Abstract:In order to make efficient use of network resources, balance the energy consumption of network topology, eliminate redundant links between network topology, reduce node loads, and extend the life cycle of network of the maximum, in this paper a topology optimization algorithm for wireless sensor networks (PGOSG) was designed based on the concept of potential game and optimal stiffness sub-graph, which considers the residual energy of nodes, the load of nodes and the redundancy of network topology links. First, according to the power change of communication between nodes, the power set of nodes was constructed as the strategy set of the game, and the potential-game function was constructed with the goal of balancing energy consumption by using the potential game theory, which converges to the Nash equilibrium point, and then constructs the preliminary network topology structure. Second, taking advantage of the fact that the number of global links with the optimal rigid graph is small and does not damage the network topology, an optimal rigid sub-graph was used to eliminate redundant links between to the network topology layer by layer to obtain the final network topology. The network topology, link communication quality, network robustness and the network life cycles of PGOSG algorithm were analyzed by the simulation experiments, and compared with the existing DEBA algorithm. The simulation results show that: in the topology, PGOSG algorithm eliminates redundant links to the network communication links, and reduces the load of some nodes in the network. In terms of energy consumption balance, PGOSG algorithm formulates node data forwarding rules, effectively utilizes network resources, balances node energy consumption and avoids redundant forwarding between nodes in the game stage. Therefore, the algorithm proposed in this paper can eliminate redundant links in the network, reduce the node load and link weight, and prolong the network lifetime.