A Comparison of Two CoMP Cell Clustering Algorithms in LTE-A Downlink Transmission

NSC 99-2221-E-009-075-MY3

Introduction Utility function

1 B CoMP

Coordinated Multi-Point (CoMP) transmission is a promising way to suppress inter-cell Sum Rate Utility (SR): J1 = Cb
B b =1
interference for fulfilling the requirements of IMT-Advanced.

The main idea of CoMP is to select several BSs which could cooperate together to B
mitigate interference. Proportional Fair Utility (PF): J 2 = ( CbCoMP )1/ B

In LTE-A downlink, there are two classes of CoMP schemes named joint transmission b =1
(JT) and coordinated beamforming (CB). B

To reduce the complexity, a linear precoder based on block diagonalization (BD) is Weighted Sum Utility (WS): J 3 = wb CbCoMP
proposed in our scheme. b =1
Simulation Parameters

Fig. 1. Illustration of JT and CB.

System Model
Simulation Results

The received signal of the kth user from the bth cell in a set G can be written as

H x j ,b + H kb,b xl ,b + nk ,b

ykG,b = H kb,b xk ,b + H kb,b xi ,b + b
k ,b
i k ,i b bG ,b b j b b G l b

where G is one of the set of selected base stations(BS) in a cluster, H kb,b is the Nr Nt
channel matrix from the BS b to the user k served by the BS b, xk,b is the Nt 1 signal
vector of user k from the BS b, and nk,b is the corresponding noise vector, each component
of which is a zero-mean complex Gaussian random variable with variance N0.
Fig. 3. Average rate versus the edge-SNR. Fig. 4. Average rate v.s. pilot SINR threshold.

.
Fig. 2. Block diagram for downlink multi-user MIMO-OFDM. Fig. 5. The CDF of the CoMP requesting users. Fig. 6. The CDF of the CoMP included users.

Algorithms

Static Clustering Algorithm

1:Specify the CoMP cluster size B;
2:Each user measures channel gains and calculates the corresponding pilot SINR
2 2
by SINRbPilot = Hbb / Hbb + N0 , b,
bb,bb Fig. 7. The CDF of the CoMP last five Fig. 8. The average rate versus different
percent users. CoMP cluster size.
where Ib is set of the first-tier interfering BSs around BS b, i.e.,
only the pilots from the neighboring BSs are regarded as the valid Conclusions
interference. If SINRbPilot < ,where is the threshold, the user asks for CoMP

CoMP is a promising way of suppressing inter-cell interference, but it requires the CSI
service from the BS b through uplink. Then BS b sends the request to its and user data sharing between BSs. In order to reduce the overhead, only a limited
distributed coordinator (DC); number of BSs form a CoMP cluster.
3:DC finds the remaining B-1 BSs which should cooperate with the BS b
A straightforward way is static clustering, but it inherits the drawback that users in the
edge of the static cluster still suffer from severe interference.
based on the clustering table, and makes them form a cluster;
The dynamic clustering method utilizes the instantaneous CSI and forms CoMP clusters
4:Go back to step 3 until all the CoMP needed users are satisfied; adaptively; there is no fixed cluster edge in the scheme and no user will always be prone
to interference. Hence, it outperforms the static one.

Nevertheless, this dynamic method demands the cost of higher computation complexity as
compared with the static one.
Dynamic Clustering Algorithm
1: Specify the CoMP cluster size B; References
pilot pilot
2: Each user calculates his SINR b . If SINR b < , then the user sends [1] 3GPP TS 36.201 V8.1.0, Evolved Universal Terrestrial Radio Access (E-UTRA); LTE Physical Layer - General Description, Nov. 2007.
[2] 3GPP2 C.S0084 v. 2.0, Ultra Mobile Broadband (UMB) Air Interface Specification, Sept. 2007.
CoMP request to the serving BS b; [3]
[4]
IEEE 802.16e-2005, Part 16: Air Interface for Fixed Broadband Wireless Access Systems, Amendment for physical and MAC layers for combined fixed and mobile operation in licensed bands, 2005.
3GPP TR 25.913 V7.0.0, Requirements for Evolved UTRA (E-UTRA) and Evolved UTRAN (E-UTRAN), June 2005.
[5] Siemens, Interference mitigation Considerations and Results on Frequency Reuse, 3GPP R1-050738, 3GPP TSG RAN WG1 Meeting #42, September 2005.
3: The central coordinator (CC) collects all the requests and chooses a CoMP [6] M. Sternad, T. Ottosson, A. Ahlen, A. Svensson, Attaining both coverage and high spectral efficiency with adaptive OFDM downlinks, in Proc. IEEE 58th Veh. Technol. Conf., vol. 4, Oct. 2003, pp.
2486-2490.
[7] Huawei, Soft Frequency Reuse Scheme for UTRAN LTE, 3GPP R1-050507, 3GPP TSG RAN WG1 Meeting #41, May 2005.
needed user who has not been chosen so far uniformly; [8] Bin Fan, Yu Qian, Kan Zheng, Wenbo Wang, A Dynamic Resource Allocation Scheme Based on Soft Frequency Reuse for OFDMA Systems, in Proc. Int. Symp. Microwave, Antenna, Propagation and
EMT Tech. for Wireless Comm. , Aug. 2007, pp. 1-4.
4: Find the remaining B - 1 BSs which maximize the utility function defined by [9]
[10]
Alcatel, Interference Coordination in new OFDM DL air interface, 3GPP R1-050407, 3GPP TSG RAN WG1 Meeting #41, May 2005.
3GPP TR 25.814 v7.1.0, Physical Layer Aspects for Evolved UTRA, Sept. 2006.
[11] Siemens, Interference Mitigation by Partial Frequency Reuse, 3GPP R1-060135, 3GPP TSG RAN WG1 Ad Hoc Meeting on LTE, Jan. 2006.
CoMP CoMP CoMP
J(C 1 , . . . ,C B ) with the user chosen in step 3, where C b is the capacity. [12]
[13]
H. Holma and A. Toskala, WCDMA for UMTS, 3rd edition, Wiley, 2002.
Renshui Zhu, Xing Zhang, Wenbo Wang, Scheduling Based Controllable Interference Coordination in OFDMA Systems, in Proc. First Internation Conference on Communications and Networking, Oct.
2006. pp. 1-5.
We let only the first-tier BSs around the selected user be the candidates, [14]
[15]
Young-June Choi, Cheol Seung Kim, Saewoong Bahk, Flexible Design of Frequency Reuse Factor in OFDMA Cellular Networks, in Proc. IEEE Int. Conf. Commun., vol. 4, June 2006, pp. 1784-1788.
Haipeng Lei, Zhang Lei, Xin Zhang, Dacheng Yang, A Novel Multi-Cell OFDMA System Structure using Fractional Frequency Reuse, in Proc. IEEE Symp. PIMRC, Sept. 2007.
[16] Texas Instruments, Performance of Inter-Cell Interference Mitigation with Semi-Static Frequency Planning for EUTRA Downlink, 3GPP R1-060368, 3GPP TSG RAN WG1 Meeting #44, February
i.e., b I b . If the number of available BSs is less than B - 1, the user cannot 2006.
[17] H. Jia, Z. Zhang, G. Yu, P. Cheng, S. Li, On the Performance of IEEE 802.16 OFDMA System Under Different Frequency Reuse and Subcarrier Permutation Patterns, in Proc. IEEE Int. Conf.
Commun., June 2007, pp. 5720-5725.
acquire CoMP service at this time slot, then CC drops this user and picks another [18] Texas Instruments, Inter-Cell Interference Mitigation for EUTRA, 3GPP R1-051059, 3GPP TSG RAN WG1 Meeting #42bis, October 2005.
[19] Shannon, C.E., Collected Papers, Edit by Sloane & Wyner, IEEE Press, 1993.
one uniformly; [20] P. Mogensen, Wei Na, I.Z. Kovacs, F. Frederiksen, A. Pokhariyal, K.I. Pedersen, T. Kolding, K. Hugl, M. Kuusela, LTE Capacity Compared to the Shannon Bound, in Proc. IEEE 65th Veh. Technol.
Conf., April 2007, pp. 1234-1238.
[21] Qualcomm Europe, Description and simulations of interference management technique for OFDMA based E-UTRA downlink evaluation, 3GPP R1-050896, 3GPP TSG RAN WG1 Meeting #42, Sept.
5: Go back to step 3 until all the CoMP needed users find their partners; [22]
2005.
NTT DoCoMo, NEC, SHARP, Intra-Node B Macro Diversity Using Simultaneous Transmission with Soft-combining in Evolved UTRA Downlink, 3GPP TSG RAN WG1 Meeting #42, September
2005.
[23] Siemens, LTE interference Coordination Downlink System Simulation Results, 3GPP R1-062374, 3GPP TSG RAN WG1 Meeting #46, Sept. 2006.
[24] C.-S. Chiu and C.-C. Huang, A Hybrid Inter-Cell Interference Mitigation Scheme for an OFDMA Downlink System, IEICE Trans. on Comm., vol. E93-B No.1 pp.73-