From dj.johnston@INTEL.COM Fri Apr 23 10:41:57 2004 Date: Thu, 22 Apr 2004 15:33:02 -0600 From: "Johnston, Dj" To: l.napoli@ieee.org Subject: Re: [STDS-802-16] Broadcast The issue of data broadcast needs to be understood in the context of what CS service you are providing. Broadcast data (as opposed to broadcast management) is a feature of the CS service. So for example 802.1 defines broadcast behaviour within the context of 802. This is not completely trivial, the behaviour of 802 broadcast data on an 802 network is subject to a number of rules. The behavior at the 802.16 MAC CPS layer is the multicast service. At present it is the job of the CS to map between the CS broadcast service and the MAC CPS multicast service. I submitted a few comments about the need to clarify the CS broadcast behavior in order to make both ends of the link behave the same. So if a CS is presenting and 802 broadcast service, it needs to know who is in the 802 broadcast group and set up a multicast group to all these SSs. You may be doing something different to the default. There may be VLANs or provider bridging or distinct, isolated 802 networks being served to users. So in 802.16, broadcast data is not broadcast data. It is a specially managed aspect of the MAC CPS multicast service to provide an equivalent to a CS layer broadcast service. If we want to go down a route of defining broadcast data in the MAC CPS, then we need to have a reason. E.G. A MAC CPS layer broadcast service will enable CS layer broadcast service with lower on air management messaging or A MAC CPS layer broadcast service will prevent redundant specification of broadcast behaviour in multiple CS specs. I don't see that either of these is true. Therefore I don't see a need for a broadcast CID on the downlink. On the uplink, there is no broadcast. Data goes only to the BS, since that is what the PHY layer supports. The CS is free to create a multicast CID to provide that function. So I see no need to a broadcast CID in the MAC CPS. DJ -----Original Message----- From: owner-stds-802-16@listserv.ieee.org [mailto:owner-stds-802-16@listserv.ieee.org] On Behalf Of Ken Stanwood Sent: Wednesday, April 21, 2004 8:48 AM To: STDS-802-16@listserv.ieee.org Subject: Re: [STDS-802-16] Broadcast The basic situation is this: 1. The "Broadcast" connection that currently exists was intended for MAC usage. There is now way to indicate a user service over it. 2. Multicast polling groups are just that, polling groups, not connections. The "CID" is a replacement for the SS's Basic CID which is actually used in maps a a shortened SS ID rather than really a connection ID. 3. Terminals are unaware whether a connection is multicast or unicast, with the exception of it having a different security association. 4. A broadcast DL connection would be established today by setting up a connection with the same parameters to all SS's. It's data would need to be transmitted in a region that all SS's could listen to, or be re-broadcast to sets of SS's if no such region existed as could possibly happen with AAS or DL OFDMA. We should consider whether a true broadcast well-known connection would be useful. If so, I recommend putting it in 16e since it's an extra feature. Ken ________________________________________________________________________________ From: owner-stds-802-16@listserv.ieee.org [mailto:owner-stds-802-16@listserv.ieee.org] On Behalf Of Pedro Neves Sent: Wednesday, April 21, 2004 4:42 AM To: STDS-802-16@listserv.ieee.org Subject: Re: [STDS-802-16] Broadcast The draft does not specify one broadcast and one multicast connection for sending data from the BS to the SSās? As Vladimir said, multicast polling groups are used, but what about data transfer from the BS to the SSās? I thought CIDs were allocated for this type of transmissions. Regards, Pedro Neves -------------------------------------------------------------------- Pedro Miguel Naia Neves Instituto Telecomunica^Ķ^Ūes - http://www.av.it.pt Aveiro - Portugal Phone: +351 234 377 900 Mobile: +351 96 618 75 82 Homepage: http://daidalos.av.it.pt/~pneves MSN contact: etneves@hotmail.com ________________________________________________________________________________ From: owner-stds-802-16@listserv.ieee.org [mailto:owner-stds-802-16@listserv.ieee.org] On Behalf Of Vladimir Yanover Sent: quarta-feira, 21 de Abril de 2004 12:06 To: STDS-802-16@listserv.ieee.org Subject: Re: [STDS-802-16] Broadcast Eyal, Seems that the standard does not preclude from sending data over broadcast connection. Another question is whether we may establish a service flow associated with broadcast connection. I think, we cannot, then there is no way for data entering CS SAP to be routed to broadcast connection. By the way, multicast connections are not related to multicast [polling] groups. Vladimir -----Original Message----- From: Eyal Verbin [mailto:everbin@AIRSPAN.COM] Sent: Wednesday, April 21, 2004 1:39 PM To: STDS-802-16@LISTSERV.IEEE.ORG Subject: [STDS-802-16] Broadcast Section 6.1 of the standard states that " In addition to individually addressed messages, messages may also be sent on multicast connections (control messages and video distribution are examples of multicast applications) as well as broadcast to all stations." Correct me if I'm wrong, but broadcast transmission is limited to MAC management messages (MAPs, DCD,...) and can't be used to transfer data. Therefore, the only way to broadcast data is to form a multicast group containing all SS's Eyal -----Original Message----- From: owner-stds-802-16@listserv.ieee.org [mailto:owner-stds-802-16@listserv.ieee.org]On Behalf Of Don Leimer Sent: Monday, April 19, 2004 8:26 PM To: STDS-802-16@listserv.ieee.org Subject: Re: [STDS-802-16] Clarification regarding SS power level control Only one more comment. The final 4dB of error will also be reduced by subsequent BS commands, and relative error diminishes to +/- 0.5dB for the final error (relative to the BS's capability to measure power) -----Original Message----- From: Raja Banerjea [mailto:RBanerjea@PROXIM.COM] Sent: Monday, April 19, 2004 9:14 AM To: STDS-802-16@LISTSERV.IEEE.ORG Subject: Re: [STDS-802-16] Clarification regarding SS power level control The power control method is a closed loop method where the Base station asks for further power control corrections if required. If the base station requests the subscriber station in the RNG-RESP to increase the power level by 30dB the SS should increase it by 30dB with a relative accuracy of 4dB. If the Base station is going to increase the power of the SS in 5 steps and the BS requests the SS to increase the power by 8dB the SS will increase it by 8dB with a relative accuracy of 4dB. In the subsequent RNG-RESP message the BS instead of requesting a power increase of 8dB will request for 8dB+(relative accuracy). Therefore after each increase requested from the BS the relative accuracy should be 4dB. This assumes that the BS can make an accurate measurement of the SS's power increase. Any comments ? Regards, -Raja -----Original Message----- From: Crozier, Eugene [mailto:Eugene_Crozier@SRTELECOM.COM] Sent: Monday, April 19, 2004 6:26 AM To: STDS-802-16@listserv.ieee.org Subject: Re: [STDS-802-16] Clarification regarding SS power level control My understanding of this is that the step size should be greater than 1 dB but less than 8 dB (I'd assumed for the relative accuracy that the 50% of the step size can be no more than 4 dB), but the number of steps is based on the step size and the relative accuracy to achieve the minimum control range, so for 1 dB steps, the number of steps can be between 60 and 20 (30/0.5 and 30/1.5) for a 30 dB range, and for 8 dB step size the number of steps between 8 and 3 for the 30 dB range. Regards Eugene Crozier -----Original Message----- From: Eyal Verbin [mailto:everbin@AIRSPAN.COM] Sent: Monday, April 19, 2004 8:22 AM To: STDS-802-16@listserv.ieee.org Subject: [STDS-802-16] Clarification regarding SS power level control Power level control for the OFDM PHY is defined in section 8.3.9.1: " For an SS not supporting subchannelization, the transmitter shall support a monotonic power level control of 30 dB minimum. For an SS supporting subchannelization, the transmitter shall support a monotonic power level control of 50 dB minimum. The minimum step size shall be no more than 1 dB. The relative accuracy of the power control mechanism is +/-50% of the step size in dB, but no more than 4 dB. As an example, for a step size of 5 dB the relative accuracy is 2.5 dB. For a BS, the transmitter shall support a monotonic power level control of 10 dB minimum." Looking at the SS (subchannelization) for example, it is possible to go from Min power to Max power either in 5 steps of 8 dB or in a single step of 50dB. In the first option the accumulated offset can reach 5*4dB (20dB) wheras in the second option the tolerance is limited to 4dB. Does anyone have a more clear interpretation of this text? Eyal Verbin This mail passed through mail.alvarion.com ******************************************************************************** **** This footnote confirms that this email message has been scanned by PineApp Mail-SeCure for the presence of malicious code, vandals & computer viruses. ******************************************************************************** **** This mail was sent via mail.alvarion.com ******************************************************************************** **** This footnote confirms that this email message has been scanned by PineApp Mail-SeCure for the presence of malicious code, vandals & computer viruses. ******************************************************************************** **** ^@