With the steady development of geostationary orbit (GEO) satellites in China and the accelerated construction of low-orbit (LEO) satellite internet, the integration application of high and low orbit heterogeneous constellations has become an important direction for future development. Current research mainly focuses on resource allocation within a single constellation, such as GEO constellations or LEO constellations, while there is insufficient attention to the collaborative allocation of heterogeneous resources in mixed high and low orbit and cross-constellation scenarios. Therefore, this paper takes the China-Sat, Asia-Pacific and LEO satellite internet systems as research objects, deeply analyzes the service transmission modes and resource characteristics of different satellite systems such as transparent forwarding, high throughput and LEO constellations. At the same time, from the current engineering construction status, a heterogeneous resource allocation strategy for cross-high and low orbit mixed satellite networks is proposed. This strategy takes dynamic communication service demands as input and collaboratively allocates beam bandwidth, frequency, time slot, power and inter-satellite links and other heterogeneous resources. The research results can provide support for the simulation modeling and business planning of high and low orbit mixed satellite networks.
Zhihao Wang, Hongbin Luo, Zhiyuan Wang et al.· Peer-to-Peer Networking and...· 0 citations
Deploying large language models (LLMs) on edge nodes enables low-latency and privacy-preserving inference, but faces severe resource constraints under high-concurrence workloads. While existing inference systems leverage intranode key–value (KV) caching to improve efficiency, they largely neglect the unique complexities of multinode edge environments. Specifically, reactive KV cache eviction policies suffer from temporal uncertainty, often discarding reusable KV caches prematurely, while the tight coupling between request scheduling and cache placement often leads to myopic decisions that exacerbate load imbalance and resource contention. To address these challenges, we propose a dynamic block-level paradigm that treats KV blocks as the fundamental units for caching and scheduling, enabling dynamic sharing, generation, and eviction of arbitrary-length prefixes. We present complete modeling of the spatiotemporal coupling between scheduling and caching under block-level granularity, capturing intricate interactions overlooked by prior work. Based on this model, we design an online joint optimization algorithm, which applies to general edge LLM serving scenarios. The algorithm decouples spatiotemporal dependencies via randomized rounding over per-slot subproblems, achieving a balance between real-time responsiveness and long-term system efficiency. Theoretical analysis establishes high-probability near-optimality guarantees, and extensive experiments show that our method reduces the average time to first token (TTFT) by up to 54.02% over existing baselines.
Xishuo Li, Wei Jiao, Jun He et al.· IEEE Internet of Things Jour...· 0 citations