Network Working Group                                             D. Liu
Internet-Draft                                              China Mobile
Intended status: Informational                                   J. Song
Expires: September 8, 2011                                        W. Luo
                                                                     ZTE
                                                           March 7, 2011


     Distributed Mobility Management Handover Frequentness Analysis
          draft-liu-distributed-mobility-handover-analysis-00

Abstract

   This document analysis the handover frequentness in distributed
   anchor scenario.

Status of this Memo

   This Internet-Draft is submitted in full conformance with the
   provisions of BCP 78 and BCP 79.

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF).  Note that other groups may also distribute
   working documents as Internet-Drafts.  The list of current Internet-
   Drafts is at http://datatracker.ietf.org/drafts/current/.

   Internet-Drafts are draft documents valid for a maximum of six months
   and may be updated, replaced, or obsoleted by other documents at any
   time.  It is inappropriate to use Internet-Drafts as reference
   material or to cite them other than as "work in progress."

   This Internet-Draft will expire on September 8, 2011.

Copyright Notice

   Copyright (c) 2011 IETF Trust and the persons identified as the
   document authors.  All rights reserved.

   This document is subject to BCP 78 and the IETF Trust's Legal
   Provisions Relating to IETF Documents
   (http://trustee.ietf.org/license-info) in effect on the date of
   publication of this document.  Please review these documents
   carefully, as they describe your rights and restrictions with respect
   to this document.  Code Components extracted from this document must
   include Simplified BSD License text as described in Section 4.e of
   the Trust Legal Provisions and are provided without warranty as
   described in the Simplified BSD License.




Liu, et al.             Expires September 8, 2011               [Page 1]


Internet-Draft            DMM Handover Analysis               March 2011


Table of Contents

   1.  Introduction . . . . . . . . . . . . . . . . . . . . . . . . .  3
   2.  Conventions used in this document  . . . . . . . . . . . . . .  3
   3.  Inter-GW Handover Frequentness Analysis  . . . . . . . . . . .  3
     3.1.  The Trend of the Inter-GW Handover Frequentness  . . . . .  4
   4.  The impact of non-optimized Routing  . . . . . . . . . . . . .  4
     4.1.  Analysis model . . . . . . . . . . . . . . . . . . . . . .  5
     4.2.  Traffic load Analysis  . . . . . . . . . . . . . . . . . .  7
     4.3.  Congestion Probability Analysis  . . . . . . . . . . . . .  8
     4.4.  Delay Analysis . . . . . . . . . . . . . . . . . . . . . .  8
   5.  Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . .  9
   6.  Security Considerations  . . . . . . . . . . . . . . . . . . . 10
   7.  IANA Considerations  . . . . . . . . . . . . . . . . . . . . . 10
   8.  Contributors . . . . . . . . . . . . . . . . . . . . . . . . . 10
   9.  References . . . . . . . . . . . . . . . . . . . . . . . . . . 10
     9.1.  Normative References . . . . . . . . . . . . . . . . . . . 10
     9.2.  Informative References . . . . . . . . . . . . . . . . . . 10
   Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 11
































Liu, et al.             Expires September 8, 2011               [Page 2]


Internet-Draft            DMM Handover Analysis               March 2011


1.  Introduction

   With the Gateway in the cellular telecom core network moving
   downwards towards the access network and being deployed
   distributively, the number of gateways is increasing.  Consequently,
   an increasing the percentage of handovers will be of inter-gateway
   handovers among all the handovers is also rising.  Due to the fixed
   anchor point used by the current mobility management, the route in
   the access network is not optimized, ; that means there is roundabout
   in the access network.  Such roundabout problems will aggravate with
   the increase of inter-gateway handovers.


2.  Conventions used in this document

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL","SHALL NOT",
   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
   document are to be interpreted as described in [RFC2119].


3.  Inter-GW Handover Frequentness Analysis

   We use a real deployment case in a metro of China as an example for
   the analysis.  There are seventeen RNC nodes covering the metro.  If
   we presume the Gateways are moved downwards to the level of RNC, we
   can calculate the frequentness of inter-GW handover based on that of
   inter-RNC handover.

   The frequentness of inter-RNC handover:

   In certain time period of j, relocNj is the total numbers of inter-
   RNC handover: I relocNj = SUM(CS relocation number i + PS relocation
   number i) i=1

   In certain time period of j, rabNj is the total number of PDP
   activation and call establishment in RNCs: I rabNj = SUM(CS call
   number i + PS call number i) i=1

   Rj is the frequentness of handover in this time period: Rj = relocNj
   / rabNj

   R is the mean handover frequentness in all the time periods: J R =
   (SUM Rj)/ J *100% j=1

   Notes:

   1) The CS calls and PS calls are counted based on the successful RAB
   establishment



Liu, et al.             Expires September 8, 2011               [Page 3]


Internet-Draft            DMM Handover Analysis               March 2011


   2) Both CS relocation and PS relocation include types of UE involved
   and UE not involved for handover out

   3) I is the number of RNC in the metro

   4) J is the number of time period, which could be one hour or more.

   The statistics shows that the mean frequentness of inter-RNC handover
   in one hour is 12.32%, that is there are 12 times inter-RNC handover
   per 100 times of CS calls or PS PDP activation.  The highest handover
   frequentness could reach 15.97%.  If the gateways are moved to the
   level of RNC, then the frequentness of inter-GW handover should be
   equivalent to that of inter-RNC.

3.1.  The Trend of the Inter-GW Handover Frequentness

   The statistics result above is based on the current deployed network.
   With the evolution of the mobile network, e.g. when the LTE is
   deployed, we can expect that inter-GW handover frequentness will
   increase, when comparied with current deployed network.

   a.  With the increment of the traffic and the user amount, the
   requirement of the bandwidth goes up consequently.  To meet this
   requirement, the density of the basestation shall increase.
   Meanwhile, with the consideration of the limited capacity and
   capability of the GWs, the density of the GWs should also increase.
   So, in a same coverage area, there will be more GWs.  As a result,
   the inter-GW handover frequentness will increase.

   b.  With the service evolution, the rate of the always online user
   will increase.  Accordingly, the rate of the service which is started
   (e.g.  PDP activation) and completed (e.g.  PDP inactivation) under a
   same GW will drop.  As a result, the inter-GW handover frequentness
   will increase.

   We expect that in the near future, the value of the inter-GW handover
   frequentness may be doubled or even tripled to the value of current
   frequentness.


4.  The impact of non-optimized Routing

   When fixed anchor point is used, the problems of roundabout will
   worsen with the movement of the user.  Especially, when service is
   deployed far from the anchor point and near the location where the
   user is moving to.  When the anchor point locates at a different
   city, the impacts of roundabout become more serious, that will
   inevitably affect the customer experience.  In the following



Liu, et al.             Expires September 8, 2011               [Page 4]


Internet-Draft            DMM Handover Analysis               March 2011


   chapters, the impacts are appraised with a traffic model from the
   aspects of traffic volume, congestion and time delay, etc.

4.1.  Analysis model

   We use the following model for the handover analysis.  In this model,
   the anchor is distributed in the network edge.  The communication
   peer may be located in the backbone or in the metro network.



                   +----------------------+
                   |       Backbone       |
                   |+-------+ +-------+   |   +-----+
                   ||RouterB|-|RouterB|---|---| Peer|
                   |+-------+ +-------+   |   +-----+
                   +----------------------+
                     |   |           |   |
                     |   |           |   +---------------+
             /-------+   |           +-----+             |
    +-------|------------|-------+ +-------|-------------|------+
    |       |            |       | |       |             |      |
    |   +-------+    +-------+   | |  +-------+      +-------+  |
    |   |RouterA|----|RouterA|   | |  |RouterA|------|RouterA|  |
    |   +-------+    +-------+   | |  +-------+      +-------+  |
    |         \         /        | |        \           /       |
    |         ,-'' -------+      | |        ,-'' -------+       |
    |       /              \     | |       /              \     |
    |      /       Metro    \    | |      /       Metro    \     +-----+
    |      |                |    | |      |                |-----|Peer |
    |      \    Network A  ,'    | |      \    Network B  ,'     +-----+
    |       `.           _,      | |       `.           _,      |
    |        /  `-..____,,' \    | |        /  `-..____,,' \    |
    |       /                \   | |       /                \   |
    |    +------+       +------+ | |    +------+       +------+ |
    |    |P-GW1 |       | P-GW2| | |    | P-GW |       | P-GW | |
    |    |S-GW1 |       | S-GW2| | |    | S-GW |       | S-GW | |
    |    +------+       +------+ | |    +------+       +------+ |
    |                            | |                            |
    +----------------------------+ +----------------------------+




           Figure 1: Distributed anchor handover analysis model

   When the UE attaches to the network through S-GW2, the forwarding
   path is illustrated in the following figures.



Liu, et al.             Expires September 8, 2011               [Page 5]


Internet-Draft            DMM Handover Analysis               March 2011


   For the case that the communication peer is located in backbone, the
   forwarding path is illustrated as the following figure.



     +----+   +-----+   +------+   +-------+   +-----------------+
     |UE  |-->|P-GW2|   |      |   |       |   |                 |
     |    |   |S-GW2|-->|Metro |-->|RouterA|-->|Metro to Backbone|
     +----+   +-----+   +------+   +-------+   +-----------------+
                                                                 |
                                                  +-------+   +-------+
                                                  |  Peer |<--|RouterB|
                                                  +-------+   +-------+


    Figure 2: Forwarding path when peer is located in backbone network

   For the case that the peer is located in metro network, the
   forwarding path is illustrated as following figure.



                  +----+   +-----+   +------+   +-------+
                  |UE  |-->|S-GW2|-->|Metro |-->| Peer  |
                  |    |   |P-GW2|   |      |   |       |
                  +----+   +-----+   +------+   +-------+



      Figure 3: Forwarding path when peer is located in metro network

   Whe the UE moves to S-GW1's serving area and handover to S-GW1 from
   S-GW2, the data forwarding path is illustrated as the following
   figures.  In this scenario, if we use current mobility solution, the
   anchoring point should be still at P-GW2.

   For the case that the peer is located in backbone network, the
   forwarding path is illustrated as following figure.



     +----+   +-----+      +------+      +-------+        +------ +
     | UE |-->|S-GW1|----->|Metro |----->| P-GW2 |------->|Metro  |
     +----+   +-----+      +------+      +-------+        +-------+
                                                                 |
                  +----+   +-------+   +------------------+   +-------+
                  |Peer|<--|RouterB|<--| Metro to backbone|<--|RouterA|
                  +----+   +-------+   +------------------+   +-------+



Liu, et al.             Expires September 8, 2011               [Page 6]


Internet-Draft            DMM Handover Analysis               March 2011


    Figure 4: Forwarding path when peer is located in backbone network
                               after handove

   For the case that the peer is located in metro network, the
   forwarding path is illustrated as following figure.



           +----+  +-----+  +------+  +------+  +------+  +----+
           |UE  |->|S-GW1|->|Metro |->| P-GW2|->|Metro |->|Peer|
           |    |  |     |  |      |  |      |  |      |  |    |
           +----+  +-----+  +------+  +------+  +------+  +----+



      Figure 5: Forwarding path when peer is located in metro network

   It is assumed that the traffic to the backbone network is 60% of the
   total traffic and the traffic to metro network is 40% of the total
   traffic.

4.2.  Traffic load Analysis

   Based on the above handover model, the un-optimized routing is
   summaried in the following table.



    +------------------------------------------------------------------+
    |Traffic | peer in Backbone | peer in Metro    |      average      |
    |------------------------------------------------------------------|
    |Before  |                  |                  |                   |
    |handover|1 volume within   |1 volume within   |1 volume within    |
    |        |metro.            |metro.            |metro.             |
    |        |1 volume between  |0 volume between  |0.6 volume between |
    |        |metro and backbone|metro and backbone|metro and backbone |
    |        |1 volume within   |0 volume within   |0.6 volume within  |
    |        |backbone          |backbone          |backbone           |
    |------------------------------------------------------------------+
    |After   |2 volume within   |2 volume within   |2 volume within    |
    |handover|Metro.            |metro.            |metro              |
    |        |1 volume between  |0 volume between  |0.6 volume between |
    |        |metro and backbone|metro and backbone|metro and backbone |
    |        |1 volume within   |0 volume within   |0.6 volume within  |
    |        |backbone          |backbone          |backbone           |
    +------------------------------------------------------------------+





Liu, et al.             Expires September 8, 2011               [Page 7]


Internet-Draft            DMM Handover Analysis               March 2011


        Figure 6: Traffic Analysis result for non-optimized routing

   From this analysis result, it can be concluded that the non-optimized
   routing will result in 100% traffic load increase in metro network.

4.3.  Congestion Probability Analysis

   We assume that the congestion probability in metro network is X, the
   congestion probability between metro network and backbone is Y, based
   on the above analysis model, we have the following result.




      +-------------------------------------------------------------+
      |congestion |peer in   |             |                        |
      |probability|backbone  |peer in metro|     average            |
      |-------------------------------------------------------------|
      |Before     |          |             |0.6*[1-(1-X)*(1-Y)]+0.4*|
      |handover   |1-(1-X)*  |     X       |X                       |
      |           |   (1-Y)  |             |                        |
      |           |          |             |                        |
      |-----------+----------+-------------+------------------------+
      |After      |          |             |0.6*[1-(1-X)(1-X)*(1-Y)]|
      |handover   |1-(1-X)*  |             |                        |
      |           |(1-X)*(1-Y)|1-(1-X)(1-X)|+0.4*[1-(1-X)*(1-X)]    |
      +-------------------------------------------------------------+




    Figure 7: Congestion Probability Analysis result for non-optimized
                                  routing

   From the above result,we can conclude that the congestion probability
   will increase after handover.  If X=3%, Y=3%, the congestion
   probability after handover will increase 2.86%.

4.4.  Delay Analysis

   The delay from UE to peer consists of three parts:delay within metro
   network: T1, the delay between metro to backbone:T2, the delay within
   backbone:T3.  Based on the above model, we have the following
   analysis result.







Liu, et al.             Expires September 8, 2011               [Page 8]


Internet-Draft            DMM Handover Analysis               March 2011


      +--------------------------------------------------------------+
      |             |                 |               |              |
      |  Delay      | Peer in Backbone| peer in Metro |Average       |
      |--------------------------------------------------------------|
      |Before       |                 |               |              |
      |handover     | T1+T2+T3        |      T1       |T1+0.6(T2+T3) |
      |--------------------------------------------------------------+
      |After        |                 |               |2T1+0.6(T2+T3)|
      |handover     | 2*T1+T2+T3      |      2*T1     |              |
      +--------------------------------------------------------------+



         Figure 8: Delay Analysis result for non-optimized routing

   From the analysis result, we can conclude that the delay within metro
   will increase 100% due to the non-optimized routing after handover.


5.  Conclusion

   According to the inter-gateway handover frequentness calculation
   model and based on the data gathered from the deployed network, an
   average of 12% of service sessions perform inter-gateway handover
   (e.g. handover from S-GW2 to S-GW1) in every survey cycle.  The
   inter-gateway handover leads to non-optimal routing and brings some
   impactions as following:

   Traffic load: referring to figure 6, a lot of wasted volume deliveres
   in the metro network.  The operator's valuable bearer resource is
   wasted.  According to the Global Mobile Data Traffic Forecast by
   CSISO, in 2015 the global mobile data traffic will be 6.3EB.  That
   means the wasted volume will be 0.75EB in 2015.

   Congestion: Many mobile users will suffer from a higher congestion
   probability than the average.  The increased congestion probability
   is X*(1-X)*(1-0.6X), in which the X is the average congestion
   probability in metro network and the Y is the average congestion
   probability between the metro network and backbone network.  For
   example, when X is 10%, Y is 10%, then the average congestion
   probability is 15.4%.  Then these mobile users have to bear the
   congestion probability of 23.86%.

   Delay: Many users will suffer from much more traffic delay.  The
   increased traffic delay is one time of the average delay in the metro
   network.





Liu, et al.             Expires September 8, 2011               [Page 9]


Internet-Draft            DMM Handover Analysis               March 2011


6.  Security Considerations

   TBD


7.  IANA Considerations

   None


8.  Contributors

   The following people contributed to this document (in no specific
   order):

      Hong Jiang
      ZTE
      Jiang.Hong3@zte.com.cn

      ZhiHai Wang
      ZTE
      Wang.Zhihai@zte.com.cn


9.  References

9.1.  Normative References

   [RFC2119]  Bradner, S., "Key words for use in RFCs to Indicate
              Requirement Levels", BCP 14, RFC 2119, March 1997.

9.2.  Informative References

   [I-D.chan-netext-distributed-lma]
              Chan, H., Xia, F., Xiang, J., and H. Ahmed, "Distributed
              Local Mobility Anchors",
              draft-chan-netext-distributed-lma-03 (work in progress),
              March 2010.

   [I-D.ietf-mext-flow-binding]
              Tsirtsis, G., Soliman, H., Montavont, N., Giaretta, G.,
              and K. Kuladinithi, "Flow Bindings in Mobile IPv6 and NEMO
              Basic Support", draft-ietf-mext-flow-binding-11 (work in
              progress), October 2010.

   [I-D.kassi-mobileip-dmi]
              Kassi-Lahlou, M., "Dynamic Mobile IP (DMI)",
              draft-kassi-mobileip-dmi-01 (work in progress),



Liu, et al.             Expires September 8, 2011              [Page 10]


Internet-Draft            DMM Handover Analysis               March 2011


              January 2003.

   [I-D.seite-netext-dma]
              Seite, P. and P. Bertin, "Dynamic Mobility Anchoring",
              draft-seite-netext-dma-00 (work in progress), May 2010.

   [RFC3775]  Johnson, D., Perkins, C., and J. Arkko, "Mobility Support
              in IPv6", RFC 3775, June 2004.

   [RFC5213]  Gundavelli, S., Leung, K., Devarapalli, V., Chowdhury, K.,
              and B. Patil, "Proxy Mobile IPv6", RFC 5213, August 2008.

   [RFC5648]  Wakikawa, R., Devarapalli, V., Tsirtsis, G., Ernst, T.,
              and K. Nagami, "Multiple Care-of Addresses Registration",
              RFC 5648, October 2009.


Authors' Addresses

   Dapeng Liu
   China Mobile
   Unit2, 28 Xuanwumenxi Ave,Xuanwu District
   Beijing 100053
   China

   Email: liudapeng@chinamobile.com


   Jun Song
   ZTE
   No.68,Zijinghua Road, Yuhuatai District
   Nanjing 210012
   China

   Email: song.jun@zte.com.cn


   Wen Luo
   ZTE
   No.68,Zijinghua Road, Yuhuatai District
   Nanjing 210012
   China

   Email: luo.wen@zte.com.cn







Liu, et al.             Expires September 8, 2011              [Page 11]