One of the key research topics to extend the network lifetime is choosing the suitable routing algorithm to increase the energy efficiency of nodes in information transfer. An improved EEABR algorithm called N-EEABR is proposed to address the problems that the current ant colony routing algorithm is easily prone to: increasing the communication cost when the long ant packet is used; the routing loop is easily formed when the short ant packet is used; and the uneven distribution of node energy consumption. The pheromone update formula of the backward ant packet is redefined and the deviation value of node energy and path energy is added. This effectively weakens the loop effect of EEABR in short ant packets and balances the residual energy of nodes in the network. "pkt_src" (ant packet source address) and "sq_num" (ant packet sequence number) are added together in the neighbor list of nodes in the EEABR algorithm. The N-EEABR algorithm extracts the RSSI value from the "hello" packet to determine the best transmission power of the nearby node, which helps to further reduce the energy consumption of the node when sending packets. To conserve transmission energy and prevent energy waste, a PCABR-based ant colony routing algorithm is presented. The simulation results demonstrate that these algorithms have some degree of increase in node energy consumption efficiency and network energy balance, with the PCABR's path optimization accuracy being more obvious than that of N-EEABR and EEABR.