Bandwidth-aware distributed ad-hoc grids in deployed wireless sensor networks

摘要

Nowadays, cost effective sensor networks can be deployed as a result of a plethora of recent engineeringudadvances in wireless technology, storage miniaturisation, consolidated microprocessor design, andudsensing technologies.udWhilst sensor systems are becoming relatively cheap to deploy, two issues arise in their typicaludrealisations: (i) the types of low-cost sensors often employed are capable of limited resolution and tendudto produce noisy data; (ii) network bandwidths are relatively low and the energetic costs of using theudradio to communicate are relatively high. To reduce the transmission of unnecessary data, there is audstrong argument for performing local computation. However, this can require greater computationaludcapacity than is available on a single low-power processor. Traditionally, such a problem has beenudaddressed by using load balancing: fragmenting processes into tasks and distributing them amongst theudleast loaded nodes. However, the act of distributing tasks, and any subsequent communication betweenudthem, imposes a geographically defined load on the network. Because of the shared broadcast nature ofudthe radio channels and MAC layers in common use, any communication within an area will be slowed byudadditional traffic, delaying the computation and reporting that relied on the availability of the network.udIn this dissertation, we explore the tradeoff between the distribution of computation, needed to enhanceudthe computational abilities of networks of resource-constrained nodes, and the creation of networkudtraffic that results from that distribution. We devise an application-independent distribution paradigm anduda set of load distribution algorithms to allow computationally intensive applications to be collaborativelyudcomputed on resource-constrained devices. Then, we empirically investigate the effects of networkudtraffic information on the distribution performance. We thus devise bandwidth-aware task offload mechanismsudthat, combining both nodes computational capabilities and local network conditions, investigateudthe impacts of making informed offload decisions on system performance.udThe highly deployment-specific nature of radio communication means that simulations that areudcapable of producing validated, high-quality, results are extremely hard to construct. Consequently, toudproduce meaningful results, our experiments have used empirical analysis based on a network of motesudlocated at UCL, running a variety of I/O-bound, CPU-bound and mixed tasks. Using this setup, we haveudestablished that even relatively simple load sharing algorithms can improve performance over a range ofuddifferent artificially generated scenarios, with more or less timely contextual information. In addition,udwe have taken a realistic application, based on location estimation, and implemented that across the sameudnetwork with results that support the conclusions drawn from the artificially generated traffic.