The focus of this paper is to analyze the energy consumption performance of
various versions of TCP, namely, Tahoe, Reno and NewReno, for bulk data tr
ansfer in an environment where channel errors are correlated. We investigat
e the performance of a single wireless TCP connection by modeling the corre
lated packet loss/error process (e.g., as induced by a multipath fading cha
nnel) as a first-order Markov chain. Based on a unified analytical approach
, we compute the throughput and energy performance of various versions of T
CP. The main findings of this study are that (1) error correlations signifi
cantly affect the energy performance of TCP (consistent with analogous conc
lusions for throughput), and in particular they result in considerably bett
er performance for Tahoe and NewReno than iid errors, and (2) the congestio
n control mechanism implemented by TCP does a good job at saving energy as
well, by backing off and idling during error bursts. An interesting conclus
ion is that, unlike throughput, the energy efficiency metric may be very se
nsitive to the TCP version used and to the choice of the protocol parameter
s, so that large gains appear possible.