An important issue affecting cellular networks is to make services available to regions of bad or no reception. Femtocells are considered to be the solution to meet the future needs for better indoor coverage, high data rates and capacity in the cellular networks. Since FAPs are usually user deployed, blind placement of FAPs is inevitable leading to problems of power spillage causing severe co‐ and cross‐tier interference and subsequent performance degradation. Even though femtocells are discovering an important role, the issue of interference as a result of blindly placed FAPs needs to be addressed. This chapter presents performance analysis of a coverage radius based power control scheme to circumvent the problems caused by blind placement of FAPs. Our co scheme does not require FAPs to be relocated to optimal positions for interference mitigation; rather a self‐update algorithm is implemented by FAPs to reduce their cell radius by adaptive adjustment of power values for interference management. Using system level simulations, the performance of the scheme has been analysed for densely deployed single and multicell femtocell scenarios. Compared to existing schemes, our scheme provides improved interference mitigation and throughput results. The results reveal that coverage radius of FAPs has considerable effect on both co‐ and cross‐tier SINR values. It is also shown that there are coverage radius bounds that provide balanced co‐ and cross‐tier SINR values. These bounds provide valuable information for effective distribution of FUE to avoid performance degradation in blindly placed densely deployed femtocells.

4.1 Introduction

Indoor cellular usage, which accounts for 50% of all voice calls and 70% of data traffic, is mostly faced with poor reception due to low signal penetration through walls and attenuation, which may lead to total loss of signal [1]. A MBS, which motivates existing wireless strategies, is supported by the deployment of smaller cells to address these growing issues. Most of the benefits of smaller cells overlaid in the MBS arise from the fact that a significantly reduced distance between transmitter and receiver results in a better‐quality link. The femtocell is a new indoor solution to cellular communication, which aims to provide better indoor network coverage and has attracted network operators and stakeholders. Femtocells are low powered, low cost and subscriber controlled units that provide a dedicated Base Station (BS) to subscribers indoors.

The concept of femtocells, also known as home BSs, Home NodeB (HNB) or Home eNodeB (HeNB) [2] and residential small cells [3] was first studied in 1999 by Bell Labs of Alcatel‐Lucent, but it was in 2002 that Motorola announced the first 3G based home BS product [4]. Femtocell units, known as FAPs, connect standard mobile devices to the network of a mobile operator through residential Digital Subscriber Line (DSL), optical fibres, cable broadband connections or wireless last‐mile technologies [4–6]. By installing FAPs indoors, the cell sites are reduced, thereby bringing the transmitter and receiver closer to each other. The use of the subscriber’s broadband network to backhaul data offer improved mobile phone coverage indoors for both voice and data compared to the MBS. The close proximity greatly lowers transmission power and increases the battery life of mobile devices and with a dedicated FAP it offers subscribers a single billing address for mobile phone, broadband and land line as they are all channelled through the same backhaul [7]. Additionally, it acts as a solution to the convergence of landline and mobile systems [8].

Due to the two‐tier architecture of femtocells and macrocells, interference is imminent. The cell sites covered by a number of FAPs (in some cases overlapping each other) are overlaid in the larger cell site of the macro BSs. Interference could be between a femtocell and macrocell, which is known as cross‐tier interference or between neighbouring femtocells known as co‐tier interference [9]. Interference can be further classified as uplink (UL) or downlink (DL) based on the sources, which besides the FAP and MBS, also include FUE – a UE served by a FAP) and Macrocell User Equipment (MUE – a UE served by a MBS). It is important to note that the scale at which interference affects a femtocell network is largely dependent on the deployment scenario. In a dedicated channel deployment, the licensed spectrum is split into different portions for each tiered network to operate in a dedicated manner whereas both tiers share the same licensed spectrum in co‐channel deployment [10]. Network operators prefer a co‐channel deployment due to the limited available bandwidth but will have to deal with the interference issues [11–13]. FAPs deployed in an open access allow connection for all users, whereas in a close subscriber group (CSG) mode only the subscribed owners of the FAP have access. A new access deployment is hybrid access, which combines open access and the CSG by allowing only a limited amount of resources to all users [14]. The power transmitted by a FAP, which comprises of pilot and traffic power values, also has a detrimental effect in a network. For instance, a high pilot power will result into a large cell coverage area, but consequently has higher chances of causing increased interference. Thus, there is a need to optimize the power levels in femtocells to avoid interference while maintaining a certain QoS. Some of the typical power control schemes employed by femtocells for interference mitigation are described in [15–17].

Nonetheless, a FAP is user deployed and is usually blindly placed in an indoor environment such as near walls and windows. In the absence of any antennas’ beam directivity and so on, the FAP power could spill out in the surrounding regions, thereby causing considerable co‐ and cross‐tier interference (Figure 4.1). As opposed to blind placement, research work carried out in [18–20] investigates optimal positions to place a FAP in an indoor environment to improve the throughput and mean capacity. However, it might not always be possible or necessary to find the optimal positions and move the FAPs to effectively mitigate the interference caused. Thus, there is a serious need for development of interference mitigation schemes for blindly placed LTE femtocells. In this chapter, the problem of interference due to blind placement of FAPs is addressed by analysis of a coverage radius based power control scheme, which adaptively varies the pilot power of FAP based on its distance from the farthest served FUE. The scheme investigated in the chapter does not require FAPs to be located or relocated at optimal positions; rather, interference is mitigated by coverage radius based power control scheme. Based on balanced co‐ and cross‐tier SINR values, the important research findings are coverage radius bounds that provide valuable information towards radial placement of FUE in densely deployed blindly placed femtocells.

The rest of the chapter is organized as follows: the coverage radius based power control scheme is described in Section 4.2 before Section 4.3 presents the system model and simulation parameters. Section 4.4 details the performance analysis, results and discussion whereas finally the chapter is summarized in Section 4.5.

4.2 Coverage Radius Based Power Control Scheme (PS)

The coverage radius based PS is described next with the help of equation 4.1 and illustrated in Figure 4.2.

(4.1)

Where P_t is the FAP power value, P_m is the MBS power value, G(θ) is the MBS antenna gain in the direction of the FAP, L_m(d) is the MBS path loss relative to a femtocell distance d, L_f(r) is the FAP path loss relative to the target radius r, and P_max is an upper limit power value of P_t set at 20 dBm. The varying value of r that is configured at each FAP is determined to be the distance between the FAP and its farthest served UE with constraints as described next.

4.3 System Model

This section describes the system model and simulation parameters based on 3GPP LTE specifications [23]. The simulation tool used in our analysis is the well‐known LTE system level simulator v1.7r1119, which is offered for free under an academic, non‐commercial use licence [24]. However, the simulator has some limitations because it only supports a fixed transmit and pilot power value algorithm implementation for femtocells. Eventually, its functionality was enhanced by the implementation of existing adaptive and proposed coverage radius based adaptive power control algorithms. The main simulation parameters are given in Table 4.1. The considered network topology consists of two simulation scenarios; a single tri‐sector and a densely deployed urban scenario that consists of seven tri‐sector hexagonal MBS cells with an inter‐site distance of 500 m. Ten FAPs are blindly distributed in each sector with four pieces of FUE attached to each FAP in a CSG fashion. Additionally, 30 MUE are randomly deployed in each sector of the MBS to investigate effects of cross‐tier interference. WINNERII+ channel modelling [25] is used and traffic load is uniformly distributed among all the users.

To account for the macrocell propagation model, our simulations employ the macroscopic path loss model as proposed in [26] in an urban environment and defined in equation 4.2.

(4.2)

Where R is the distance between BS and UE in km, Dhb is the height of the BS antenna above ground in metres and f is the carrier frequency in MHz. The path loss model implemented at the femtocell is the dual slope path loss for urban deployment while ignoring shadowing and penetration losses as defined in equation 4.3.

(4.3)

where d_2D,indoor in this context is the indoor distance between an FAP and its serving FUE.

The azimuth antenna gain pattern used is proposed in TS36.942 [23] and given as follows in equation 4.4.

(4.4)

where ° is the gain pattern angle and is the side lobe gain.

4.4 Performance Analysis

The performance analysis carried out is to investigate and compare our coverage radius based scheme with other power control schemes in both single‐ and multicell scenarios. However, the main purpose is to investigate the coverage radius bounds and subsequently their impact on SNIR for both these scenarios. Simulations have thus been performed for all the possible values of R_f between 10 and 3 m for our coverage radius based scheme and compared with three existing power control schemes. The first is a baseline scheme where all FAPs are assigned a fixed value of 20 dBm. It is important to note that this scheme is used for simulation and comparison analysis and not for implementation because a fixed maximum power is not an ideal solution for mitigating interference in femtocells. The second scheme assigns FAP power value based on the power it receives from its closest MBS while maintaining a target femtocell radius of 10 m [27]. The third scheme is a distance based power control scheme proposed in [28] that intends to limit the impact a FAP has on the aggregate macrocell downlink throughput. In the distance based scheme the MBS is divided into three regions with power values assigned for FAPs in each region as defined in equation 4.5. d_FAP here denotes the distance between a FAP and the closest MBS.

(4.5)

A graphical comparison of the schemes is presented in Figure 4.3 where sector 1 illustrates the distance based power control algorithm as defined in equation 4.5. Sector 2 illustrates the constant radius scheme and sector 3 illustrates our coverage radius based PS. The value of P_t in sectors 2 and 3 is as defined in equation 4.1.

The result of the fixed power value of 20 dBm for all FAPs is denoted ‘FP’ while the distance based power scheme is denoted ‘DB’. The constant radius power scheme is denoted ‘CR’ and for our coverage radius based adaptive PS, the results are shown for the values of R_f at 7, 6 and 5 m denoted PS‐7 and PS‐6 and PS‐5, respectively. The results provided in the chapter are from simulations performed for single‐ and multicell scenarios, whereas the single cell scenario is composed of single tri‐sector hexagonal MBS cell compared to a multicell scenario that consists of seven tri‐sector hexagonal MBS cells with an inter‐site distance of 500 m. The reason single and multicell scenarios are chosen is to investigate the effect of coverage radius bounds on the variations of SINR in each scenario as described next.

4.4.1 Results and Discussion

4.4.1.1 SINR Cross‐Tier (Single Cell)

Figure 4.4 shows the cumulative distribution function (CDF) plot of SINR value for all MUE. The proposed scheme takes the cross‐tier interference impact it has on MUE fully into consideration. The transmit power value in proposed scheme is directly proportional to the coverage radius. Due to its low transmit power (P_t)value for smaller FAP coverage area, our scheme at PS‐5 with a mean SINR value of 10.35 dB performs far better compared to other schemes. With a slightly increased coverage area the mean SINR values of PS‐6 and PS‐7 are 8.75 and 6.13 Db, respectively. On the other hand, due to maximumP_tin case of FP, MUE experiences heavy cross‐tier interference with a low mean SINR value (−3.60 dB). The mean SINR values for DB and CR are 7.95 and 2.26 dB, respectively. As a whole, our proposed scheme at PS‐5 improves SINR by 13.90 dB while compared to FP, 8.09 dB compared to CR and 2.04 dB compared to DB schemes.

4.4.1.2 SINR Co‐Tier (Single Cell)

Figure 4.5 shows the SINR results for all the FUE. FP performs better compared to our proposed scheme because some of the femtocells in the simulated scenario are standalone with a maximum fixed value of P_t, thus resulting in better co‐tier SINR. This improvement is due to the fact that in FP, FAPs satisfy their serving FUE, however, completely disregarding neighbouring FAPs and MUE (causing serious cross‐tier interference as already shown in Figure 4.4).

The mean SINR values of DB, CR, PS‐5, PS‐6 and PS‐7 are −0.16, 7.93, 3.45, 4.99 and 6.24 dB, respectively. The slightly lower SINR values in our proposed scheme as compared to CR are attributed to lower values of P_t because of smaller coverage radius.

4.4.1.3 Downlink Throughput (Single Cell)

Figure 4.6 shows the CDF plot for downlink throughput over all FUE. In accordance with SINR results, the baseline scenario FP performs better due to its high P_t value with a mean throughput of 12.06 Mbps. With mean throughput values of 6.72, 7.64 and 8.42 Mbps for PS‐5, PS‐6 and PS‐7, respectively, our scheme shows that a significantly high throughput can still be achieved with a varying coverage radius as compared to CR (8.64 Mbps) with fixed coverage radius. DB with a mean throughput value of 3.99 Mbps performs lower than the other schemes.

4.4.1.4 Co‐ and Cross‐Tier SINR (Single Cell versus Multicell)

Figures 4.7 and 4.8 show how co‐ and cross‐tier SINR, respectively, vary for all the mentioned schemes compared against single‐ to multicell scenarios. Compared to a single‐cell scenario, clearly for all the schemes both co‐ and cross‐tier SINR values dropped in a multicell scenario because of enhanced interference experienced by any cell due to increased number of blindly placed FAPs. Figure 4.7 further reveals that no matter what the scenario is (single‐cell or multicell), the DB scheme resulted into extremely poor co‐tier SINR values. Compared to other schemes, even though FP scheme exhibited better co‐tier SINR values for both scenarios (Figure 4.7), however, as expected, FP due to its very nature resulted into very bad cross‐tier SINR values (Figure 4.8). Despite the fact that CR scheme compared to our PS resulted into better co‐tier SINR values (Figure 4.7), however, it is important to note that CR scheme suffered serious degradation for its cross‐tier SINR value (Figure 4.8). Lastly, even though all schemes resulted into droppage in SINR values compared against single and multicell scenarios, our proposed scheme (PS) clearly proven to be the best scheme in terms of SINR values.

4.4.1.5 Droppage in SINR (Single Cell versus Multicell)

Figure 4.9 compares mentioned schemes for a percentage drop in SINR values for single‐ and multicell scenarios. Table 4.2 shows how SINR values have reacted to the change in scenario (single to multicell).

Scheme	Co‐Tier SINR (%age)	Cross‐Tier SINR (%age)
FP	22	33
DB	95	71
CR	60	90
PS‐5	53	54
PS‐6	72	50
PS‐7	68	42

The shaded fields in Table 4.2 provide very important information about the schemes and how change of scenario has affected their SINR values. It might appear that compared to other schemes FP scheme suffered the lowest droppage in its SINR values, however, careful consideration would reveal that irrespective of the scenarios, cross‐tier SINR for FP has never been of any importance because it has always stayed negative (−3.63 dB for single‐cell and −5.425 dB for multicell, Figure 4.8). It further proves that a fixed power scheme such as FP is not at all a suitable scheme for cross‐tier interference mitigation in densely deployed blindly placed femtocells.

On the same lines, DB scheme behaved very poorly for co‐tier SINR values (highlighted in Table 4.2, it suffered a drop of 95% with the change of scenario). Co‐tier SINR values for DB scheme stayed at −0.160 dB for single‐cell and −2.977 dB for multicell scenarios (Figure 4.7). Lastly, even though the CR scheme showed some promise in terms of co‐tier SINR values, it suffered droppage of 90% for its cross‐tier SINR values (highlighted in Table 4.2). Despite the droppage due to change in scenario, our proposed scheme has always been promising both for co and cross‐tier SINR values. The bounds of coverage radius and its impact on SINR values for our PS are further described in Section 4.6.

4.4.1.6 Coverage Area Bounds and Impact on SINR (Single Cell versus Multicell)

Figure 4.10 plots how co‐ and cross‐tier SINR values responded to the change in coverage radius and also the change in the scenario for our proposed scheme.

Irrespective of the scenario (single‐ or multicell), lower coverage radius gave higher cross‐tier SINR values, whereas increase in coverage radius resulted into better co‐tier SINR and vice versa. This further validates the results already shown in Figures 4.4, 4.5, 4.7 and 4.8. Higher cross‐tier SINR values are obtained because of the fact that with reduced coverage radius more MUE is left out, thereby resulting in lower interference and better cross‐tier SINR. On the other hand, since coverage radius is directly proportional to the transmit power, lower coverage radius resulted into lower signal power per FUE compared to increased interference due to densely deployed blindly placed femtocells, thus resulting in reduced co‐tier SINR. It is fair to conclude that in our proposed interference mitigation scheme, for both scenarios (single‐ and multicell) a lower coverage radius favours cross‐tier SINR whereas higher coverage radius values favour co‐tier SINR. However, from our simulations it is shown that there are coverage radius bounds (i.e. ~6 m for single‐cell and ~4.7 m for multicell), which resulted into balanced (optimum) value for both co‐ and cross‐tier SINR values. These findings are very important in blindly placed densely deployed femtocells for the possible distribution and radial mobility of FUE to avoid significant performance degradation.

4.5 Summary

Femtocells aim to improve poor indoor network coverage in cellular communication, which has attracted network operators and stakeholders. Even though femtocells are discovering an important role, the issue of interference as a result of blindly placed FAPs needs to be addressed. In this chapter, a coverage radius based adaptive power control scheme to mitigate interference for blindly placed LTE femtocells was investigated. The proposed scheme does not require FAPs to be relocated on optimal locations for effective interference mitigation, rather it implements a self‐update algorithm for FAPs to reduce their cell radius and adjust power values in an adaptive manner. The performance of the scheme was analysed using system level simulations for single‐ and multicell scenarios. The results have shown that our proposed scheme has an improved value of cross‐tier SINR, throughput and lower co‐tier SINR compared to baseline and existing adaptive interference mitigation schemes. Further, the results have proven that, irrespective of the scheme, the change of scenario from single to multicell, affected and resulted in lower co‐ and cross‐tier SINR values for multicell compared to single‐cell values. It was also found that our proposed adaptive power control scheme contributed towards coverage radius bounds, which provide balanced co‐ and cross‐tier SINR values. In terms of densely deployed blindly placed LTE femtocells, coverage radius bounds are a very important finding because they can be helpful in the effective distribution of FUE to achieve balanced co and cross‐tier SINR values while maintaining other performance parameters too; for example, throughput.

References

1. 1 V. Chandrasekhar, J. Andrews and A. Gatherer, “Femtocell networks: a survey”, Communications Magazine, IEEE, vol. 46, no. 9, September 2008, pp. 59–67.
2. 2 3GPP. A Global Initiative, Website, available at: www.3gpp.org (accessed December 2017).
3. 3 Small Cell Forum. Website, available at: www.smallcellforum.org (accessed December 2017).
4. 4 J. Zhang and G. Roche, Femtocells: Technologies and Deployment, John Wiley & Sons, Ltd, 2010.
5. 5 F. Liu, E. Bala, E. Erkip and R. Yang, “A framework for femtocells to access both licensed and unlicensed bands”, International Symposium on Modeling and Optimization in Mobile, Ad Hoc and Wireless Networks (WiOpt), 9–13 May 2011, pp. 407–411.
6. 6 J.G. Andrews, H. Claussen, M. Dohler S. Rangan, and M.C. Reed, “Femtocells: Past, present, and future”, IEEE Journal on Selected Areas in Communications, vol. 30, no. 3, April 2012, pp. 497–508.
7. 7 H.A. Mahmoud and I. Guvenc, “A comparative study of different deployment modes for femtocell networks”, IEEE 20th International Symposium on Personal, Indoor and Mobile Radio Communications, 13–16 Sept. 2009, pp. 1–5.
8. 8 N. Saquib, E. Hossain, L.B. Le and D. Kim, ‘“Interference management in OFDMA femtocell networks: issues and approaches”, Wireless Communications, IEEE, vol. 19, no. 3, 2012, pp. 86–95.
9. 9 T. Zahir, K. Arshad, A. Nakata and K. Moessner, “Interference management in femtocells”, Communications Surveys & Tutorials, IEEE, vol. 15, no. 1, Feb. 2013, pp. 293–311.
10. 10 Y. Bai, J. Zhou and L. Chen; “Hybrid spectrum usage for overlaying LTE macrocell and femtocell”, IEEE Global Telecommunications Conference, GLOBECOM, Nov. 2009, pp. 1–6.
11. 11 D.L. Perez, A. Valcarce, G.D.L. Roche, E. Liu, and J. Zhang, “Access methods to WiMAX femtocells: A downlink system‐level case study”, in Proc. IEEE Int. Conf. Commun. Syst. (ICCS), Guangzhou, China, Nov. 2008, pp. 1657–1662.
12. 12 L.T.W. Ho and H. Claussen, “Effects of user‐deployed, co‐channel femtocells on the call drop probability in a residential scenario”, in Proc. IEEE Int. Symp. Personal, Indoor, Mobile Radio Commun. (PIMRC), Athens, Greece, Sep. 2007, pp. 1–5.
13. 13 V. Chandrasekhar and J.G. Andrews, “Uplink capacity and interference avoidance for two‐tier cellular networks”, in Proc. IEEE Global Telecommunication Conference, Nov. 2007, pp. 3322–3326.
14. 14 G. de la Roche, A. Valcarce, D. Lopez‐Perez, and J. Zhang, “Access control mechanisms for femtocells”, IEEE Commun. Mag., vol. 48, no. 1, Jan. 2010, pp. 33–39.
15. 15 Z. Wang, W. Xiong, C. Dong, J. Wang and S. Li, “A novel downlink power control scheme in LTE heterogeneous network”, International Conference on Computational Problem‐Solving (ICCP), 21–23 Oct, 2011, pp. 241–245.
16. 16 P. Mach and Z. Becvar, “Dynamic power control mechanism for femtocells based on the frame utilization”, International Conference on Wireless and Mobile Communications (ICWMC), Sept. 2010, pp. 498–503.
17. 17 D. Roche, G. Ladányi, D. López‐Pérez, D.C. Chong and J. Zhang, “Self‐organization for LTE enterprise femtocells”, IEEE GLOBECOM Workshops (GC Wkshps), Dec. 2010, pp. 674–678.
18. 18 J. Liu, Q. Chen and H.D. Sherali, “Algorithm design for femtocell base station placement in commercial building environments”, INFOCOM, 2012 Proceedings IEEE, 25–30 March 2012, pp. 2951–2955.
19. 19 S. Wang, W. Guo and T. O’Farrell, “Optimising femtocell placement in an interference limited network: theory and simulation”, Vehicular Technology Conference (VTC Fall), 2012 IEEE, 3–6 Sept. 2012, pp. 1–6.
20. 20 W. Guo and S. Wang, “Interference‐aware self‐deploying femto‐cell”, Wireless Communications Letters, IEEE, vol. 1, no. 6, Dec. 2012, pp. 609–612.
21. 21 H.O. Kpojime, G.A. Safdar, “Efficacy of coverage radius‐based power control scheme for interference mitigation in femtocells”, Electronics Letters, vol. 50, no. 8, April 10 2014, pp. 639–641.
22. 22 G.A. Safdar, “Analysis of time correlated channel model for simulation of packet data networks”, Antennas and Propagation Conference (LAPC), 2011 Loughborough, Nov. 2011, pp. 1, 4, 14–15.
23. 23 Technical Specification Group RAN, “E‐UTRA; LTE RF system scenarios”, 3GPP, Tech. Rep. TS 36.942, 2008–2009.
24. 24 J.C. Ikuno, M. Wrulich, and M. Rupp, System level simulation of LTE networks, in Proc. 2010 IEEE 71st Vehicular Technology Conference, Taipei, Taiwan, May 2010. [Online] Available: http://publik.tuwien.ac.at/files/PubDat_184908.pdf (accessed December 2017).
25. 25 L. Hentila, P. Kyosti, M. Kaske, M. Narandzic, and M. Alatossava, “MATLAB implementation of the WINNER Phase II Channel Model ver1.1”, Dec. 2007.
26. 26 ETSI TR 136 942 V10.3.0 “LTE; Evolved Universal Terrestrial Radio Access (E‐UTRA); Radio Frequency (RF) system scenarios” July 2012.
27. 27 H. Claussen, “Performance of macro‐ and co‐channel femtocells in a hierarchical cell structure”, in Proc. IEEE Int. Symp. Personal, Indoor, Mobile Radio Commun. (PIMRC), Athens, Greece, Sep. 2007, pp. 1–5.
28. 28 Small Cell Forum, “Interference Management in OFDMA Femtocells”, Small Cell Forum, Mar. 2010.