MULTI-HOP RADIO ACCESS CELLULAR CONCEPT FOR FOURTH-GENERATION MOBILE COMMUNICATION SYSTEMS

Transcription

1 MULTI-HOP RADIO ACCESS CELLULAR CONCEPT FOR FOURTH-GENERATION MOBILE COMMUNICATION SYSTEMS MR. AADITYA KHARE TIT BHOPAL (M.P.) PHONE , Abstract This paper proposes a multi-hop radio access cellular (MRAC) scheme for achieving both high-speed/highcapacity and good area coverage in fourth generation mobile communication systems. In this scheme, we assume two kinds of hop stations, one is a dedicated repeater station installed at a good propagation location such as a rooftop, and the other is user terminal that temporarily experiences good propagation conditions. For both cases, the path diversity effect can be obtained between single-hop and multi-hop paths. Multi-hop packet radio transmission is an effective technique for extending the range of communications at maximum power. In order to reduce interference from high-power communication users at the cell edge and to enhance the coverage, we propose multi-hop radio access cellular (MRAC) concept. By choosing the optimum path to the BS, the transmit power for the MS can be reduced for the same transmission speed, or the MS can communicate at higher speeds using the same transmit power. Utilizing these modes, results in decreasing the interference from the MS in the power domain or in the time domain, and enhances the cell size or area coverage by relaying the transmissions. Since future generation mobile communication systems will focus on IP-based service, a broadband packet transmission scheme will be employed. Considering this, the path selection can be performed packet by packet, and path diversity is effective for shadowing. Therefore we can expect further improvement by path diversity. INTRODUCTION The IMT rd generation mobile communication system (3G) was launched in Japan in 2001 and will soon be introduced worldwide. The system can provide a variety of mobile multimedia communication services through its high-speed transmission and wide-area coverage capability. Growing demand for mobile multimedia services including mobile internet and mobile e-commerce will result in a remarkable increase in 3G traffic. The fourth generation mobile communication systems (4G) will be introduced from around 2010 and are expected to have significantly higher performance than that of IMT The transmission speed of 50 to 100 Mbit/s, which is two orders of magnitude higher than that of currently operated 3G services, is under discussion. The bit cost should be dramatically decreased so that people can enjoy the convenience of broadband communication without paying high communication charges. However it is quite difficult to achieve both very high-speed/high-capacity and good area coverage at a reasonable cost. Since the target data rate of future systems is more than two orders of

2 magnitude higher than the present systems, the equivalent noise bandwidth in a receiver also increases by two orders of magnitude. This results in a smaller cell. On a simple assumption that the urban mobile radio propagation index is around 3.5 to 4, the noise bandwidth increase causes a cell radius decrease of ¼. This means a large number of base station (BS) sites with the current cellular configuration will be required, which will push costs high. Therefore, we should research more efficient ways of achieving good area coverage and high capacity. This paper proposes a multi-hop radio access cellular (MRAC) scheme that assumes two kinds of hop stations (HS), one is a dedicated repeater station installed at an advantageous propagation position such a roof-top, and the other is a user terminal that temporarily experiences a good propagation conditions. For both cases, the path diversity effect can be obtained between the single-hop and multi-hop paths. Thus, transmit power for a mobile station (MS) can be reduced for the same transmission, or the ms can communicate at higher speeds using the same transmit power. Consequently, the cell radius can be increased. Moreover, the interference currently caused by high-power users located at the edge area of a cell will decrease and system capacity will be enhanced. Current mobile communication systems including IMT-2000 employ a cellular configuration. A wide service area is divided into many small cells and the frequency spectrum is reused among areas that are separated from each other. Thus, this configuration satisfies two important requirements for mobile communications: seamless area coverage and spectrum economy. On the other hand, it has the disadvantage of inter-cell interference, which degrades the communication quality and decreases system capacity or throughout to half or less. Therefore, a great number of BSs (or sectors) are required to maintain the target capacity, which raises the cost of the systems. The effect of inter-cell interference is remarkable at the edge area of a cell, both in generating interference and in being affected by it. This is due to the near-far problem, which is quite common to radio communications. Figure 1 shows the average transmit power required for an ms in a cell. Radio path loss L (db) for an MS varies over time and is expressed as L=L p + L s + L f (1) Where L p denotes large scale propagation loss, L s denotes the loss due to shadowing, and L f denotes the instantaneous fading loss. The farther the location of an MS is from the corresponding BS, the greater is the propagation loss, L p. Therefore, MSs at the cell edge area require the highest transmitting power both in up and downlinks, which results in generating a certain amount of interference to other users. The two other loss components, Ls and Lf, are generally assumed to be independent of the distance to the BS, but dependent on the local environment such as the arrangement of the streets and buildings. The shadowing loss, Ls, depends on the micro-level location of the MS. Recent digital access technologies such as wideband CDMA mitigate the effect of instantaneous fading with a wide-band spectrum spread and accurate transmitter power control. However, propagation loss Lp, which dominates the near-far problem, is not influenced by the access technologies. Also, limits on transmit power for both BS and MS cause imperfect transmitter power control in the cell edge area, which results in frequent transmission error. Therefore, other approaches from the viewpoint of cellular architecture design should be studied.

3 MULTI-HOP RADIO ACCESS CELLULAR CONCEPT Multi-hop packet radio transmission is an effective technique for extending the range of communications at minimum power. In order to reduce interference from high-power communication users at the cell edge and to enhance area coverage, we propose the multi-hop radio access cellular (MRAC) concept. Figure 2 shows a graphical representation of the concept. It adds hop stations (HS) such as radio repeaters or wireless packet routers between users and corresponding BSs. The HS can be dedicated for relaying signals or any user terminal experiencing good propagation conditions can assume the role of the HS. The MSs can choose one of the two path modes. For the first, an MS can directly access a BS when the propagation conditions are sufficient to satisfy the communication requirements ( speed, error rate, etc.), we call this the single-hop path. When the propagation loss is high, the MS selects an available HS and communicates with the HS at a lower transmit power than that of a single-hop path. The HS relays the communication to the BS. This is called the double-hop path. By choosing the optimum path to the BS, the transmit power for the MS can be reduced for the same transmission speed, or the MS can communicate at higher speeds using the same transmit power. Utilizing these modes, results in decreasing the interference from the MS in the power domain or in the time domain, and enhances the cell size or area coverage by relaying the transmissions. Since future generation mobile communication systems will focus on IP-based service, a broadband packet transmission scheme will be employed. Considering this, the path selection can be performed packet by packet, and path diversity is effective for shadowing. Therefore we can expect further improvement by path diversity. In double-hop modes, the HS requires some extra radio resources for relaying transmissions. This may offset the interference reduction effect. When dedicated repeaters are deployed at the edges of a roof or at the corners of the sides of buildings, they are set to maintain the line of sight(los) to the BSs and directional antennas can be employed at both the HS and BS. Therefore, interference from the HS-BS hop is relatively low. On the other hand, if a user terminal is employed as an HS, the interference due to the HS-BS hop will be greater than that using the dedicated HS. However, in a highly populated urban area, many user-terminal-based HS candidates may be found. By choosing the best one, the total amount of interference from the MS-HS and HS-BS paths can be minimized. Other issues for employing a user terminal as an HS are realizing the relay function in the terminal, as well as applying the frequency spectrum to the BS-HS and HS-MS paths. In

4 order to avoid adding an extra transceiver and spectrum to the terminal, time division duplex relay (TDDR) operation is desirable. Although TDDR operation is limited with respect to the transmission speed due to transceiver time-sharing, it is mitigated by asymmetric speed TDDR operation. For example, when the propagation conditions of the HS-BS path are good, a higher transmission speed is possible for the path, which results in conservation of the forwarding time in the HS-BS path. Thus, a high percentage of time can be reserved for the HS-MS path communication and the speed can be maintained. Propagation Models Interference and area coverage performance in cellular systems strongly depend on the propagation conditions. Therefore, we should study what specific propagation models are suitable for evaluating the target system. Figure 3 shows the assumed propagation models. In the figure, the BS is typical cellular site with a relatively high antenna tower(~50 m high and the antenna gain of 20dBi), HS1 is a dedicated repeater placed at the edge of the roof of a building(~30 m high and the antenna gain of 15 dbi), and the MS is held by a human (height~ 1.5 m and the antenna gain of 0 dbi). HS2 is a user terminal repeater with the same height and antenna gain as that of the MS, but it is in LOS to the MS. Table 1: Propagation Models and Antenna Parameters Propagation loss model BS-HS1 Free space propagation BS-HS2 Urban mobile communication propagation considering multiple diffractions caused by buildings HS1-MS Urban mobile communication propagation considering multiple diffractions caused by buildings HS2-MS Two-ray(direct and ground-reflected way) propagation with line of sight assuming street cell micro cellular systems Antenna parameters Gain (dbi) Height (m) BS HS HS2 MS The proposed propagation models with antenna parameters are summarized in Table 1.The BS-HS1 path is considered to be free space. Urban area mobile propagation models are applied to the BS-HS2 and HS1-MS paths. HS2 and MS are assumed to be on a street, and the two ray (direct wave and ground reflected wave) propagation model is used. Figure 4 shows the propagation loss examples for the paths.

5 The curves in Fig. 4(a) represent the basic propagation losses versus distance (using an isotropic antenna for both transmission and reception). In order to analyze the effective range from the HSs, the antenna gains of the HSs should be taken into account. Thus, the antenna gains are added to the basic propagation losses and are shown in Fig. 4(b). Hereafter, these compensated losses referred to as transmission losses. From Fig. 4(b), it is clear that the transmission loss of the BS-HS1 path is lower than the other paths by 30 to 50 db. Therefore, the required transmit power is very low and the interference to other stations is not serious. On the other hand, the BS-HS2 and HS1-MS propagations are assumed to be normal mobile propagations with a loss exponent index of 3.8; thus the losses are relatively high. With regard to the HS2-MS path, its propagation loss in Fig. 4(a) is lower than that of the BS-HS2 because the two-ray propagation model has a LOS component. However, its transmission loss in Fig. 4(b) is higher than that of BS-HS2 due to low antenna gains of HS2 and the MS. Compared with the single-hop path, which is identical to the BS-HS2 path, the transmission loss of the HS1-MS path is higher because of its lower antenna height. This means that the double-hop area created by HS1 is narrower than the BS cell. However, HS1 can be located not only inside the BS cell area, but also outside, which is important to enhance area coverage.

6 MRAC DEPLOYMENT MODELS Based on the propagation models, suitable MRAC operation conditions are discussed in this section. Figure 5 shows the double-hop area created by HS1. HS1 is located 800 m. from the BS. All the transmission loss curves are copied from Fig. 4(b). As discussed in the last section, the transmission loss of the BS-HS1 path is very low and we can position HS1 anywhere. However, improvement by employing the double-hop transmission is remarkable when HS1 is located at the cell edge, since the transmission loss of the singlehop path is very high in that area and high transmit power is required. In this figure, the transmission loss of the HS1-MS (double-hop) path is lower than that of the BS-MS (single-hop) path when the MS is located at a distance from around 600 m. to 1100 m. Certainly, path diversity is effective at the crossover area at the distance of 600 m. and 1100 m.; therefore, the MRAC effective area (possible double-hop area) is depicted beyond the border. In this area, the transmit power required for communication can be reduced.the relative transmit power for each path, which is proportional to the transmission loss, is depicted in the same figure. They are P HM,P BH and P BM, which correspond to the BS-MS (single-hop) path, HS1-MS path and BS-HS1 path, respectively. In this area, P HM + P BH P BM (2) can be achieved. In order to evaluate interference generated from the paths, it is necessary to consider transmitting antenna gains. For the up-link, total radiated power can be reduced into the area that satisfies the condition: G M P HM + G H1 P BH G M P BM (3) Where GM and GH1 are antenna gains for the MS and HS1, respectively. Since the second term of the above inequality is small, the area is almost same as that for the condition (2). For the down-link, the area that satisfies the following condition generates less interference in MRAC: G H1 P HM + G B P BH G B P BM (4) Where GB is the antenna gain for the BS. Since the second term of the above inequality is small, and also G H1 G B (5) the condition (4)can be achieved easier than the condition (3). Thus, the interference can be reduced while enhancing the area coverage.

7 A similar discussion is possible for HS2, shown in Fig. 6. In this case, HS2 is a user terminal and the BS-HS2 path loss curve is identical to that of BS-MS (single-hop) path. In order to reduce the transit power for the MS by using the double-hop, HS2 is located at a better than the MS. For example, HS2 is located closer to the BS than the MS in this figure. Considering the same conditions in Equation(2),the MRAC effective area is around HS2 and further until the end of LOS. In this area, however, it is considered that the total transmit power reduction will be achieved mainly by the path diversity effect between the single-hop and double-hop paths because the macro scale power difference between the MS-BS path and HS2-BS path is not large The cell layout of the MRAC is shown in Fig. 7. The layout is very similar to honeycomb cells. The center area of each cell is a single-hop dominant area. The cell border area is the MRAC area, in which many HSs exist for relaying communications, and both single-hop and double-hop paths are used to access the BSs.

8 CONCLUSION In order to achieve high-speed/high-capacity and good area coverage at a reasonable cost, the Multi-hop Radio Access Cellular (MRAC) architecture was proposed for the Fourth Generation Mobile Communication Systems. Two kinds of hop stations were proposed and four propagation models were studied representing the paths among the base station, hop stations and mobile station. Based on the propagation models, suitable MRAC operation conditions were discussed. For mobile stations located at the cell edge area, MRAC operation is effective in reducing the transmit power. Consequently, MRAC reduces the interference in cellular systems and enhances the area coverage. Detailed protocol studies including a relay duplex scheme, MAC (Media Access Control) and a performance analysis on the interference reduction remain as a future work. REFERENCES [1] A.J.Titerbi, CDMA Principles of Spread Spectrum Communication, Addision-Wesley,Massachussets, 1995 [2] Radio Access Network Design Concept for the Fourth Generation Mobile Communication Systems By Y. Yamao [3] An Approach to 4 th Generation Wireless Infrastructures-Scenarios and Key Research Issues By M. Flament [4] Radio Propagation Characteristics for Line of Sight Micro cellular and Personal Communications By H.Xia [5] Digital Communications By John G. Proakis. [6] Mobile Cellular Telecommunications Analog and Digital systems 2e By William C.Y.Lee [7] Wireless Communication Principle and Practices By Rapport [8] [9] [10] [11] [12]