A forward planning situated protocol for data propagation in wireless sensor networks based on swarm intelligence techniques

• Authors: Ioannis Chatzigiannakis;Sotiris Nikoletseas
• Venue: Proceedings of the seventeenth annual ACM symposium on Parallelism in algorithms and architectures
• Year: 2005

Wireless sensor networks are comprised of a vast number ofultra-small fully autonomous computing, communication and sensingdevices, with very restricted energy and computing capabilities,which co-operate to accomplish a large sensing task. Such networkscan be very useful in practice in applications that requirefine-grain monitoring of physical environment subjected to criticalconditions (such as inaccessible terrains or disaster places).Features including the huge number of sensor devices involved, thesevere power, computational and memory limitations, their densedeployment and frequent failures, pose new design andimplementation aspects. The efficient and robust realization ofsuch large, highly-dynamic, complex, non-conventional, environmentsis a challenging algorithmic task.We here present the Forward Planning Situated Protocol (FPSP),for scalable, energy efficient and fault tolerant data propagationin situated wireless sensor networks. To deal with the increasedcomplexity of such deeply networked sensor systems, instead ofemphasizing on a particular aspect of the services provided, i.e.either for low-energy periodic, orlow-latency event-driven, orhigh-success query-based sensing, FPSP usestwo novel mechanisms that allow the networkoperator to adjust the performance of theprotocol in terms of energy, latency and success rate on a per-taskbasis. We emphasize on distributedness, direct or indirectinteractions among relatively simple agents, flexibility androbustness.The protocol operates by employing a series of plan& forward phases through which devices self-organizeinto forwarding groups that propagate data over discovered paths.FPSP performs a limited number of long range,high power data transmissions to collectinformation regarding the neighboring devices. The acquiredinformation, allows to plan a (parameterizablelong by λ) sequence of short range, low power transmissionsbetween nearby particles, based on certain optimization criteria.All particles that decide to respond (based on localcriteria) to these long range transmissions enter theforwarding phase during which information ispropagated via the acquired plan. Clearly, the duration of theforwarding phases is characterized by the parameter λ, thetransmission medium and the processing speed of the devices. Infact the parameter λ provides a mechanism to adjust theprotocol performance in terms of the latency--energytrade-off. By reducing λ the latency is reduced atthe cost of spending extra energy, while by increasing λ,the energy dissipation is reduced but the latency is increased.To control the success rate--energytrade-off, particles react locally on environment andcontext changes by using a set of rules that are based onresponse thresholds that relate individual-levelplasticity with network-level resiliency, motivated by thenature-inspired method for dividing labor, a metaphor of socialinsect behavior for solving problems [1]. Each particle has anindividual response threshold Θ that is related to the"local" density (as observed by the particle, [2]); particlesengage in propagation of events when the level of thetask-associated stimuli exceeds their thresholds. Lets be the intensity of a stimulus associated witha particular sensing task, set by the human authorities. We adoptthe response functionTθ(s)= snoversn +θn, the probabilityof performing the task as a function of s, wheren 1 determines the steepness of thethreshold. Thus, when θ is small (i.e. the network is sparse)then the response probability increases; when sincreases (i.e. for critical sensing tasks) the responseprobability increases as well.This role-based approach where a selective number of devices dothe high cost planning and the rest of the network operates in alow cost state leads to systems that have increased energyefficiency and high fault-tolerance since these long range planningphases allow to bypass obstacles (where no sensors are available)or faulty sensors (that have been disabled due to power failure orother natural events).