Abstract: With the improvement of launch technology and the expansion of the demand for space missions，LEO internet constellation has gradually become a research hotspot in the aerospace industry.Computing resource awareness at the network layer helps to build the shortest computing delay path and effectively use the node resources on the edge or path for task scheduling.The traditional networking protocol did not consider the computing power resources of neighbor nodes，it was difficult to achieve resource awareness and inter satellite collaborative networking to complete the unloading task.In order to solve the problems mentioned above，the LEO satellite network was modeled as a MANET network，and the node computing resource was introduced into the active protocol OLSR to sense the computing power，CPU，memory and load of the surrounding network nodes，and the MPR selection algorithm and routing table update algorithm were modified according to this index.Simulation results show that the proposed routing protocol reduces task computing latency by 15% to 30% in inter satellite collaborative computing，and verifies the advantages of inter satellite collaborative computing by comparing it with terrestrial cloud computing.

Key words: Mobile Ad Hoc Network（MANET）, OLSR, computing power networks, MPR selection algorithms, computational resource awareness