The Distributed Jammer Network (DJN) is composed of a large number of tiny, low-power jammers, which are distributed inside a target network and emit radio energy to disrupt its communications. Recent advancement in microelectromechanical system (MEMS) technology makes it possible to make jammers sufficiently small that a DJN can take the form of a dust suspending in the air, thus the name Jamming Dust. Miniaturization of jammers should be less challenging than that of wireless sensors since jammers just emit noise signal without requiring complex modulation, filtering, and other signal processing functions. Therefore, new miniature devices such as nanotube radio may find their first application in jamming dust.
DJN, in a sense, forms a mirror image to a distributed wireless network (DWN), e.g. distributed sensor network (DSN), for DWN communicates information, whereas DJN disrupts such communications. DJN can be deployed to form a lowpower (possibly airborne) jamming dust that is more attractive than traditional high-power jammers because of its low deployment profile (the naked eye can not see the nanotube radios) and its much reduced effect on self-interference.
Advantages of DJN are reminiscent of those of DSN. First, DJN is robust because it is composed of a large number of devices with ample redundancy. Second, DJN nodes emit low power, which is advantageous because of health concerns. Third, DJN is hard to detect because of the nodes’ small size and low power emission. Fourth, DJN provides extended coverage with high energy efficiency. Using the same total amount of power, a DJN of n nodes covers an effective area n1-2/α times larger than that of a single jammer, where α is the path loss exponent with a typical value of 4. So the typical power efficiency gain of DJN is n1/2, which is unbounded as n goes to infinity.
A simulation was performed with the following setup: The target network (DWN) has 100 nodes deployed in an area of 1000 by 1000 meters, with half of the nodes having CBR UDP sessions with the other half of the nodes. The MAC protocol used is IEEE802.11, and two routing protocols are used – AODV and DSR. Ten jammers were deployed with the same transmitting power as the target device (15 dbm). Then, the number of jammers was increased while reducing jamming power, holding the total power consumed by the jammers constant. The performance metric is the ratio of DWN throughput with jamming, versus that without jamming. A phase transition occurs at roughly 20 jammers, where throughput ratio drops precipitously. Given that the total power consumption is constant, the benefit of using a large number of lowpower jammers (i.e., DJN) is evident.
As for the jamming mechanism, it is assumed that jammers in DJN are reactive jammers, which are among the most effective jamming methods. A reactive jammer senses a wide frequency range and jams the channel on which it detects radio activity. Recent advancement in software-defined radio and UWB radio provides technology to make such jammers. The jammer has two parameters: period P and duty cycle D. That is, after sensing a busy channel, a jammer becomes active in D percentage of time in a period of P seconds, after which the jammer restarts sensing again.
It is assumed that transmission powers of both DWN and DJN radios are adaptable up to their respective maximum transmitting powers. If the maximum transmitting powers of DWN and DJN nodes are the same, a DWN receiver is considered jammed if there exists a DJN jammer that is closer to the receiver than the DWN transmitter. In such a case, the jammer can always adapt its power to overpower the transmitter; however, the transmitter adapts its own power.
This work was done by Hong Huang, Nihal Ahmed, and Santhosh Pullurul of New Mexico State University for the Army Research Laboratory. ARL-0130
This Brief includes a Technical Support Package (TSP).
Low-Power Distributed Jammer Network
(reference ARL-0130) is currently available for download from the TSP library.
Don't have an account? Sign up here.