Network Energy Driven Wireless Sensor Networks

Introduction

A common challenge in wireless sensor networks is the limited energy resource of field nodes. Existing techniques on increasing network lifetime aim at efficient processor design and network protocol level solutions, none of which can help replenish the drained battery of a field sensor node. Many deployment scenarios in hazardous and inaccessible terrains demand that once the sensor nodes are deployed they cannot be physically accessed for replacing the drained batteries. Also, since the field nodes could be hundreds of thousands in number in many cases, once deployed, it is not feasible to physically access all of them for recharging. This restriction puts a huge cost of network operation and waste of technology resources, as new nodes need to be redeployed to replace the existing nodes that have drained batteries but are otherwise functionally capable. Even if the energy drained nodes can be collected and recharged, off-line recharging solution adds disruption to the network operation. Certainly, a disruption-free network operation, which requires online recharging of the sensor nodes, would add a significant value to the physical world networking via field sensors.

One technology that serves the purpose of disruption-free operation is energy harvesting -- a process by which energy is derived from an external ambient source. The available online recharging solutions in the research literature propose to use natural energy resources, such as solar (Raghunathan, et al., 2005), vibration (Meninger, et al., 2001; Roundy, et al., 2004), wind (Weimer, et al., 2006), thermal gradient and strain (Starner, 1996; Shenck and Paradiso, 2001; Gonzalez, et al., 2002), etc., as well as a combination of them (Kansal, et al., 2007), (Paradiso and Starner, 2005; the references therein), which are available subject to the specific deployment scenarios. The possibility of charging wireless node batteries from ambient (other than in-network) radio frequency (RF) energy sources is also being investigated via suitable broadband rectenna designs, e.g., (Hagerty, et al., 2004). Since the non-network ambient RF signal can be of any polarization and frequency, the rectenna design has to be adaptive and uniquely designed for wireless recharging only.

Wireless energy transfer from dedicated RF sources is being studied (Paing, et al., 2007) and commercialized (Powercast Corp., 2010; WildCharge Technology, 2010; E-coupled technology, 2010). Wild charge technology is based on MIT's wireless electricity technology. E-coupled technology provides wireless energy transfer using electromagnetic induction. The RF technique is also being investigated for solar energy transfer from space to the earth at a mega scale (MIT Space Solar Power Workshop, 2007). While RF energy transfer technique at a low power domain has been demonstrated to be feasible, in general energy transfer by RF waves involves significant waste of energy due to power law decay of RF signal power with distance. Also, in a remote area deployment setting such an approach to power a large number of spatially dispersed nodes may prove inefficient.

As an alternative and possibly a more generic and cost-effective solution, we propose to harvest the available network RF energy in a field deployment scenario for powering the field sensors. This approach does not require any special antenna design, as the RF energy is available in the same frequency band and with same polarization as the desired information signal. To this end, we use our observations from the surrounding nature and propose to put in effect the learning from practices of the living world.