Backscatter Communications for the Internet of Things: A Stochastic Geometry Approach

Powering the massive number of small computing devices in the Internet of Things (IoT) is a major challenge because replacing their batteries or powering them with wires is very expensive and impractical. A viable option to enable a perpetual network operation is through the use of far-field Wireless Power Transfer (WPT). We study a large network architecture that uses a combination of WPT and backscatter communication. The network consists of power beacons (PBs) and passive backscatter nodes (BNs), and their locations are modeled as points of independent Poisson point processes (PPPs). The PBs transmit a sinusoidal continuous wave (CW) and the BNs reflect back a portion of this signal while harvesting the remaining part. A BN harvests energy from multiple nearby PBs and modulates its information bits on the composite CW through backscatter modulation. The analysis poses real challenges due to the double fading channel, and its dependence on the PPPs of both BNs and PBs. With the help of stochastic geometry, we derive the coverage probability and the capacity of the network in tractable and easily computable expressions. These expressions depend on the density of both PB and BN, both forward and backward path loss exponents, transmit power of the PB, backscattering efficiency, and number of PBs in a harvesting region. We observe that the coverage probability decreases with an increase in the density of the BNs, while the capacity of the network improves. We compare the performance of this network with a regular powered network in which the BNs have a reliable power source and show that for certain parameters the coverage of the former network approaches that of the regular powered network.