Transport protocol for smart grid infrastructure

There is an emerging class of applications in which there is a need to reliably transport data from a large number of low rate devices (e.g., hundreds of thousands) to a central server. Atop the list of such applications is smart grid infrastructure. Other applications include meteorological applications, such as monitoring of weather, pollution, and allergy conditions. These applications are characterized by the existence of a large number of small Internet Protocol (IP) devices sending small packets at a low rate to the same server. Even though individual data rates are low, aggregation of these rates produce a significant amount of traffic that can disrupt the communication function of other applications. Because individual devices produce data at a low rate, the transmission control protocol's (TCP) congestion control mechanism is rendered ineffective, thereby causing (i) excessive retransmission of packets; and (ii) degrading the throughput of other competing applications. To address the above problems, we introduce the concept of aggregation at the communication transport level. A TCP aggregator node is added between the data sources and the central server. It splits the TCP connections between the sources (e.g., smart meters) and the data collection server. The aggregator collects data from the individual sources and reliably forwards the data to the collection server. The new strategy provides better response to traffic conditions and, most importantly, makes congestion control and flow control useful. Through extensive NS-2 simulations, we show the effectiveness of the TCP aggregator approach to mitigating the problems in terms of the throughput and packet drop rate performance metrics.