Throughput of wireless networks powered by energy harvesting

Designing mobile devices for harvesting ambient energy such as kinetic activities or electromagnetic radiation (EMR) will enable mobile networks to self sustain besides alleviate global warming. The throughput of a mobile ad hoc network powered by energy harvesting is analyzed in this paper using a stochastic-geometry approach. The transmitters powered by energy harvesting are modeled as a Poisson point process (PPP); each transmits to a receiver at an unit distance using either a random-access protocol or the time-hopping multiple access (THMA) and satisfying an outage-probability constraint. Consider non-EMR energy harvesting where energy packets of random sizes arrive at a transmitter following a stationary random process. By applying Mapping Theorem, the network (spatial) throughput for random access and in the limit of a long harvesting interval is derived in simple closed-form functions of the energy-arrival rate, transmitter density and coding rate. These results show that the throughput of a sparse network increases logarithmically with the energy-arrival rate and linearly with the transmitter density. Moreover, dense energy arrivals provide marginal throughput gain as the network becomes interference limited but this gain can be enhanced using THMA. Next, EMR energy harvesting is also considered where transmitters harvest energy from transmissions in coexisting networks modeled as independent PPPs. The corresponding expressions of the network throughput can be modified from their non-EMR counterparts such that the harvested EMR power per mobile is equal to a sum of coexisting-network densities weighted by corresponding transmission power and harvesting efficiencies.