Search

The forthcoming 6G technology aims to ensure widespread connectivity through Space-Air-Ground Integrated Networks (SAGINs), crucial for regions lacking proper terrestrial infrastructure. Utilizing High-Altitude Platform Stations (HAPSs) as aerial base stations, 6G facilitates low-latency internet access. However, the exponential growth in terrestrial wireless devices puts a strain on the available frequency bands shared by terrestrial and non-terrestrial networks. Therefore, in this research, we aim to optimize the resources between HAPS-to-satellite and HAPS-to-terrestrial uplinks. One challenge that arises is the interference caused by simultaneous uplink and downlink transmissions, especially from uplink back lobes disrupting ground user downlinks. To mitigate this interference, current methods like the Zero-Forcing (ZF) proposed a technique to steer nulls toward ground users. However, current methods randomly allocate limited nulls, potentially leaving many users outside null zones and experiencing interference. To address this issue, we propose a solution to design null directivity based on user locations. This is achieved by creating a codebook of antenna patterns that are linked to weight matrices. Selecting suitable antenna patterns from the codebook and adjusting weight matrices accordingly enables null formation at positions that maximize user communication satisfaction. Numerical analysis confirms the effectiveness of this approach, showing increased user satisfaction compared to existing methods.

Hybrid Intelligent Reflecting/Refracting Surfaces (HIRS) offer a promising solution to the limitations faced by Intelligent Reflecting Surfaces (IRS) in high-frequency bands, owing to their versatile utilization of reflective and refractive techniques across indoor and outdoor surfaces. This paper investigates a hierarchical beam-sweeping (HBS) approach tailored for HIRS-assisted millimeter-wave networks, aiming to pinpoint user locations with reduced overhead. However, the HBS method typically entails numerous coarse steps for user identification, primarily due to its initial exploration starting with broader beam widths at the lowest detectable signal-to-noise ratio (SNR), gradually narrowing down with increasing SNR. Such a process raises concerns about inadequate time for user discovery within HIRS coverage and available maximum beam search time. To mitigate this, we propose a novel method to optimize beam width allocation at the outset of beam searching, prioritizing maximum SNR for all users in the initial step to avoid multiple steps of HBS. Through extensive simulations and analyses across varied indoor and outdoor environments, encompassing diverse user counts and coverage distances, we demonstrate the efficacy of our proposed approach in enhancing network performance. By dynamically adjusting beam widths, our method adeptly accommodates the requirements of diverse indoor and outdoor environments within the allocated maximum beam searching time.

Intelligent reflecting surfaces (IRSs) have attracted attention for realizing ultrahigh-speed and large-capacity communication in the next-generation 6 G communication standard. Still, research has indicated that it is difficult for an IRS to communicate simultaneously with multiple users. Therefore, the common method for communication via IRSs is to allocate a wide bandwidth to one user to perform large-volume communication in a short period per user and to support the communication of many users by repeating the same process. However, allocating broadband to a single user causes a reflected beam shift owing to the beam squint, reducing the frequency efficiency. Therefore, we propose a control and frequency allocation method for IRSs that can support multiple users simultaneously without reducing the frequency efficiency by effectively utilizing a beam squint. Specifically, the proposed method maximizes the frequency efficiency by controlling the IRS reflection direction, reflected beam width, and resource block allocation to user equipment (UE), exploiting the shift in the reflection direction caused by the beam squint. In this study, we evaluated the proposed method with regard to frequency efficiency and simultaneous multiple-connection performance through simulations.

WiGig networks and 60 GHz frequency communications have a lot of potential for commercial and personal use. The high-frequency bands can provide high transmission rates, but their high amplitude makes it so the signal cannot go through any walls or obstacles. The signal also has a strong path loss element caused by the high frequency, significantly limiting the reach of connections because the signal is too weak at moderate distances. Due to these issues, users can easily lose connection with the access point while moving and need to connect to a new device, making WiGig systems unstable as they need to rely on frequent handovers to maintain a high-quality service. However, this solution is problematic as it forces users into bad connections and downtime before they are switched to a better access point. In this work, we use machine learning to identify patterns in user behaviors and predict user actions. This prediction is used to do proactive handovers, switching users to access points with better future transmission rates and a more stable environment based on the future state of the user. Results show that not only the proposal is effective at predicting channel data, but the use of such predictions improves system performance and avoids unnecessary handovers.

Recently, mobile traffic has increased and is expected to increase further. Therefore, the deployment of high-density networks with many small-cell base stations within the coverage area of the macro-cell base station is being considered as a network that can increase the capacity of mobile networks. The open-radio access Network (O-RAN), which divides the functions of the radio access network, has also attracted attention. However, in the networks, the function of the base station stops presumably because of power outages or failures in the event of disasters, resulting in the failure of user equipments (UEs) to communicate with external networks. Then, use of Electric Vehicles (EVs) with base station functionality as alternatives to base stations during disasters is being considered. Therefore, this study examines networks using the aforementioned EVs and highlights the problems caused by inoperable EVs on UEs that cannot communicate with external networks and the need for power consumption by network operation and movement during disasters. Furthermore, we propose a method to rebuild networks by considering the power consumed by network operation and movement. In the proposed method, we model the system, express each power consumption mathematically, and optimize the placement of EVs and transmission power. In addition, the results of the experiments validate the effectiveness of the proposed method in various environments. The reduction in the power consumed by the network operation and movement improved the feasibility of these networks.

Recently, Internet traffic has surged due to the widespread demand for teleworking and flat-rate video distribution services. It is expected that such enormous traffic with diverse patterns will be managed by employing beyond fifth-generation backbone networks, such as satellite networks. Satellite communication resources are scarcer than those of terrestrial communication. Therefore, this study has focused on ensuring efficient satellite network resource operations. Traffic forecasting is a promising approach to facilitate optimal resource allocation. While there are several methods for traffic prediction in terrestrial communication using machine learning, an insufficient number of studies have been conducted regarding satellite communications. Existing traffic prediction approaches consume high bandwidths, which can be a problem for the bandwidth used by users over the limited bandwidth of satellite networks. Therefore, this study proposes a lightweight machine-learning- based traffic prediction method using transfer learning to reduce bandwidth consumption. Furthermore, we demonstrated the effectiveness of the proposed method via simulations by comparing its accuracy with conventional approaches.

Intelligent reflecting surfaces (IRS) have garnered attention for the high-capacity and reliable communication required by Beyond 5G. However, IRS control in existing studies assumes information sharing and control instructions from base stations (BSs), thus requiring a backhaul link to the core network and BSs. This requires the construction of a network system, ruining the advantages of IRS, such as its low cost and installation flexibility. Therefore, a standalone-IRS, which is completely independent of the core network and BSs, is expected to maximize the performance of the IRS. However, a standalone-IRS cannot obtain sufficient information for control decisions, such as channel state information, which is frequently used in existing studies; therefore, inefficient control methods, such as exploration, must be used. Thus, this study proposes a significant reduction in the overhead required for exploration by hierarchical exploration and a reduction in the exploration itself using codebooks to store the results of exploration. Furthermore, to prevent the degradation of communication quality owing to the use of hierarchical exploration and codebooks, we propose an effective beamwidth expansion method for hierarchical exploration and an efficient codebook construction method based on user distribution as elemental methods. In this study, we evaluated the overhead reduction and communication quality of the standalone-IRS control by combining these methods through simulations.