In view of the traditional positioning defects based on received signal strength, a new K-NNC location algorithm is proposed for IOT, based on K- neighbor node coverage (K-NNC). K-NNC is divided into two stages, select neighbor nodes and positioning stage. In order to reduce the influence of long distance nodes on the positioning, by adjusting the transmit power to select the nearest K neighbor node. In positioning stage, the node calculates the weight by the received signal strength of the K beacon node. The coordinates of the nodes are calculated by the weighted sum. Using the self-correcting method the position coordinates are compensated. The K-NNC positioning model can effectively avoid the influence of environmental factors on the positioning, and the positioning algorithm is simple. Experiment shows the K-NNC positioning model positioning is highly accurate.