ACM Transactions on Embedded Computing Systems (TECS) - Special Section ESFH'12, ESTIMedia'11 and Regular Papers
| Hi-index | 0.00 |
Wireless sensor and actuator networks (WSANs) form an integral part of Cyber-Physical Systems (CPSs). Large-scale coordination and control problems in WSANs are often expressed within the networked optimization model. While significant advances have taken place in both first- and higher-order optimization techniques, their widespread adoption in practical implementations has been impeded by the complexity of inter-node coordination and lack of programming support for the same. Currently, application developers build their own elaborate coordination mechanisms for synchronized execution and coherent access to shared resources by distributed and concurrent controller processes. However, they typically tend to be error-prone and inefficient due to tight constraints on application development time and cost. This is unacceptable in many CPS applications as it can result in expensive and often irreversible side-effects in the environment due to inaccurate or delayed reaction of the control system. Subsequently, many programming and middleware frameworks have been proposed in prior literature to simplify application development, but they are either incomplete, inefficient or inapplicable with respect to support for distributed iterative optimization algorithms. This dissertation explores the design of a distributed shared memory (DSM) architecture that abstracts away the details of inter-node coordination from the programmer resulting in simplified application design. The DSM system maintains data coherency through explicit use of mutual exclusion lock primitives and supports synchronized execution through the use of barriers. Subsequently, the proposed design is realized through the implementation of Hotline Application Programming Framework which reduces programmers' burden by managing the underlying routing, messaging and discovery of nodes for transparent and consistent access to shared resources. Therefore, it not only enables rapid implementation of distributed optimization algorithms, but dramatically increases their performance through the design of DSM customized to the spatial locality inherent in actuation. This drives faster and scalable application execution through opportunistic data parallel operation and automatic formation of optimal information exchange graphs that are derived from the contextual relationship between sensors, actuators and controllers. The results demonstrate a 75% reduction in lines of code and 50-99% reduction in application latency depending on the algorithm, network and deployment characteristics. Moreover, the proposed coordination mechanisms in Hotline are shown to be correct, thereby resulting in error-free implementation of algorithms.