For future networks in the 6G, it will be important to maintain a ubiquitous connection, bring processing heavy applications to remote areas, and analyze big amounts of data to efficiently provide services. To achieve such goals, the literature has utilized satellite networks to reach areas far away from the network core, and there has even been research into equipping such satellites with edge cloud servers to provide computation offloading to remote devices. However, analyzing the big data created by these devices is still a problem. One could transfer the data to a central server, but this has a high transmission cost. One could process the data through distributed machine learning, but such a technique is not as efficient as centralized learning. Thus, in this paper, we analyze the learning costs behind centralized and distributed learning and propose a hybrid solution that adaptively uses the advantages of both in a cloud server-equipped satellite network. Our proposal can identify the best learning strategy for each device based on the current scenario. Results show that the proposal is not only efficient in solving machine learning tasks, but it is also dynamic to react to different configurations while maintaining top performance.

Hybrid Centralized and Distributed Learning for MEC-Equipped Satellite 6G Networks

Unmanned aerial vehicles (UAVs) mobility enables flexible and customized federated learning (FL) at the network edge. However, the underlying uncertainties in the aerial-terrestrial wireless channel may lead to a biased FL model. In particular, the distribution of the global model and the aggregation of the local updates within the FL learning rounds at the UAVs are governed by the reliability of the wireless channel. This creates an undesirable bias towards the training data of ground devices with better channel conditions, and vice versa. This paper characterizes the global bias problem of aerial FL in large-scale UAV networks. To this end, the paper proposes a channel-aware distribution and aggregation scheme to enforce equal contribution from all devices in the FL training as a means to resolve the global bias problem. We demonstrate the convergence of the proposed method by experimenting with the MNIST dataset and show its superiority compared to existing methods. The obtained results enable system parameter tuning to relieve the impact of the aerial channel deficiency on the FL convergence rate.

Characterization of the Global Bias Problem in Aerial Federated Learning

The on-orbit processing of massive satellite-native data relies on powerful computing power. Satellite computing has started to gain attention, with researchers proposing various algorithms, applications, and simulation testbeds. Unfortunately, a practical platform for deploying satellite computing is currently lacking. As a result, the industry needs to make relentless efforts to achieve this goal. We suggest using cloud-native technology to enhance the computing power of LEO satellites. The first main satellite of the Tiansuan constellation, BUPT-1, is a significant example of a cloud-native satellite. Prior to delving into the details of BUPT-1, we define the essential concepts of cloud-native satellites, i.e., the cloud-native load and cloud-native platform. Afterwards, we present the design scheme of cloud-native satellites, including the architecture of BUPT-1 and the experimental subjects it can support. Two validation tests are shown to reflect the operation and capability of BUPT-1. Besides, we predict possible research fields that could shape the future of satellites in the next decade.

Satellite Computing: A Case Study of Cloud-Native Satellites

Space-ground integration is a key component of 6G with diverse topologies, time-varying topologies, broad geographical and temporal scales, and limited spatial node resources, bringing new development opportunities and scope for mobile communications. Before designing the 3D 6G space-ground integrated network architecture, we first examine the main difficulties of space-ground integrated networks, and then examine the common use cases and requirements of 6G-oriented space-ground integrated networks. Finally, we propose a space-ground integrated network architecture based on a decentralized core network and a ground-centralized core network. For the former, the core network is placed on the ground while the airborne platform performs certain access network activities. To meet the needs of various application scenarios, the latter airborne platform performs access network operations and loads a portion of the core network components. Dynamic deployment of key network functions becomes possible as the network functions are reconfigured to meet the changing application scenario needs and network operation requirements.

Design of 6G Space-Ground Integrated Network Architecture Based on Ground Core Network

In this paper, we consider a task scheduling problem for the freshness-critical services in the Internet of Remote Things scenario (IoRT). In the IoRT scenario, a gateway collects status updates from the surrounding devices and then makes a scheduling decision, in which the status updates would be offloaded to a specific satellite for on-orbit processing. Our objective is to propose a task scheduling scheme which can minimize the age of information of the system. To this end, we use the promising hybrid geosynchronous earth orbit and low earth orbit (hybrid GEO-LEO) satellite networks and design an age-aware task scheduling scheme to utilize heterogeneous communication and processing resources. The issue of task scheduling is considered as cooperation between gateway association and resource management problem. To cope with this complicated problem, we formulate it as a Markov Decision Process with minimum peak age and decompose it into two sub-problems, which are resource management with fixed gateway association indexes and scheduling decisions for gateway association. The convex optimization algorithm is utilized to obtain optimal resource management results, and the deep reinforcement learning network is used to achieve the optimal gateway association indexes. Extensive simulation results demonstrate that the peak age of the designed strategy has an advantage over other referred strategies.

Age-Aware Task Scheduling Scheme in Hybrid GEO-LEO Satellite Networks

Geosynchronous Earth Orbit (GEO) satellites, which can relay image data for Low Earth Orbit (LEO) satellites, play an important role in remote sensing. With the development of satellite technologies, the significantly improved computation capabilities of GEO satellites have enabled space service computing, through which GEO satellites can provide data processing services before forwarding to reduce the quantity of transmitted data. In the presence of multiple LEO satellites, how to make effective use of limited communication and computation resources in GEO satellites has become crucial. At present, the research on satellite resource management typically focuses on either communication or computation resources. Existing resource management algorithms are usually of slow convergence speed, which limits their applicability in real-time remote sensing scenarios. Therefore, we propose an aggregated resource management method for remote sensing applications. We first propose models for transmission tasks and processing tasks of remote sensing images. Then we formulate the aggregated resource management for satellite edge computing as a hybrid Stackelberg game and simplify the problem to speed up its convergence speed. Then we propose a distributed resource management algorithm to determine the optimal strategies. Simulation results show that the proposed method can quickly obtain the optimal resource allocation strategy and outperforms typical dynamic iterative algorithms in terms of service quantity and throughput.

On the Aggregated Resource Management for Satellite Edge Computing

Federated edge learning is often used in delay-sensitive big data machine learning. However, in cellular networks with limited communication resources, efficient scheduling of large groups of clients has become vital. There is evidence that the impact of the number of participating clients in initial communication rounds is less significant than that of subsequent communication rounds in global model aggregations. Inspired by this, in this paper, we propose a Stochastic Client Scheduling Strategy with Dynamic SINR Thresholds (STUART). In particular, we apply a higher SINR threshold for initial communication rounds in the base station to prevent low-rate clients from participating in the model aggregation, which can effectively reduce local model upload time. Thereafter, the SINR threshold is gradually lowered for the following communication rounds to allow more clients to take part in, hence improving convergence performance. With a convergence constraint, the optimal client scheduling problem is formulated and solved using the genetic algorithm. Finally, we verify the effectiveness of the proposed algorithm through extensive experiments. The results reveal that compared with the fixed threshold method, the STUART algorithm can reduce communication time by more than 1/3 while marginally degrading the convergence performance.

Stochastic Client Scheduling with Dynamic SINR Thresholds for Fast Federated Learning

Remote sensing is one of the main applications of satellite communication. With limited communication time between low earth orbit (LEO) sensing satellites and the ground, geostationary orbit (GEO) satellite relaying has become an effective solution. In practical applications, one GEO communication satellite usually needs to provide data forwarding services for multiple satellites or even multiple satellite constellations. Therefore, communication resource management of GEO satellites has become crucial. Existing methods only focus on resource allocation without taking data forwarding task scheduling into consideration, which limits their applicability in real-time applications. In this paper, we propose a real-time resource management approach for GEO relaying, aiming to maximize network throughputs as well as to reduce transmission delays. In this approach, resource allocation is modeled as a Stackelberg game while task scheduling is modeled as a real-time queuing problem. Furthermore, we propose two real-time algorithms, including an adaptive gradient descent method to find optimal price for data forwarding services with low computational complexity, and a queue-jumping algorithm for data forwarding scheduling to minimize transmission delays. With comparison with existing methods, simulations verify the effectiveness of our proposed method with respect to convergence speed of the algorithm, network throughputs, and transmission delays.

Chang Liu

Papers

On the Aggregated Resource Management for Satellite Edge Computing

Stochastic Client Scheduling with Dynamic SINR Thresholds for Fast Federated Learning

Resource management of GEO relays for real-time remote sensing