., shielded, lower loss) can be considered for 40m Discussion on Cable length • In terms of accessories – is there data on what accessories are needed on different type of automobiles (bus, trucks, cars, vans) – Cameras were shown in the presentation (more cameras on bus as opposed to Vans), as for other accessories – need input from Daimler • Are buses the only use case for 40m – No – Buses and trucks • Is it possible to have a intermediatary intelligence system in between for the 40m long haul – No Intermission • 12PM – Team departed for lunch • 1:30PM – Team reconvened Wake up for Automotive Communication Networks • Presenter: Thomas Hogenmuller • Summary – Different Terminal control (eg on/off , on/sleep mode etc.,) – Sleep mode current consumption considerations – Wake up requirements based on back over avoidance – Need for Fast wake‐up mechanism to guarantee 100ms link acquisition time – Need for Reliable wake up – Need for power efficient sleep mode – Need to solve automotive power voltage issue (12V) Tutorial on Lifetime Requirements and Physical Testing for Automotive ECUs • Presenter – Thomas Hagenmuller • Summary ‐ Typical ECU requirements for ECUs used in engine compartment (Presentation is not a complete summary of requirements) • Discussions – ECU to subdevices has no isolation, there are some ESD requirements though – Is the bus connection star or a single bus with multiple drops on it (this might define the requirement on multi‐drop PoE) • It is single bus with multiple drops on it Technical feasibility of Gigabit transmission on one or two pair cabling based on category 6a technology • Presenter : Richard Mei • Chair asked for permission to present updated presentation on “Technical feasibility of Gigabit transmission on one or two pair cabling based on category 6a technology” ‐ None • Summary – Test results were presented {Test results covered 1‐pair/2‐pair:8m,3 connectors; 12m, 5 connectors;40m,5 connectors} – PHY feasibility study – Mode conversion data presented – Both 1 pair and 2 pair systems are technically feasible, this data suggests that number of pairs will be determined by other parameters • Discussion – Connector used is not automovite connector {Coming up with the connector is not part of RTPGE SG} – Feasibility of physically solving the problem, because the worst case scenario of aggressors (6 around 1) where all aggressors are same type, might be corner case {aggressors might be of different type} – Twisted pairs in cable should be kept together – Jacketed twisted pairs were investigated – bare twisted pairs were not (for 6 around 1 testing) – Measured impedance of the differential pair is ~103Ohms – Aging, mechanical reliability not considered in testing – commerically available cat6a cables were used for testing PHY Feasibility study for one or two pairs RTPGE • Presenter:Joseph Chou, Benson Huang • Few backup slides are new – chair asked if the team has opposition against Realtek using this in the presentation if need be – None • Realtek will upload new slide set with these backup slides • Summary – Cable model from Commscope – Salz SNR for performance Evaluation – Alien crosstalk dominates overall noise – Bidirectional signal on each twisted pair assumed for comparison purposes – 1pair Vs 2pair comparison at PAM‐4 and no NEXT/FEXT cancellation was shown • 2‐pair has better SNR margin compared to 1‐pair without alien NEXT/FEXT • 1‐pair solution requires further study on channel coding to enhance SNR margin – Complexity comparison between 802.3ab,2 pair RTPGE and 1 pair RTPGE presented • 1 pair is comparable to 2 pair – Both 1 and 2 pairs are technically feasible – 1 pair 40 meter deserves further study of performance impact caused by environment Discussions on PHY Feasibility Study from Realtek • 40nm technology or shrink more?