New Algorithm Balances Load in Dynamic LEO Satellite Networks

Researchers have developed a dynamic resource allocation method for low Earth orbit (LEO) satellite networks designed to address the challenges posed by satellite mobility and beam interference. The study, published in the journal Discover Space, introduces a method known as LDCB-MSHBRA, which stands for dynamic low Earth orbit multi-satellite hopping beam resource allocation considering load and dynamic coverage balance. This approach aims to improve overall system throughput while ensuring fair distribution of time, frequency, and power resources among multiple satellites with overlapping coverage areas. The work is significant because traditional ground-based communication systems often struggle to meet the needs of remote, mountainous, and maritime areas, making satellite communication a critical component of modern wireless infrastructure.

The core problem addressed by the LDCB-MSHBRA method is the uneven system load and changing coverage caused by the movement of LEO satellites. In multi-satellite and multi-beam communication scenarios, the mobility of satellites leads to continuous changes in their coverage areas, which can result in beam interference and reduced resource utilization. The proposed method uses time slicing to dynamically allocate resources, allowing the system to adapt to these changes in real-time. By focusing on load balancing and dynamic coverage balance, the algorithm seeks to enhance the efficiency and reliability of satellite communication services, particularly in scenarios where multiple satellites provide overlapping coverage.

The LDCB-MSHBRA method involves a six-step calculation process. First, a LEO satellite system with three identical configuration and motion speed parameters is established. Second, values are assigned to the time slicing weight coefficients, with the optimization objective being to minimize the difference in demand for dual-star services based on the principle of load balancing. Third, an initial power resource allocation is obtained with the goal of maximizing system capacity. Fourth, considering the mobility of the satellites, weight coefficients are updated, and power resources are dynamically allocated to obtain a time resource dynamic allocation matrix and the optimal service cell set. Fifth, the independent coverage cell set of the third satellite is updated, and the Lagrange equation of the weighted fair objective function is used to obtain the time resource allocation optimization matrix. Finally, with the goal of maximizing system capacity, the optimal solution for the power resource allocation matrix of the third satellite is determined.

The study utilizes a beam hopping mechanism, where satellite beams are dynamically directed to different ground areas, or wave positions, over time. This technique allows a limited number of beams to serve a larger number of wave positions by multiplexing in the time domain. In the proposed system, the number of beams is smaller than the number of wave positions, necessitating the adjustment of beam pointing positions to ensure all areas are illuminated. This process, known as beam hopping, provides communication opportunities to more users, although it comes with the cost of beam adjustment loss. The method employs gaze beam technology, where the beam pointing wave position is fixed and unchanged, and the wave position division on the ground is planned in advance.

To evaluate the effectiveness of the LDCB-MSHBRA algorithm, the researchers conducted experimental simulations on LEO satellite beam hopping technology. The simulation scenario involved a service area of 6,670 by 6,670 square kilometers, divided into 49 equally sized areas. The satellite network consisted of a predetermined orbit and a LEO satellite with 7 point beams. The physical parameters of the simulation were based on the Iridium system, with an average packet size of 50 kbit and packet arrivals following a Poisson distribution. The arrival rate range was set between 1 and 21 packets per second.

The performance of the LDCB-MSHBRA algorithm was compared with two other algorithms: a deep learning method for beam hopping in multi-beam satellite systems and a resource scheduling algorithm based on beam hopping in LEO satellite networks. The comparison focused on the number of time slots allocated, business processing, and average packet delay under changing business demand patterns in different regions. The results showed that the LDCB-MSHBRA algorithm significantly improved business processing performance in high-demand areas while meeting regional quality of service (QoS) requirements. This indicates that the proposed method is more effective in ensuring that user demands are met within the specified service levels.

In terms of average packet delay, the LDCB-MSHBRA algorithm demonstrated superior performance compared to the other two algorithms. The overall packet delay in the LDCB-MSHBRA algorithm region was significantly lower than that of the algorithms in references [14] and [15]. The average packet delay of each region in the LDCB-MSHBRA algorithm was reduced by 28.15% and 19.56% compared to the algorithms in references [14] and [15], respectively. The algorithm’s ability to minimize packet delay is attributed to its dynamic resource allocation strategy, which adapts to the current queue situation, channel status, and power allocation situation.

The study also investigated the impact of different service arrival rates on system throughput and average packet delay. As the system’s business arrival rate increased, the number of data packets to be processed increased, leading to an increase in the total throughput of the system. However, due to the accumulation of queues, the average packet delay also increased. The results showed that the total system throughput and average packet delay of the LDCB-MSHBRA algorithm were better than those of the algorithms in references [14] and [15]. Moreover, the performance degradation rate of the LDCB-MSHBRA algorithm was slower as the service arrival rate gradually increased, indicating its robustness under high load conditions.

Furthermore, the researchers examined the impact of different user numbers on the performance of system resource allocation. The user numbers were set to a range from 120 to 220, with users following a non-uniform distribution and multiple hotspot areas. The results showed that as the number of system users increased, the total system throughput under different algorithms also increased. This is primarily because an increase in the number of users leads to an increase in the total demand of the system. In the case of fixed communication resources, the number of cases where the demand is less than the resource allocation decreases, thereby reducing resource waste. However, the system throughput did not increase proportionally with the number of users, mainly due to increased co-channel interference within the system, which reduces the transmission rate of some users and constrains communication performance.

The study also explored the changes in system throughput performance under different numbers of wireless resources. In this simulation, the number of users was kept constant at 160, while the number of assignable subcarriers was set to a range from 20 to 40, and the assignable power resources were set to a range from 20 to 20.8 dBw. The results indicated that as the number of assignable subcarrier resources increased, the system throughput also increased. This is because an increase in the number of assignable subcarriers leads to finer resource allocation granularity and a decrease in corresponding co-channel interference. Additionally, as the system power increased, the throughput of the system also increased. These findings demonstrate that the proposed dynamic resource allocation algorithm can effectively utilize resources and improve the system’s business processing performance, even when the available resources vary.

The LDCB-MSHBRA method is based on a fairness model that considers load balancing and dynamic coverage range changes. By ensuring the fairness of resource allocation, the method helps to improve the throughput and resource utilization of satellite systems. The study highlights the importance of considering the mobility of LEO satellites and the dynamic nature of their coverage areas in the design of resource allocation algorithms.

The researchers noted several areas for future work. One direction is exploring beam resource scheduling solutions for mobile terminals, particularly in scenarios where user terminals are moving at high speeds and their wave positions are not fixed. This includes applications in aviation, where planned flight routes can change in real-time due to factors such as weather, traffic control, and mechanical malfunctions. Another area of future work is the introduction of inter-satellite links to optimize the layout of the signal station, enabling it to be shared by all LEO satellites and reducing deployment and operational costs. However, the number of inter-satellite communication transmission hops should not be too high due to limitations in onboard resources and communication latency.

The study also addressed the issue of queuing delay caused by time slot allocation, which can lead to higher latency differences and affect user experience. In further scheduling strategies, dynamic allocation of bandwidth, power, and time slots is considered to optimize the delay jitter rate. The scalability of the proposed method for larger satellite constellations and different configurations was also discussed. While the method has a certain degree of scalability, it is important to note that it is an algorithm for near-Earth satellites. According to the experimental results, the average packet delay is between 200 and 300 ms, which can meet the transmission delay requirements of near-Earth satellites. However, as the satellite altitude increases, it is necessary to comprehensively consider the requirements of transmission delay and achieve a better balance between resource allocation and transmission delay.

In conclusion, the LDCB-MSHBRA method offers a promising solution to the challenges of resource allocation in dynamic LEO satellite networks. By addressing the issues of changing coverage and beam interference caused by satellite mobility, the method improves system throughput and resource utilization while ensuring fairness in resource distribution. The experimental results demonstrate the effectiveness of the algorithm in various scenarios, including different service arrival rates, user numbers, and wireless resource configurations.

Subject of Research: Dynamic low earth orbit multi satellite hopping beam resource allocation considering load and dynamic coverage balance (LDCB-MSHBRA)

Article Title: Dynamic low earth orbit multi satellite hopping beam resource allocation considering load and dynamic coverage balance (LDCB-MSHBRA)

Article References: Li, X., Zhang, X., & Chang, J. (2024). Dynamic low earth orbit multi satellite hopping beam resource allocation considering load and dynamic coverage balance (LDCB-MSHBRA). Discover Space, 128(1), Article 3. https://doi.org/10.1007/s11038-024-09557-5

Image Credits: AI Generated

DOI: 10.1007/s11038-024-09557-5

Keywords: Dynamic, earth, orbit, multi, satellite, hopping, beam, resource, allocation, considering, load, coverage