From mcdonald@ECE.NEU.EDU Thu Jul 11 12:35:19 2002 Date: Thu, 11 Jul 2002 12:30:14 -0400 (EDT) From: A. Bruce McDonald To: tudball@ECE.NEU.EDU Subject: FAMA (fwd) -------------------------------------------------------- A. Bruce McDonald f:617.373.8970 Assistant Prof. v:617.373.3028 ECE Dept., 417 Dana mcdonald@ece.neu.edu 360 Huntington Ave. www.ece.neu.edu/faculty/mcdonald Northeastern University Boston, MA 02115 -------------------------------------------------------- ---------- Forwarded message ---------- Date: Mon, 4 Mar 2002 09:59:36 -0500 From: Emre Safak To: Fang Yue Cc: A. Bruce McDonald Subject: FAMA Dear Yue, I read your paper and the first thing I noticed is the discrepancy between the results in the original FAMA paper and "The Effects of MAC Protocols in Ad Hoc Network Communication". Garcia-luna-aceves claims that FAMA is better than all the others -- including IEEE 802.11 -- but Royer et al. say otherwise. Even worse, simply using handshaking seems to exacerbate the problem. Should handshaking not improve throughput by alleviating the hidden-terminal problem? How can CSMA outperform MACA and FAMA? This seems ridiculous; I question Royer et al.'s results! CORRECTIONS pg.3: You need to define BSS and DSSS. pg.4: 1. "The throughput of the system is defined as the amount of data successfully transmitted from the source to the destination". I have never heard of this definition before! 2. Lemma 1 shows poor notation; e^6 does not mean 10^6. 3. Are you intentionally not deriving (2) and (3)? Section 4: What is the physical distribution of the nodes in the simulated network? Section 4.1: "For example, given a load of ~ 25 %, the analytical model over-estimate maximum throughput by ~ 150 kbps.". Over-ESTIMATES THE throughput ... As a general complaint, refrain from using mathematical notation like "~" in sentences when words suffice ("approximately"). "Without coupling the simulation validates that 100% throughput can be sustains to twice that load" should read "Without coupling, (COMMA) the simulation ... SUSTAINED to (or perhaps 'under') twice that load." Section 4.2: "Each of the logarithmatic ..." LOGARITHMIC "Without control mechanisms the best throughput that can be expected is ~36% form slotted-ALOHA" FROM slotted ALOHA Anticipating your thoughts, Emre.