Adaptive Soft Frequency Reuse Scheme for In-building Dense Femtocell Networks

Transcription

1 SELECTED PAPERS FROM IEEE ICCC'12 Adaptive Soft Frequency Reuse Scheme for In-building Dense Femtocell Networks CHEN Jiming 1, WANG Peng 2, ZHANG Jie 3 1 Ranplan Wireless Network Design Ltd., UK 2 Centre for Wireless Network Design, University of Bedfordshire, Luton, UK 3 The Communications Group, University of Sheffield, Sheffield, UK Received: Revised: Editor: NIU Zhisheng Abstract: Femtocell networks have emerged as a key technology in residential, office building or hotspot deployments that can significantly fulfill high data demands in order to offload indoor traffic from outdoor macro cells. However, as one of the major challenges, inter-femtocell interference gets worse in 3D in-building scenarios because of the presence of numerous interfering sources and then needs to be considered in the early network planning phase. The indoor network planning and optimization tool suite, Ranplan Smallcell, makes accurate prediction of indoor wireless RF signal propagation possible to guide actual indoor femtocell deployments. In this paper, a new adaptive soft frequency reuse scheme in the dense femtocell networks is proposed, where multiple dense femtocells are classified into a number of groups according to the dominant interference strength to others, then the minimum subchannels with different frequency reuse factors for these groups are determined and transmit powers of the grouping sub-channels are adaptively adjusted based on the strength to mitigate the mutual interference. Simulation results show the proposed scheme yields great performance gains in terms of the spectrum efficiency relative to the legacy soft frequency reuse and universal frequency reuse. Key words: in-building femtocell networks; soft frequency reuse; adaptive interference coordination I. INTRODUCTION Recently, it has been reported that about 80% of wireless communication traffic comes from indoors [1]. However, penetration loss and complex indoor environment make that it is difficult for existing cellular network to sufficiently follow the demand of indoor mobile traffic. The deployment of femtocells is envisioned to be a key solution for providing high wireless data-rates, offloading the macrocell traffic and enhancing the coverage of existing networking [2]. Femtocell is a low-power consuming base station connected to the service provider s network via broadband such as DSL or cable. It can improve the indoor coverage and data performance by means of frequency reuse. Femtocells can be densely deployed in a small area such as an office building, hotspot area, or residential area. Nevertheless, an unplanned or random user-installed deployment faces several problems and challenges in terms of interference management and backhaul constraints [3]. In co-channel deployments, all femtocells reuse the spectrum resources, and interference from other femtocells might significantly deteriorate the overall system performance by multiple dominant interferences, 44 China Communications January 2013

2 which are from femtocells not only on the same floor due to adjacent deployment, but also on the upper and lower floors due to cross-floor signal penetration. Hence, interference modeling used in traditional cellular scenario is not proper in a building environment. Currently, several interference mitigation schemes in macro base station scenario have been widely studied mainly by means of interference randomization, interference control, interference suppress, and interference coordination. In the first class the interference is averaged across the whole spectrum via spreading sequences (e.g. scrambling and interleaving), and therefore is not actually cancelled out. The second class uses the power control and static beamforming to reduce the interference level. But by contrast, in the third class, the interference is successfully suppressed by using advanced signal processing techniques [4]. Although these techniques are becoming popular, however, the complexity at the receiver side and backhaul constraint are still the challenging issues particularly in the presences of multiple dominant interferences. Inter-Cell Interference Coordination (ICIC) techniques [5], on the other hand, present pragmatically a more feasible solution. Unlike the macro base station, femto base stations can be installed by users in a random manner, making it difficult to handle the interference problem. The traditional methods can be applied to the mitigation of inter-femtocell interference when femtocells are deployed in a systematic way with low density [6]. However, when multiple femtocells are densely deployed in a building environment, the interference source will be greatly increased, and the interference scenario will drastically vary due to a large number of dominant random interfering nodes. In order to mitigate the inter-femtocell interference in the dense environment, several methods have been proposed, e.g. Fractional Frequency Reuse (FFR) [7-8], which uses the flexible Frequency Reuse Factors (FRF) in the cell- edge area and cell-centre area. But in Ref. [7], dense inter-femtocell interference is not specially considered. While in Ref. [8], a femtocell can be granted admission into a group only when it interferes with all the femtocells already admitted by that group. However, it should be noted that it has high computational complexity and the graph based method is very flexible. Therefore, different admission control criterions can be developed to tune the size of each group, and spectrum resource can be fully used for each group to improve the spectrum efficiency. In this paper, a new soft frequency reuse scheme is proposed for interference management in dense femtocell networks, which was partly presented in Ref. [9]. Based on the Reference Signal Received Power (RSRP) from the serving users, the multiple interfering femtocells can be determined to form several groups, and then the minimum subchannels with different soft frequency reuse factors for these groups are allocated to provide optimal performance near the cell edge. After the allocation of sub-channels to femtocells in each group, the transmit power on the sub-channel part for the group is adaptively adjusted in term of the interference strength. By making the use of higher frequency reuse factor with the least number of orthogonal sub-channels according to the deployment and interference environment, the dense femtocells can perform interference mitigation, and consequently improve the overall system performance. Simulation results show the SINR performance gain is more than 3 db and the average spectrum efficiency outperforms 10.12% comparing with the legacy soft FFR scheme. The remainder of this paper is organized as follows. Sections II and III describe the system model and existing frequency reuse schemes. Section IV proposes the adaptive soft frequency reuse scheme in a multi-femtocell deployment environment. In Section V, the simulation results of the proposed scheme based on the Ranplan Small-cell tool suite [10] is presented, and finally, Section VI In the proposed adaptive soft frequency reuse scheme, multiple dominant interference from in-building dense femtocells is classified into a number of groups, and different frequency reuse factors and transmit powers for these groups are adjusted adaptively to mitigate the mutual interference. China Communications January

3 summarizes the conclusions and further works. II. SYSTEM MODEL Considering the downlink transmission of a femtocell network in a multi-floor office building, as shown in Figure 1, M femtocells and K User Equipments (UEs) are deployed, respectively. These femtocells use an Orthogonal Frequency Division Multiple Access (OFDMA) technique. Here, we assume that the building is located in the macrocell coverage area, and femtocells use separated frequency band so that interference from macrocell can be neglected. For any transmission between a femtocell m and one of its UE k, the received signal at UE k over the subcarrier of OFDM systems can be denotes as where M k = k, m m + k, i i + k i= 1, i¹ m y H x å H x n (1) y k denotes the received signal vector by the UE k; H ki, are complex channel state information from femtocell i to UE k; x m is the transmitted data stream of femtocell m; and n k represents the noise at UE k which is considered as a power spectrum density N 0 additive white Gaussian noise vector (AWGN). Then the average Signal to Interference plus Noise Ratio (SINR) of UE k from femtocell m at sub-channel n can be represented as n k = M å i= 1, i¹ m n -L 10 2 k m10 Hm, n n -L 10 2 k i 10 i, n + f 0 p P H b N (2) n here, P m denotes the transmit power of femtocell m at sub-channel n; b f is the sub-channel a spacing; and Lk = 10 log 10 ( dm) + Lw, m denotes the path loss in db, where d m denotes the distance from femtocell m to user, L wm, denotes the wall loss in db, and α denotes the path loss exponent. Fig.1 3D office building model 46 China Communications January 2013

4 III. PREVIOUS FREQUENCY REUSE SCHEMES In this section, we will review several frequency reuse schemes [5] that are related to our work. 3.1 Frequency reuse 1 (FR 1) Frequency reuse 1 is well known as universal frequency reuse scheme, where the entire frequency spectrum is reused over all cells, as shown in Figure 2. However, by employing this scheme only the cell centre users experience good channel quality whereas cell edge users suffer from low radio conditions due to severe inter-cell interference. edge users are scheduled in the protected resources and cell centre users are scheduled in the shared resources with a lower transmit power, as shown in Figure 4. Fig.4 The soft frequency reuse scheme IV. PROPOSED ADAPTIVE SOFT FREQUENCY REUSE SCHEME 4.1 Optimal threshold Fig.2 The frequency reuse 1 scheme 3.2 Hard frequency reuse 3 Figure 3 illustrates how the power-frequency resource restrictions are applied at each cell in frequency reuse 3 scheme. It is important to note that the frequency reuse factor is restricted by a list of integer numbers: {1, 3, 4, 7,, i 2 + i j + j 2 i, j N }. Fig.3 The hard frequency reuse 3 scheme 3.3 Soft frequency reuse Soft frequency reuse scheme suggests that cell In the coverage of femtocell m, considering the inter-femtocell interference, especially in the cell-edge region, we can partition the cell coverage into cell centre and cell edge regions. Users in the centre and at the edge of femtocell are identified based on the received SINR when the frequency resource is reused. A user is regarded as in the centre of a femtocell if its SINR is above a threshold; otherwise, it is regarded as at the edge. Denote M C and M E to be the index sets of the centre users and the edge users, respectively. Then Mc = { k: k >G th} and ME = { k: k G th} (3) where G th represents the SINR threshold. In Ref. [11], the distance-based and SINR-based thresholds are optimized to maximize the mean to variance ratio of SINR, i.e. maximizing the mean to variance ratio of the received SINR for a given user. Thus, this parameter is used to achieve an optimal dimension of the centre region and an efficient use of the available bandwidth. The objective function can be written as G ( ) ( ) = V ( ) (4) China Communications January

5 where denotes the mean and V is the variance of SINR, respectively. The objective function is maximized by G th which determined as G = max G ( ) (5) 4.2 Femtocell grouping th In the dense femtocell networks, a user at the edge will be interfered by multiple dominant femtocells, especially in the multi-floor office building, as shown in Figure 1, a user on the floor 3 will be interfered by femtocells not only on the same floor but also on the upper and lower floor, such as floor 1, 2, 4, and 5, and so on due to the cross-floor penetration, which is obviously different from the traditional macro cellular scenario. Based on the cell-edge regions of femtocells m, the cell-edge UE k will receive the signal from M femtocells, which can be classified into a number of groups to determine the frequency reuse factors and to set the priority order for the sub-channel allocation of each group. But for each cell-edge user, there have the different interfering femtocells, so the same interfering femtocells set is selected to do the grouping based on the femtocell neighboring relationship, which is similar to the legacy soft FFR scheme in Ref. [5]. Figure 5 gives an example of femtocell grouping. Here, the femtocells grouping is based on the interference graph modeling method. Based on the RSRP of femtocells set, define Q s as the received signal strength from serving femtocell, and Q i, i=1,2,...,m-1 as the received signal strength from interference femtocells. Let G={g 1,, g L } be a set of Fig.5 An example of femtocell grouping groups, where g l, l=1,,l, comprises a sets of femtocells interfering with each other. So define gl Tl-1 < Qs-Qi Tl, for all i= 1, 2,, M -1 (6) where T l is the grouping threshold in db of lth group. 4.3 Grouping-based frequency reuse factor selection From the grouping results, this subsection introduces the selection of grouping-based frequency reuse factor. Figure 6 illustrates an example of the proposed scheme, where femtocells are classified into three groups. Instead of having only a group on top of the cell-edge region, it is possible to form up to multiple virtual groups each using different parts of spectrum and different frequency reuse factors. Following the concept, the group with the strongest interference strength has the lowest frequency reuse factor, the group with the weakest interference strength has the highest frequency reuse factor, and the others are in the middle. Therefore, for Figure 6, the frequency reuse factor for group 1 is 1/3, the group 2 is 2/3, and the group 3 is 3/4. Accordingly, if there are more groups and more femtocells per group, different frequency reuse factors and frequency partition are used based on cyclic difference sets, where the cyclic property allows a quick FRF adjustment. 4.4 Group reuse pattern algorithm After grouping in Subsections 4.2 and 4.3, next step is to allocate the sub-channel for femtocells in these groups based on the different frequency reuse factor, i.e. selecting the frequency reuse pattern for the groups. If the interference cells are sorted out according to the serving cells, the neighborhood information of each serving cell can be obtained as a result. If we consider each cell as a node in a graph, G(V, E), where V represents the set of grouped femtocells and E is the set of edges. It is expected that two nodes that have an edge between them select different sub-channel 48 China Communications January 2013

6 Fig.6 An example of proposed soft frequency reuse scheme patterns and frequency reuse factor according to grouping criterion in order to improve the cell edge performance. The problem of finding out the smallest number of sub-channels in the in-building scenario can be formulated as a graph coloring problem. Each node chooses a color different from that of neighbor nodes if there has an edge between them. The smallest number of colors will be the solution of the problem. In the network planning phase, Ranplan Smallcell tool suite is able to offline solve the problem and perform the sub-channel pattern assignment for each cell through the femtocell management system. The node coloring problem is an NP-complete problem and a heuristic algorithm is presented to address the problem to seek a near optimal solution. The value of currentcolor is considered as the smallest number of colors used in the scenario. The whole frequency band or resource blocks in LTE should be divided into currentcolor sub-channels and each sub-channel is assigned to each cell for cell edge users with higher transmit power according to the color of the node. Note that if there are not enough cell edge users to be scheduled at the reserved sub-channel, the sub-channels can be also used for cell-centre users to achieve high resource efficiency. 错误! Pseudo-code of the Developed Heuristic Algorithm: INPUT G=(V, E) 1 Compute Degree(v) for all v in V, and V is classified into L groups 2 Set uncoloredcells = V sorted in ascending order of Degree(v) 3 Set currentcolor = 0 4 For each element in uncoloredcells: 5 currentcolor = currentcolor Set u = first element of uncoloredcells 7 If u is in the group l: 8 Set cellcolor(u) = currentcolor in group l 9 Set coloredcells = {u} 10 Set coloredgroup = {l} China Communications January

7 11 Set frequencyreusefactorcells 12 Remove u from uncoloredcells and group l 13 For each v in uncoloredcells: 14 If v is not adjacent to any cells in coloredcells of group l: 15 Set cellcolor (v) = currentcolor 16 Add v to coloredcells 17 Remove v from 18 End if 19 End for 20 End if 21 End for uncoloredcells In this way, the proposed scheme allows not only to control the degree of coverage but also to adaptively adjust the transmit power between the cell-edge, cell-centre region, and groups, which effectively improves the frequency reuse efficiency of UEs on the celledge region. 4.5 Adaptive power control of sub-channel on the edge region In order to improve the performance of cell-edge UE, we adjust the transmit power on the sub-channels of the group based on the RSRP, which should be adaptively select the transmit power of sub-channel for cell-centre and celledge [12]. So the transmit power of femtocell m in the group l at sub-channel n can be expressed as where Ng l n åårsrpi n iî gl n= 1 total l : m = M N m n åårsrpi i= 1 n= 1 g P P (7) N g l denotes the number of sub-channel in the group g l ; N is the number of total sub- channels; and total P m denotes the total transmit power of femtocell m, i.e., total m P N n= 1 n m = å P That means the transmit power of each group is adjusted based on the RSRP in a distribution manner after the allocation of sub-channels to. femtocells in each group. V. SIMULATION RESULTS The modeling platform is based on the Ranplan Small-cell tool suite [10], which is an in-building network planning & optimization software tool produced by Ranplan Wireless Network Design Ltd., and path loss model is Ranplan Radio-wave Propagation Simulator (RRPS). RRPS implements Intelligent Ray Launching Algorithm (IRLA) [13], which is based on ray tracing/ray launching techniques to model the physical characteristics of the rough terrain and urban building features, performs the electromagnetic calculations, and then evaluates the signal propagation characteristics. The propagation model offers to predict coverage and other channel characteristics such as Angles of Arrival (AoA), Power Delay Profile (PDP). LTE system simulation is based on 9-floor Office building with 4 femtocells deployment on each floor, where 12 db penetration loss for a heavy wall and 20 db penetration loss for a floor are considered. The deployment of femtocell network is shown in Figure 1, and the simulation parameters are given in Table I [14]. In the proposed scheme, how to select the grouping threshold is a key step, which will affect the system performance. Coverage radius-based grouping method [8] can obtain good performance in the outdoor scenario, but in the building, due to the irregular coverage area, the performance will be degraded. Minimizing capacity-based method can find the optimal threshold, but it has high complexity. In the paper, simulation-based method is used to find the suboptimal threshold, where the femtocells are classified as 4 groups, and the grouping thresholds are 1dB, 2.5 db, and 6 db, respectively, and the number of femtocells in these groups are 3, 3, 4, and 2, respectively, i.e. for a UE on the cell-edge region, total 12 femtocells should be considered to group due to the strength interference from each other, and the femtocells, which are Qs- Qi 1dB, are added in the first group, which has the 50 China Communications January 2013

8 strongest interference, the femtocells, which are 1< Qs- Qi 2.5, are added in the second group, the third group is based on 2.5 < Qs- Qi 6, and the fourth group is based on 6 < Qs - Qi, which has the weakest interference. Table I Simulation parameters Parameter Cell layout Area Path loss model Shadowing Bandwidth # of resource blocks (RB) Femtocell Tx power Femtocell antenna UE antenna gain UE noise figure Downlink scheduler UE number User traffic model Value 36 Femtocells m 2 with 9 floors ibuildnet RRPS propagation modeling Lognormal shadowing with 3 db std for LOS, 4 db for NLOS 2 GHz, 10 MHz 50 8 dbm 0 dbi, Omnidirectional 0 dbi 5 db Round Robin 25 UEs per cell/5 RBs per UE Full Buffer Figure 7 shows the best signal level of one floor, where 2 femtocells is deployed. From this figure, we can see that the signal levels to one location are up to 60dBm and 64dBm, respectively, which means that there have severe interference at the femtocell edge point. Figure 8 shows the cross-floor transmission of femtocell signal, where a femtocell on the 3rd floor is shown. From this figure, comparing to the 41 dbm signal level on this floor, it can be seen that the signal strength on the 2nd and 4th floors are up to 77 dbm and 66 dbm, respectively, and the signal strength are also up 100 dbm on the 1st and 5th floors, which shows cross-floor signal penetration will greatly affect the system performance. Figures 7-8 mean that a UE on the 4th floors will receive the interference from other femtocells not only on the 4th floor, but also on the 3rd and 5th floors. Therefore, the signal penetrating from other floors will make inter-cell interference more complicated in the building, which is a 3-dimensional interference environment with dense interference sources, up to 11 interference femtocells. Therefore, using legacy soft frequency reuse will decrease the system spectrum efficiency because too much frequency partition will result in the low frequency reuse gain. Fig.7 Best signal level Fig.8 Cross-floor signal level Figure 9 depicts the objective function as function of the instantaneous SINR γ. It can be seen that the optimal size of the centre region is achieved with G th = 6.5dB, which means the region where SINR is less than 6.5 db is considered as a cell-edge region, and UE in this region will be scheduled based on the proposed adaptive soft frequency reuse scheme. Figure 10 presents the SINR and data rate map with legacy soft frequency reuse scheme. From the figures, it can be observed that the in-building environment is more complex, and China Communications January

9 Fig.9 Optimal SINR threshold encounters more penetration loss and shadowing fading. But due to the transmission loss of wall and pillar, the signals between the femtocells can be well isolated, so the interference can be reduced. Therefore, in the femtocell deployment, walls and pillars can be utilized to form directional transmission. But at the cell-edge region there have low SINR and data rate. Figure 11 presents the SINR and data rate map of proposed scheme when G th = 6.5dB. Comparing to Figure 10, it can be observed the proposed adaptive soft frequency reuse scheme can greatly improve the cell centre and cell-edge performance up to 3.7 db and 3.2 db, respectively, as shown in Figure 12. Figure 13 shows the CDF of the spectrum efficiency when G th = 6.5dB. It can be seen that the universal frequency reuse scheme can have the good system spectrum efficiency but Fig.10 SINR and data rate map based on legacy soft FFR Fig.11 SINR and data rate map of proposed scheme according to G = 6.5dB th 52 China Communications January 2013

10 Fig.12 CDF of SINR for legacy soft FFR and proposed scheme Fig.13 CDF of spectrum efficiency for different FFR schemes it cannot provide desired cell-edge performance. On the other hand, using the legacy soft frequency reuse [5] provides relatively low spectrum efficiency due to the multiple dominant interferences in the dense femtocell networks, which means that too much frequency partition will degrade the frequency reuse gain, and then degrades the system performance. It can be also observed that the system performance of proposed scheme is close to that of universal frequency reuse in the centre of femtocell while cell-edge performance is better than that of legacy soft frequency reuse. From the table II, comparing with the legacy soft FFR scheme, the average performance gain is up to 10.12%, and cell-edge performance gain is up to 25.3%. It is reasonable because the proposed scheme can adjust the FRF and transmit power for different interference groups, and in the severe interference region, the FRF is the smallest, which is close to the FRF of legacy soft frequency reuse. Table II Spectrum efficiency comparison Average spectrum efficiency (bps/hz/cell) Cell edge spectrum efficiency (bps/hz) Universal FR Legacy soft FFR Proposed scheme VI. CONCLUSIONS AND FUTURE WORKS Small cell networks, particularly dense femtocell networks, are promising to provide extra capacity for high indoor data demands in enterprise environments. However, dense femtocell networks will introduce excessive inter-cell interference in the 3D building scenarios. More irregular neighbor cell boundaries will be created due to cross-floor signal penetration and then efficient interference mitigation schemes are required. In this paper, an adaptive soft frequency reuse scheme was proposed for interference management in dense femtocell networks. Based on RSRP, it classifies all interfering femtocells into a number of groups, and then selects the frequency reuse factor and the transmit power for different groups according to the interference strength. Furthermore, a power adaptive scheme in which can effectively control the level of coverage between the cell-edge and cell-centre region was devised, and optimal SINR threshold was given by simulation. The final simulation results show that the proposed scheme achieves not only an enhanced cell-edge spectral efficiency, but also minimal degradation of a forthcoming co-channel femtocell deployment. However, Due to the different load and re- China Communications January

11 source scheduling of each femtocell, the interference environment is time-varying; thus, inter-femtocell interference analysis is also a new challenge for interference coordination or cancellation. These problems need to be further investigated in the future works. At the same time, how to group the femtocells optimally will be the work of next step. ACKNOWLEDGEMENT This work was supported by the EU-FP7 iplan under Grant No and EU-FP7 IAPP@RANPLAN under Grant No References [1] CERWALL P, BERGQVIST S. Ericsson Traffic and Market Data Report[R]. Novenmber, [2] 3GPP. Requirement for Further Advancements for E-UTRA (LTE-Advanced)[S]. 3GPP Technical Report (TR ) v10.0.0, March, [3] LOPEZ-PEREZ D, VALCARCE A, ZHANG Jie, et al. OFDMA Femtocells: A Roadmap on Interference Avoidance[J]. IEEE Communications Magazine, 2009, 47(9): [4] 3GPP, Coordinated Multi-point Operation for LTE: Physical Layer Aspects[S], 3GPP Technical Report (TR ) v1.1.0, August, [5] CHANG R Y, TAO Zhifeng, ZHANG Jinyun, et al. A Graph Approach to Dynamic Fractional Frequency Reuse (FFR) in Multi-Cell OFDMA Networks[C]// Proceedings of the IEEE International Conference on Communications: June 14-18, 2009, Dresden Germany, 2009: 1 6. [6] FEMTO FORUM WHITEPAPER. Interference Management in UMTS Femtocells[R]. www. femtoforum.org, December, [7] KOSTA C, IMRAN A, QUDDUS A U, et al. Flexible Soft Frequency Reuse Schemes for Heterogeneous Networks[C]// Proceedings of the IEEE 73rd Vehicular Technology Conference (VTC Spring): May 15-18, 2011, Budapest Marriott, Budapest, Hungary, 2011: 1-5. [8] LEE H C, OH D C, LEE Y H. Mitigation of Inter-Femtocell Interference with Adaptive Fractional Frequency Reuse[C]// Proceedings of the IEEE International Conference on Communications: May 23-27, 2010, Cape Town, South Africa, 2010: 1-5. [9] CHEN Jiming, WANG Peng, and ZHENG Jie. Adaptive Soft Frequency Reuse Scheme for Inbuilding Dense Femtocell Networks[C]// Proceedings of the IEEE International Conference on Communications: August 15-17, 2012, Beijing, China, 2012: [10] RANPLAN. SMALL-CELL [EB/OL]. [2001] http: // [11] NAJJAR A, HAMDI N, BOUALLEGUE A. Efficient Frequency Reuse Scheme for Multi-cell OFDMA Systems[C]// Proceedings of the IEEE Symposium on Computers and Communications: July 5-8, 2009, Sousse, Tunisia, 2009: [12] PROAKIS J G. Digital Communications (5th Edition)[M]. Mc Graw-Hill, [13] LAI Zhihua, BESSIS N, ZHANG Jie, et al. An Intelligent Ray Launching for Urban Propagation Prediction[C]// Proceedings of the 3rd European Conference On Antennas and Propagation: March 23-27, 2009, Berlin, Germany, 2009: [14] 3GPP, Further Advancements for E-UTRA Physical Layer Aspects[S], TR V.9.0.0, March, Biographies CHEN Jiming, received his M.S. and Ph.D. degrees in communication and information system from University of Electronic and Science Technology, China in 2003 and 2006, respectively. Now he is a senior research fellow at Ranplan Wireless Network Design Ltd. (UK) from Prior to the time, he was a research scientist in Bell labs at Alcatel-Lucent Shanghai Bell Ltd. from 2006 to 2011, and then became a member of Alcatel-Lucent Technical Academy in In Bell labs his research interests were the key techniques of cooperative processing and small cells for 3GPP LTE-A system. Currently, he focuses on the small cell technologies and multiple wireless system-level simulation, such as UMTS/LTE/WiFi, for indoor and outdoor scenarios, and implement in Ranplan planning and optimization tools. Up to now, he held about 21 international patents and 23 publications. Jiming.chen@ranplan.co.uk WANG Peng, received his B.S. and M.Phil. degrees from Centre for Wireless Network Design (CWiND) of University of Bedfordshire (UK). He joined Ranplan Wireless Network Design Ltd. (UK) as an engineer in 2008, and became senior engineer in He participated in executing and managing EU 7th Framework projects at Ranplan. He was also involved in developing of a world leading automatic indoor radio network planning and optimization tool suiteibuildnet. peng.wang@beds.ac.uk 54 China Communications January 2013

12 ZHANG Jie, is a full professor and holds the Chair in Wireless Systems at the Department of Electronic and Electrical Engineering, the University of Sheffield, UK. His research interests are focused on radio propagation, indoor-outdoor radio network planning and optimization, small/femto cell, HetNet, Self-Organising Network (SON) and smart building/city. Since 2006, he has been an Investigator of over 20 research projects worth over 17 million funded by the Engineering and Physical Science Research Council (EPSRC), the European Commission (EC) FP6/FP7 and the industries etc. He is a lead author of the book "Femtocells: Technologies and Deployment" (Wiley, Jan. 2010). He and his colleagues published a widely cited femtocell paper "OFDMA femtocells: A roadmap on interference avoidance". He was involved in the setting up of RANPLAN Wireless Network Design Ltd. ( that produces the world leading Small Cell and HetNet planning and optimization tools. jie.zhang@sheffield.ac.uk China Communications January