US 12,238,013 B1
Communication and computation integrated network architecture and resource adaptation method
Weiting Zhang, Beijing (CN); Dong Yang, Beijing (CN); Hongke Zhang, Beijing (CN); Peixi Liao, Beijing (CN); Jiadong Ren, Beijing (CN); and Ruibin Guo, Beijing (CN)
Assigned to BEIJING JIAOTONG UNIVERSITY, Beijing (CN)
Filed by BEIJING JIAOTONG UNIVERSITY, Beijing (CN)
Filed on Oct. 4, 2024, as Appl. No. 18/906,863.
Claims priority of application No. 202410065417.X (CN), filed on Jan. 17, 2024.
Int. Cl. G06F 9/46 (2006.01); G06F 9/48 (2006.01); G06F 9/50 (2006.01); H04L 47/80 (2022.01); H04L 47/83 (2022.01)
CPC H04L 47/805 (2013.01) [G06F 9/4887 (2013.01); G06F 9/5044 (2013.01); H04L 47/83 (2022.05); G06F 2209/501 (2013.01); G06F 2209/5019 (2013.01); G06F 2209/503 (2013.01)] 6 Claims
OG exemplary drawing
 
1. A communication and computation integrated network architecture, comprising a computing service layer, a resource integration adaptation layer and a computation and network component layer;
wherein the computing service layer is configured to extract features for each computation task of a computing user according to a quality of service requirement and a quality of experience requirement of the computation task to obtain an attribute feature of the computation task; and the computation task comprises determining start and stop of a switch of a train and processing audio/video signals of the train;
the resource integration adaptation layer is configured to carry out task dispatching and resource allocation according to total computing resources and attribute features of all computation tasks to obtain an optimal task dispatching decision and an optimal resource allocation decision, which comprises:
according to the total computing resources and the attribute features of all the computation tasks, by using the inter-domain controller and adopting a quantum genetic algorithm, making decisions on the computation tasks to obtain the optimal task dispatching decision; wherein the optimal task dispatching decision is a joint dispatching scheme of the computation tasks, which comprises computing domain allocation of each of the computation tasks; the total computing resources comprise available resource information of all computing domains;
for each of the computing domains, according to the available resource information of the computing domain and the optimal task dispatching decision, by using the intra-domain controller and adopting the quantum genetic algorithm, allocating resources to the computation tasks corresponding to the computing domain to obtain the optimal resource allocation decision corresponding to the computing domain; wherein the available resource information is acquired by the intra-domain controller from a heterogeneous computing platform;
the computation and network component layer, including a plurality of types of network components, is configured to forward and process the computation tasks according to the optimal task dispatching decision and the optimal resource allocation decision.