TY - GEN
T1 - Impact of Transport Layer Protocols on Reliable Information Access in Smart Grids
AU - Shahid, Kamal
AU - Saeed, Aamir
AU - Kristensen, Thomas le Fevre
AU - Olsen, Rasmus Løvenstein
PY - 2017
Y1 - 2017
N2 - Time is critical for certain types of dynamic information (e.g. frequency control) in a smart grid scenario. The usefulness of such information depends upon the arrival within a specific frame of time, which in other case may not serve the purpose and effect controller’s performance. In this context, transport layer offers different levels of end-to-end communication services to the applications. For instance, TCP guarantees the transport of messages between two ends, however, at the cost of high end-to-end delays due to the retransmission mechanism. Whereas UDP offers minimum end-to-end delays at the cost of unreliable, best-effort data transportation service. The research question raised in this paper is thus, which is preferred for the delay-critical applications of smart grids, and to what degree of packet losses and round trip times, TCP is preferable to UDP and vice versa. The question is addressed by analyzing the performance of UDP and TCP over imperfect network conditions to show how the selection of transport layer protocol can dramatically affect controller’s performance. This analysis is based on a quality metric called mismatch probability that considers occurrence of events at grid assets as well as the information update strategy in one single metric which otherwise is not very intuitive and difficult to allow a similar useful comparison. Further, the analysis is concluded by providing a clear guide on the selection of the transport protocol to meet application requirements.
AB - Time is critical for certain types of dynamic information (e.g. frequency control) in a smart grid scenario. The usefulness of such information depends upon the arrival within a specific frame of time, which in other case may not serve the purpose and effect controller’s performance. In this context, transport layer offers different levels of end-to-end communication services to the applications. For instance, TCP guarantees the transport of messages between two ends, however, at the cost of high end-to-end delays due to the retransmission mechanism. Whereas UDP offers minimum end-to-end delays at the cost of unreliable, best-effort data transportation service. The research question raised in this paper is thus, which is preferred for the delay-critical applications of smart grids, and to what degree of packet losses and round trip times, TCP is preferable to UDP and vice versa. The question is addressed by analyzing the performance of UDP and TCP over imperfect network conditions to show how the selection of transport layer protocol can dramatically affect controller’s performance. This analysis is based on a quality metric called mismatch probability that considers occurrence of events at grid assets as well as the information update strategy in one single metric which otherwise is not very intuitive and difficult to allow a similar useful comparison. Further, the analysis is concluded by providing a clear guide on the selection of the transport protocol to meet application requirements.
KW - Mismatch Probability (mmPr); smart grids; TCP; UDP
UR - http://sites.ieee.org/isgt-europe-2017/files/2017/09/IEEE_PES_ISGT_Europe_2017_Final_programme_22Sept2017.pdf
U2 - 10.1109/ISGTEurope.2017.8260250
DO - 10.1109/ISGTEurope.2017.8260250
M3 - Article in proceeding
T3 - IEEE Power & Energy Society Innovative Smart Grid Technologies Conference (ISGT)
BT - IEEE 2017 International Conference on Innovative Smart Grid Technologies - ISGT EUROPE 2017
PB - IEEE
T2 - IEEE International Conference on Innovative Smart Grid Technologies
Y2 - 26 September 2017 through 29 September 2017
ER -