Abstract
Purpose - Wireless sensor networks (WSN) have many important practical applications. A significant limitation for WSN is its short lifetime due to the limited capacity of the battery. Renewable energy can significantly extend the lifetime of WSN. In this paper, we investigate the whole sequence of lifetimes of every sensor in WSN, since different application scenarios have different requirement on how many sensors can die until the WSN is no longer functional. Design/methodology/approach - We use linear programming formulation to investigate both the conditional and unconditional lifetime sequence of WSN. We study the lifetime sequences of WSN without, and with differ levels of solar power.Findings - Our investigation of lifetime sequences discovers three interesting phenomena: 1) the sensors that die first are on the peripheral of the network rather close to the base station; 2) multiple sensors tend to die simultaneously; 3) the lifetimes of sensors that die later can be extended by renewable energy much more significantly than those that die early, which is very good news to applications that can tolerate the death of a fraction of sensors.Originality/value - In this paper, we provide the first optimization formulation for maximizing both unconditional and conditional lifetime sequences of WSNs with renewable energy sources. Only the conditional lifetime sequence has been investigated in a previous work, but our method runs n times faster than the previous work, with n being the number of nodes in the WSN.
Purpose - Wireless sensor networks (WSN) have many important practical applications. A significant limitation for WSN is its short lifetime due to the limited capacity of the battery. Renewable energy can significantly extend the lifetime of WSN. In this paper, we investigate the whole sequence of lifetimes of every sensor in WSN, since different application scenarios have different requirement on how many sensors can die until the WSN is no longer functional. Design/methodology/approach - We use linear programming formulation to investigate both the conditional and unconditional lifetime sequence of WSN. We study the lifetime sequences of WSN without, and with differ levels of solar power.Findings - Our investigation of lifetime sequences discovers three interesting phenomena: 1) the sensors that die first are on the peripheral of the network rather close to the base station; 2) multiple sensors tend to die simultaneously; 3) the lifetimes of sensors that die later can be extended by renewable energy much more significantly than those that die early, which is very good news to applications that can tolerate the death of a fraction of sensors.Originality/value - In this paper, we provide the first optimization formulation for maximizing both unconditional and conditional lifetime sequences of WSNs with renewable energy sources. Only the conditional lifetime sequence has been investigated in a previous work, but our method runs n times faster than the previous work, with n being the number of nodes in the WSN.