation signal based on the fifth generation NR standard, and propose a nonlinear interference cancellation (NLIC) technique that works on a per-carrier component (CC) basis for multi-carrier systems. To deal with intermodulation distortion, we propose to include inter-CC terms for NLIC and confirm that the proposed FD and NLIC designs provide sufficient self-interference cancellation performance.

Reconfigurable Intelligent surface (RIS) has emerged as a candidate technology for enhancing coverage in millimeter wave wireless networks at a lower energy and cost footprint. RIS has several antenna elements that can each be configured to reflect impinging electromagnetic waves after imparting a chosen phase shift with possibly some amplitude attenuation (a.k.a. chosen reflection coefficient). Over the canonical RIS-enabled communications scenario, an optimal choice of reflection coefficients (or optimal RIS pattern) can be efficiently determined for an ideal unit-amplitude unconstrained phase alphabet. However, for most practical RIS that entail finite alphabets with amplitude imbalance and per-group-of-elements control, efficiently determining optimal patterns remain open problems. In this paper we resolve two such open problems by designing optimal and efficient pattern determination algorithms for binary and quaternary alphabets. We show that our algorithms can noticeably improve over the state-of-art conventional heuristic, especially in the presence of high amplitude imbalance and more restrictive per-group control. The designed algorithms also yield companion sets, which we show offer very significant advantages in RIS pattern selection under interference limit (leakage suppression) constraints.

The current approach to provide cellular service is for a Mobile Network Operator (MNO) to build and operate its own network. As the density of the deployed devices increases, it becomes necessary to deploy more cells and/or utilize more bandwidth. Finding new sites for deploying new cells is getting increasingly difficult, due to various reasons, such as permitting process, deployment, and operating costs. Utilization of more bandwidth is generally associated with use of higher frequency bands, such as mmW due to the lack of available spectrum in low to mid-band range associated with good capacity coverage tradeoff. Use of mmW for large macro cell sites, may not always result in adequate coverage due to increased losses through modern building materials, tree foliage, rain etc. In this paper, to address the limitations of the current approach, we propose and study a new approach, exploiting sharing and cooperation among UEs. Compared to a typical relaying approach, UE cooperation feature increases radio capabilities of the aggregated devices and could be largely implemented in a manner transparent to the network, allowing the devices to improve Key Performance Indicators (KPIs), such as coverage and capacity, without any further major capital investment in a cellular network.

Artificial intelligence (AI) or machine learning (ML) based beam prediction is currently studied in the 3rd Generation Partnership Project (3GPP) fifth generation (5G)-Advanced new ratio (NR) standardization for future commercialization and standard evolution towards sixth generation (6G) communications, wherein time domain (TD) beam prediction is an important use case. The targets for such 3GPP studies and standardization are to lower power consumed at user equipment (UE) and reference signal (RS) overhead that are currently needed by frequent beam measurements due to UE rotation and mobility. To meet such targets, in this paper, we investigate AI/ML based algorithms facilitating TD beam prediction suitable for 5G-Advanced beam management (BM), including RS receive power (RSRP) prediction and beam change prediction. The proposed AI/ML algorithms are first evaluated through computer simulations with new UE mobility models based on recent standard evolutions in 3GPP. Then we further present over-the-air test results achieved by such AI/ML algorithms, using based station (BS) and UE compliant with 3GPP standards. Evaluation results show that the proposed schemes can accurately predict future beams and reduce large amount of the power consumed at the UE for BM, which also demonstrate feasibility of AI/ML based BM for 5G-Advanced and future 6G communications.

Analog beamforming is the predominant approach for millimeter wave (mmWave) communication given its favor-able characteristics for limited-resource devices. In this work, we aim at reducing the spectral efficiency gap between analog and digital beamforming methods. We propose a method for refined beam selection based on the estimated raw channel. The channel estimation, an underdetermined problem, is solved using compressed sensing (CS) methods leveraging angular domain sparsity of the channel. To reduce the complexity of CS methods, we propose dictionary learning iterative soft-thresholding algorithm, which jointly learns the sparsifying dictionary and signal reconstruction. We evaluate the proposed method on a realistic mm Wave setup and show considerable performance improvement with respect to code-book based analog beamforming approaches.

Full-duplex (FD) radios, which are expected to increase the spectral efficiency and reduce the latency, can help meet the ever-increasing capacity demands, in combination with the millimeter-wave (mmWave) band. In order to utilize the benefits of the FD technology, it is important to understand the characteristics of the self-interference (SI) in the mmWave band. In this paper, we study the impact of SI in the 60 GHz band and design self-interference cancellation (SIC) techniques. Using a 60 GHz hardware platform, we investigate passive SI isolation with beam steering. We also propose a nonlinear interference cancellation (NLIC) technique of feasible complexity with a limited number of kernels. Over-the-air link performance confirms that the proposed FD transceiver design provides sufficient SIC performances.

In this paper, we discuss UE power saving (PS) techniques for eXtended Reality (XR) applications in 5G NR system. XR traffic is characterized by pseudo periodic packet arrival, high bit rate, and low latency requirement. It is challenging to support such traffic in a small device form factor with limited battery size, e.g., head-mounted display (HMD) or XR glasses. Therefore, it is desired for 5G NR to support effective UE PS techniques. We first discuss a few PS techniques defined in 5G NR specifications Release 15 and 16. Then, we introduce new PS techniques that are evaluated via system level simulations with the UE power consumption model defined in a 3GPP. Evaluation results show that UE power consumption can be substantially reduced by the new PS techniques, up to 40%.

We present system-level evaluation results to characterize the performance of eXtended Reality (XR) applications over a 5G NR system. A split-rendering framework is assumed, where the XR device uses 5G to offload computation associated with rendering and encoding of video frames to an edge server. The XR traffic model includes rendered video frames on the downlink and user pose and control updates on the uplink. We present results for both sub-6 GHz (FR1) and millimeter-wave (FR2) bands. Additionally, we also discuss potential enhancements to further improve user experience for XR applications over 5G.

Joint transmit beamforming, power control and the MMSE receiver are proposed for downlink channels of CDMA communication systems. Variability of the total transmit power while satisfying each user's quality of service is studied. We show that the total transmit power has a significant variation when we consider the combination of MMSE receiver and power control. When transmit beamforming is added, such variation is reduced significantly and great transmit power saving is achieved through the proposed scheme. The signature diversity concept is revealed for joint transmit beamforming and the MMSE receiver. An optimum MMSE, transmit beamforming and adaptive frequency hopping scheme is proposed to maximize the benefits of the MMSE receiver and transmit beamforming. Two sub-optimal algorithms are proposed to solve the optimization problem with reduced computational requirements.

A new double-filter receiver for spread spectrum wireless systems is presented. The receiver exploits the inherent diversity due to the small delay difference between the multipaths which may not be exploited by the conventional RAKE receiver. The bit error rate performances of both the coherent and the noncoherent detectors are analyzed and compared with the ideal RAKE receiver and the matched filter bound. We also examine the best and the worst cases of a three-path fading channel for the proposed receiver. Optimum pulse shapes under various channel conditions are designed. Implementation issues in code-division multiple access systems are discussed. The numerical results show that the proposed receiver achieves significant gains for a given spreading factor under a complexity constraint.

We study the interference effects in a multirate DS-code-division multiple-access (CDMA) system. Optimum chip waveform selection with arbitrary shapes is analyzed using a time domain approach. The problem is posed as an interference minimization problem under energy and time-bandwidth constraints and prolate spheroidal wave functions are used to arrive at a solution. Various factors affecting the interference are identified and the trade-off between competing factors is analyzed. The effect of the interchip interference on the optimum chip waveform design is also quantified under a practical bandwidth constraint. We study the benefits of employing two different chip waveforms for two classes of users. We compare the performance of systems employing two different chip waveforms with that of a single-chip waveform system such as IS-95. We show that when the power imbalance is large, it is advantageous to employ two different chip waveforms for different classes of users.

A new double-filter receiver for spread spectrum (SS) wireless systems is presented in this paper. The receiver exploits the inherent diversity due to the small delay difference between the multipaths which may not be exploited by the conventional Rake receiver. The bit error rate (BER) performances of both the coherent and the non-coherent detectors are analyzed and compared with the ideal Rake receiver and the matched filter bound. We also examine the best and the worst cases of a three-path fading channel for the proposed receiver. Optimum pulse shapes under various channel conditions are designed, Implementation issues in CDMA systems are discussed. The numerical results show that the proposed receiver achieves significant gains for a given spreading factor under a complexity constraint.

This paper studies the benefits of combining beamforming and power control in the downlink channel of a CDMA communication system employing a variable length orthogonal spreading code (OLSC). It is demonstrated that the high data users cause significant interference to the low data users if we do not employ beamforming. Furthermore, if we perform beamforming for both types of users, most of the gain we achieve is due to the high data users if the total power of the high data users is much higher than that of the low rate users. Thus we may reduce the computational requirement significantly if we perform beamforming for the high data users only with a small increase in total power.

Optimum chip waveform selection with arbitrary shape is analyzed and a time domain approach is employed. The problem is posed as an interference minimization problem under energy and time-bandwidth constraints and prolate spheroidal wave functions are used to arrive at a solution. We propose to employ two chip waveforms for two classes of users. A 3 dB gain is possible for one class of users without penalizing another class of users. Various factors affecting the interference are analyzed. The tradeoff between competing factors is quantified. The effect of the interchip interference on the optimum chip waveform design is also quantified under a practical bandwidth constraint.