| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
|
The max allowed RoE latency limit depends on how IEEE 1914.1 splits the functional layers between the radio unit and the central station. I believe placing
the HARQ processing at the radio unit is one of the options that will be considered. This would mean that something other than HARQ could set the max RoE latency limit. Kevin, I’m not sure why this limit would affect the presentation time offset. Our current timestamp field is 32-bits, which covers 999,999,999.75ns and is
well beyond any possible RoE latency limit. However, this field cannot be reduced in size without making it incompatible with all existing time-of-day standards, which roll over at each integer second. Rich From: stds-1904-3-tf@xxxxxxxx [mailto:stds-1904-3-tf@xxxxxxxx]
On Behalf Of Akhter, Mohammad Hi Kevin and Jouni, Although, HARQ latency in LTE based system is as you mention, this is expected to be reduced as we move towards 5G (e.g., could go towards 1msec). In addition,
smaller cell size is expected to co-exist and this will reduce the link traversal time for such smaller cells. Assuming the specification is forward looking, it would be good to have profile based latency targets, e.g., profile for LTE and legacy system could stay with
1 msec one way latency, whereas, profile for 5G and small cell could target lower latency. Regards,
-
mohammad Mohammad Akhter | Director, Architecture | Direct: 1 (613) 595 6279 | Mobile: 1 (613) 286 1840 |
www.idt.com From:
stds-1904-3-tf@xxxxxxxx [mailto:stds-1904-3-tf@xxxxxxxx]
On Behalf Of Jouni Korhonen Kevin, Within the current mappers and use cases <1ms one way delay should be just fine. I think we can assume that in this version of the to-be-spec safely. Jouni
On Mar 28, 2016, at 3:09 PM, Bross, Kevin <kevin.bross@xxxxxxxxx> wrote:
|