Four-Sector Cross-Shaped Urban Microcellular Systems with Intelligent Switched-Beam Antennas

Transcription

1 592 IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, VOL. 50, NO. 2, MARCH 2001 Four-Sector Cross-Shaped Urban Microcellular Systems with Intelligent Switched-Beam Antennas Ho-Shin Cho, Member, IEEE, Jae Hoon Chung, Student Member, IEEE, and Dan Keun Sung, Senior Member, IEEE Abstract A four-sector cross-shaped urban microcellular system with intelligent switched-beam antennas is proposed. Each sector covers a street block with a base station located at an intersection, and an intelligent beam-switching scheme is used to locate mobile users in the most suitable beam coverage. Due to directional narrow-beam patterns and waveguide effects of tall buildings, radio signals along vertical and horizontal streets do not interfere with each other. Therefore, a channel can be reused simultaneously in multiple neighboring cells as long as cochannels do not encounter each other along the line of sight. The proposed scheme has a channel reuse efficiency of 0.95 for a traffic load of 0.02 [new-call arrivals/s/cell]. The system also increases system capacity more than three times with a blocking probability of 1% and considerably reduces handoff traffic when compared with a conventional cross-shaped microcellular system with an omnidirectional beam pattern. Index Terms Cross-shaped cell, intelligent switched-beam antenna, reuse efficiency, system capacity. I. INTRODUCTION CAPACITY demands for mobile communications are rapidly increasing. One solution to this problem is to increase radio resource utilization. Microcell-based systems have been proposed in order to augment spectrum utilization using a short cochannel reuse distance, especially in urban areas [1] [4]. At the same time, sectored cells with directional antennas have also been studied in order to reduce cochannel interferences and to increase system capacity [5], [6]. Urban microcellular and macrocellular systems have different features. Radio signals in tall building environments experience difficult propagation phenomena such as fading, shadowing, and corner effect [7] [11]. These phenomena cause urban microcell shapes to approximate street patterns [12] [14]. Both a cross-shaped and a cigar-shaped microcell pattern have been considered based upon the position of a base station [15], [16]. Therefore, the classical channel reuse scheme in a cluster of seven cells may not be valid in an urban microcellular environment [17]. Moreover, unpredictable radio propagation in urban areas requires real-time channel control based upon measured data in order to minimize cochannel interferences [18]. Three- and six-sector cells have been considered in hexagonal macrocell environments [19], [20]. Sectored cells can reduce cochannel interference using directional antennas and thus can increase the carrier to interference ratio. However, sectored cells may generate undesirable intersector handoffs, which may be more severe in microcellular systems. Moreover, the three- and six-sector cell structures are not suitable for cross- and cigarshaped urban cellular environments. Lopez et al. [21] and Ho et al. [22] used a switched-beam intelligent antenna with trunkpool techniques to reduce the number of intersector handoffs and to increase system capacity. This scheme manages a user s intersector movement by simple beam switching instead of using a complex handoff procedure. Channel switching concepts were also proposed by Lee [23] and Cho et al. [24]. In this paper, a four-sector cross-shaped urban microcellular system with intelligent switched-beam antennas is proposed. Four directional narrow-beam antennas, i.e., one beam per sector, are used. Intersector handoffs can be avoided using trunkpool techniques [22]. Due to directional narrow-beam patterns and waveguide effects of tall buildings, radio signals along vertical and horizontal streets do not interfere with each other. Therefore, a channel can be reused simultaneously in multiple neighboring cells as long as cochannels do not encounter each other along the line of sight. Moreover, some intercell handoffs can be replaced by simple intercell beam-switching if all time clocks of base stations (BSs) are synchronized. The proposed system is evaluated and compared with a conventional cross-shaped cellular system with omnidirectional antennas in terms of channel reuse efficiency, blocking probability, intercell/intracell handoff rates, and intercell/intracell beam-switching rates. In this analysis, typical low-tier user mobility patterns in urban areas, such as straight movement along streets and 90 turns at intersections or driveways, are considered. The rest of this paper is organized as follows. Section II describes the proposed system; Section III evaluates system performance using a multidimensional Newton method; Section IV gives numerical examples; and Section V presents conclusions. Manuscript received May 6, 1998; revised March 29, H.-S. Cho was with the Department of Electrical Engineering, Korea Advanced Institute of Science and Technology, Taejon, Korea. He is now with the School of Electronics, Telecommunications and Computer Engineering, Hankuk Aviation University, Koyang, Korea. J. H. Chung and D. K. Sung are with the Department of Electrical Engineering, Korea Advanced Institute of Science and Technology, Taejon, Korea. Publisher Item Identifier S (01) II. SYSTEM DESCRIPTION A. Architecture A grid-structured urban area like Manhattan is considered. Fig. 1 illustrates a four-sector cross-shaped urban microcellular system using intelligent switched-beam antennas. Each beam covers a street block, which is considered to be a sector. A BS /01$ IEEE

2 CHO et al.: FOUR-SECTOR CROSS-SHAPED URBAN MICROCELLULAR SYSTEMS 593 Fig. 2. Cochannel interference model. agation in urban areas [2], [25] [28], the average received power over multipath fading is expressed as (1) where transmitter power; ; Fig. 1. Architecture of a four-sector cross-shaped urban microcellular system using intelligent switched-beam antennas. is located at every other intersection, thereby yielding crossshaped cells. An intelligent switched-beam antenna, which is implemented in a BS, controls the connections between radio channel units and antenna beams to provide a mobile user with any radio channel in a cell independent of the current sector. Every BS needs to exchange information on channel assignment states with the neighboring BSs in four directions: up, down, left, and right. B. Radio Propagation Model It is assumed that radio signals that propagate along vertical and horizontal streets do not interfere with each other due to a directional narrow-beam pattern from the BS and waveguide effect of neighboring tall buildings. In real environments, however, there are some signal leakages between tall buildings. These signal leakages may interfere with cochannel users staying in neighboring perpendicular streets. However, the amount of cochannel interference may be negligible. To verify this argument, we numerically compare the received signal strength from the home BS with cochannel interfering signal strength from a neighboring BS. In Fig. 2, BS_I is a home BS and BS_II is an interfering BS. A line-of-sight (LOS) exists between the caller and the home BS, but not between the caller and the interfering BS. Based on previous studies on LOS prop- breakpoint distance ( and are the transmit and receive antenna heights, respectively, and is the wavelength) [14], [28], [29]; distance between BS and a caller; log-normal random variable with a standard deviation of approximately 3 db [8]. Based on previous experimental data [8], [29], [30] and a diffraction-based theoretical model [14], a simple empirical model for the received power under urban NLOS propagation conditions yields where shortest distance from BS to a caller; attenuation exponent; log-normal random variable with a standard deviation of approximately 5 db [8]. Therefore, the ratio of the caller s received power for the desired signal to that for the cochannel interfering signal is given by where is a block distance and is the distance between a caller and home BS. is also a random variable with a Cauchy density function (2) (3) (4)

3 594 IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, VOL. 50, NO. 2, MARCH 2001 Fig. 3. Cochannel error probabilities with narrow-beam directional and omnidirectional antennas. Then, the cochannel error probability that this ratio is less than a predetermined threshold can be expressed as (5) Fig. 3 shows the cochannel error probability on forward link for (for 900 MHz), m, m,, and db according to the caller s position when the directional beam antenna gain of BS is 20 db. The dotted line and the solid line in Fig. 3 represent the cochannel error probabilities with narrow-beam directional and omnidirectional antennas, respectively. Both of the cochannel error probabilities are maintained below 0.1%, which is tolerable, and the assumption that radio signals that propagate along vertical and horizontal streets do not interfere with each other is acceptable. C. Channel Assignment Scheme A radio channel used in a given sector can be simultaneously reused in its neighboring cells, excluding an interfering sector, as shown in Fig. 4. A BS checks a lookup table, which has channel assignment states of the interfering sectors included in adjoining cells, and then determines whether a radio traffic channel is available considering cochannel interferences from adjoining cells. Fig. 5 shows an example of the lookup table. Fig. 4. Reusable sectors and an interfering sector when a channel is used in a given sector. Whenever channel assignment states change, the BS informs the four adjoining cells of the changes and the lookup tables of the adjoining cells are updated. D. Call Management Schemes 1) Intracell Beam Switching: An intersector movement within a cell is called intracell migration in this paper. When a caller moves into a new sector, a BS switches the connection between a radio channel unit and an antenna beam in order to

4 CHO et al.: FOUR-SECTOR CROSS-SHAPED URBAN MICROCELLULAR SYSTEMS 595 (a) (b) Fig. 5. An example of a lookup table for the channel assignment states of four neighboring cells: (a) lookup table and (b) channel assignment state. cover the caller using the most suitable antenna beam. An intracell migrationcallthatdoesnotencounteranycochannelinterferences from adjoining cells holds the current channel using simple antenna beamswitching, asshowninfig. 6. This procedure iscalled intracell beam switching. During an intracell beam switching, callers do not recognize any changes, and therefore no signaling messageisneededbetweenacallerandabs. 2) Intracell Handoff: If an intracell migration call encounters any cochannel interferences from neighboring cells, then a new channel that does not cause any cochannel interference is assigned to the caller through a new antenna beam, as shown in Fig. 7. This procedure is called an intracell handoff, which may require signaling messages between a caller and a BS. 3) Intercell Beam Switching: Crossing a cell boundary and entering a new cell that has been referred as a handoff yields an intercell migration in this paper, as shown in Fig. 8. An intercell migration call requires a seamless service from a new BS. In conventional cellular systems, a handoff procedure tries to change the current channel to a new one that belongs to a new BS to keep the call going. On the other hand, all BSs share the same channel group in this system. Therefore, if an intercell migration call finds that the current channel is idle in the target cell and that does not cause any cochannel interferences to callers staying in neighboring cells, then the call does not need to change the current channel. In synchronized systems where all time clocks of BSs are synchronized, an intercell migration call can maintain the current channel with a few signaling messages. We call this procedure an intercell beam switching, which may save signaling traffic loads, especially on radio links, compared to conventional complex handoff procedures. 4) Intercell Handoff: As shown in Fig. 9, if an intercell migration call encounters cochannel interference in a new cell or causes cochannel interference to a caller in a neighboring cell, a new available channel must be assigned to the call through a conventional handoff procedure. This is called an intercell handoff in this paper.

5 596 IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, VOL. 50, NO. 2, MARCH 2001 Fig. 6. An example of an intracell beam switching. Fig. 8. An example of an intercell beam switching. A caller turning a corner [case (1)] or moving straight through an intersection within a cell [case (2)] requires either an intracell beam switching or an intracell handoff. On the other hand, crossing a cell boundary [cases (3) and (4)] requires either an intercell beam switching or an intercell handoff. Movement into a building [case (5)] requires a handoff to an indoor wireless system. It is assumed that all buildings have private indoor wireless systems, which interact with public outdoor systems through intersystem handoffs. In this analysis, only a public outdoor system is considered. Out-of-building calls are considered to be identical to new calls, and into-building calls cause a channel release. Cases (1), (2), and (6) represent call terminations within a cell. III. SYSTEM ANALYSIS Fig. 7. An example of an intracell handoff. Table I summarizes the four call-management schemes: 1) intracell beam switching, 2) intracell handoff, 3) intercell beam switching, and 4) intercell handoff. E. User Mobility Low-tier users in urban areas typically move linearly along streets and sometimes enter buildings or turn 90 at intersections. Fig. 10 shows typical user motions in a cross-shaped cell. In the proposed system, every cell uses a common channel set in order to increase a reuse efficiency of radio resources. Therefore, a call may be blocked even though there are idle channels in a cell if the idle channels are all occupied in the interfering sectors of neighboring cells. A call blocking probability and a channel reuse efficiency are important performance issues in this paper. The call blocking probability is evaluated in terms of call movement probabilities, call arrival rates, call service rates, and channel steady-state probabilities. In order to analyze the proposed system, the following assumptions are made.

6 CHO et al.: FOUR-SECTOR CROSS-SHAPED URBAN MICROCELLULAR SYSTEMS 597 TABLE I COMPARISON OF FOUR CALL-MANAGEMENT SCHEMES 1) New calls, including out-of-building calls, are uniformly generated in a cell according to a Poisson process with a rate of. 2) The call holding time is exponentially distributed with a mean of 1. 3) Users move with a constant speed of. 4) No priority is given to handoff calls. A. Modeling Fig. 11 shows a mathematical user movement model in a cross-shaped cell. denotes the block distance. The distance from a call generation point to the nearest cell boundary is a random variable. Random variable represents the distance from a call generation point to the cell boundary in the caller s direction of movement. Random variable is the distance from a call generation point to a building gate through which the caller enters a building. The variable is assumed to be exponentially distributed with a mean of. B. Intercell Migration Probabilities Let denote the system state to represent the number of channels in use in a cell, the total number of channels in a system. An intracell migration call requires a handoff in a new sector when a cochannel user is staying in the interfering sector. The probability that an intracell migration call requires an intracell handoff is written as (6), shown at the bottom of the page. A new call requires an intercell migration if a caller does not enter any building and the call lasts until the caller crosses a cell boundary. Therefore, the probability that a new call requires an intercell migration at state is given by (7), also shown at the bottom of the page, where is a probability that an intracell migration call fails in handoff at state, and is later given by (17). The cumulative distribution functions,, and in (7) are given by (8) The state of the adjoining cell which includes the interfering sector The current channel is used at the interfering sector (6) (7)

7 598 IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, VOL. 50, NO. 2, MARCH 2001 Fig. 10. Typical user motions in a cross-shaped cell structure. Fig. 9. An example of an intercell handoff. (9) (10) Similarly, the probability that an intercell migration call requires another intercell migration at state is written as (11) where represents the probability that an intercell immigration call does not terminate and that the caller does not enter buildings until another intercell migration. represents the probability that a caller succeeds in an intracell migration without call dropping. C. Other Probabilities If an intracell migration call encounters any cochannel interference from a new sector s neighboring cells, the BS replaces Fig. 11. A mathematical user movement model in a cross-shaped cell. its current channel with a new available channel. A new call requires an intracell handoff under the following conditions: 1) The user moves in direction, as shown in Fig ) The user does not enter any building and the call lasts until an intracell migration. 3) The current call requires a new channel because a beam switching is unavailable.

8 CHO et al.: FOUR-SECTOR CROSS-SHAPED URBAN MICROCELLULAR SYSTEMS 599 Then, the probability that a new call requires an intracell handoff is derived as (12) where is the probability that an intracell migration call requires an intracell handoff and is the probability that a caller moves in direction from a call generation point. Similarly, the probability that an intercell migration call requires an intracell handoff can be obtained as Fig. 12. Intercell migration movement. (13) (15) Among intercell migration calls, the calls whose channels are idle in a new cell can maintain the current channels through intercell beam switching if they do not encounter cochannel interference from interfering sectors. Case of an intercell migration shown in Fig. 12 is processed by beam switching only if the current channel is idle in Cell B. On the other hand, Case is processed through beam switching between Cell A and Cell C (Cell D) if the current channel is idle in both Cell C (Cell D) and the north sector (south sector) of Cell D (Cell C). It is assumed that a direction choice has an equal probability of 1/3 for directions of,, and. Then, the probability that an intercell migration call experiences intercell beam switching is given by Therefore, the blocking probability is written as (16) Similarly, and (the probability that an intercell migration call fails in both a beam switching and a handoff attempt at state ) can be obtained as (17) (14) where the first term is for case and the second term is for cases and. In the proposed system, a call may be blocked even though there are idle channels in the current cell if all idle channels are occupied in an interfering sector. A new call generated at state is blocked under the following conditions: 1) the state of the adjoining cell that includes an interfering sector is at least larger than ; 2) all idle channels of the current cell are occupied in the interfering sector. The call blocking probability at state is expressed as (18) Notations related to all the probabilities are summarized in Table II. D. Migration Call Rate The mean intercell migration call rate is represented by the probability law [31]. where an intercell migration call rate at state, by (19),is given

9 600 IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, VOL. 50, NO. 2, MARCH 2001 TABLE II NOTATIONS OF ALL PROBABILITIES Equation (19) can be rearranged as (20) As mentioned earlier, all intercell migration calls do not experience handoffs. The arrival rate of intercell handoff calls is given by 1) Since the variable is equal to, is obtained from (8) as (26) 2) The variable is equal to, and the variable has a uniform distribution (0, 2 ) for new calls and a constant value of 2 for intercell migration calls. Therefore, has a probability density function (pdf) for new calls and for intercell migration calls, as shown in Fig. 13. is given by (21) Therefore, the arrival rate of intercell beam-switching calls is expressed as The arrival rate of intracell handoff calls is given by (22) for new calls (27) for intercell migration calls where is a unit step function. The average service rate, which is the reciprocal of average channel holding time, is expressed as (28) (23) E. Channel Holding Time There are three cases where a caller releases a channel: 1) call completion, 2) upon entering a building, and 3) migration to another cell. The channel holding time is derived as (24) where and denote the cell sojourn times for building entrance and intercell migration calls, respectively. Therefore, the cumulative distribution function of is expressed as F. Steady-State Probability The system considered herein is represented by an model characterized by Poisson call arrivals, generally distributed service time, finite channels, and no queue. The steady-state probability of the system can be obtained by the balance equation for an system [32], [33]. The steady-state probability at state can be obtained as (29) (30) In (25), the cumulative distribution functions are derived as follows. (25) and where is the total offered load and is given by [34]

10 CHO et al.: FOUR-SECTOR CROSS-SHAPED URBAN MICROCELLULAR SYSTEMS 601 (a) Fig. 13. The pdf of T : (a) new calls and (b) intercell migration calls. (b) and and denote the average service rates for new calls and intercell migration calls, respectively. A Newton method is used to evaluate the steady state probability. Equations (29) and (30) can be represented as a function in the following form: TABLE III LIST OF SYSTEM PARAMETERS (31) Let and be the column vectors representing the following: (32) (33) where represents the number of iterations in the Newton method. Then, the Newton method [35] is applied to obtain a recursive relation where is a Jacobian matrix of and is given by (34) Fig. 14. Channel reuse efficiency. in the proposed system is actually less than one. The channel reuse efficiency can be redefined as Maximum number of available channels in a cell Total number of channels in a system With an initial vector, a convergent value of can be obtained after several iterations. IV. NUMERICAL EXAMPLES The parameter values used in numerical examples are listed in Table III. In general, the channel reuse efficiency can be defined as a ratio of the number of channels assigned to one cell to the total number of channels in the system. For example, four and seven cell cluster systems have a channel reuse efficiency of 1/4 and 1/7, respectively. In the proposed system, a cell manages all system channels of. However, a call can be blocked before channels are fully occupied because of cochannel interferences from neighboring cells. Therefore, the channel reuse efficiency (35) The definition (35) is coincident with a general case. Fig. 14 shows the channel reuse efficiency versus the new call arrival rate. As the traffic load becomes heavier, calls are more likely to experience interference, and therefore, the channel reuse efficiency decreases. The channel reuse efficiency, however, is much higher than for conventional fourand seven-cell cluster systems. Fig. 15 shows the blocking probabilities [which are represented in (16)] versus the call holding time for various. As the call holding time increases the traffic load increases, and, therefore, the blocking probability increases. The greater the value of, the greater the blocking probability because the channel holding time becomes longer.

11 602 IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, VOL. 50, NO. 2, MARCH 2001 Fig. 15. Blocking probability versus call holding time. Fig. 17. Comparison with the conventional cross-shaped cellular system in terms of blocking probability. Fig. 18. Call arrival rates. Fig. 16. A possible channel assignment scheme in the cross-shaped cell system with a frequency reuse efficiency of 0.5: dark cells for channel group A and bright cells for channel group B. The proposed system is now compared with a conventional cross-shaped cell system with an omnidirectional antenna in terms of blocking probability and handoff call rates. A channel assignment scheme in the conventional cross-shaped cell system is considered to have a channel reuse efficiency of 0.5, as shown in Fig. 16. The blocking probability and the handoff call rate of the conventional cross-shaped cell can be obtained by referring to [36]. Fig. 17 illustrates the blocking probabilities of the proposed system and the conventional cross-shaped cell system for. With a blocking probability of 1%, the proposed system can accommodate new calls with, while the conventional cross-shaped cell system has. This example shows a system capacity increase of approximately 310%. Fig. 18 shows various call arrival rates versus new call arrival rate. ( ) [which is represented in Fig. 19. The probability P. (20) (22)] corresponds to the handoff call rate in conventional cellular systems. In the proposed system, some intercell migration calls ( in Fig. 18) can be handled by simple intercell beam switching instead of by handoffs. [which is represented in (14)] is shown in Fig. 19. Therefore, the handoff call

12 CHO et al.: FOUR-SECTOR CROSS-SHAPED URBAN MICROCELLULAR SYSTEMS 603 rate in the proposed system is less than for the conventional cross-shaped cell system. Although there is additional handoff traffic load ( ) due to intracell handoff calls, the proposed system reduces handoff traffic load compared with the conventional cross-shaped cell system because. V. CONCLUSION A four-sector cross-shaped urban microcellular system with intelligent switched-beam antennas is proposed and evaluated. In the proposed system, a radio traffic channel can be reused simultaneously in multiple adjoining cells and intra/intercell handoffs can be handled with simple beam switching between the corresponding beams. Although intracell handoffs occur, the additional traffic load is insignificant when compared with the gain obtained by reducing intercell handoffs. The proposed system increases system capacity more than three times with a blocking probability of 1% compared with a conventional cross-shaped microcellular system, which has an omnidirectional antenna beam pattern. REFERENCES [1] J. Shapira, Microcell engineering in CDMA cellular networks, IEEE Trans. Veh. Technol., vol. 43, no. 4, pp , [2] M. V. Clark, V. Erceg, and L. J. Greenstein, Reuse efficiency in urban microcellular networks, IEEE Trans. Veh. Technol., vol. 46, no. 2, pp , [3] L.-C. Wang, G. L. Stüber, and C.-T. Lea, Architecture design, frequency planning, and performance analysis for a microcell/macrocell overlaying system, IEEE Trans. Veh. Technol., vol. 46, no. 4, pp , [4] R. Steele and M. Nofal, Teletraffic performance of microcellular personal communication networks, Proc. Inst. Elect. Eng., ser. I, vol. 139, no. 4, pp , Aug [5] W. C. Y. Lee, Elements of cellular mobile radio systems, IEEE Trans. Veh. Technol., vol. VT-35, pp , [6] M. Mahmoudi and E. S. Sousa, Sectorized antenna system for CDMA cellular networks, in Proc. IEEE VTC, 1997, pp [7] K. Mahbobi, Radio wave propagation in urban microcellular environment, in Proc. IEEE Vehicular Technology Conf., 1992, pp [8] A. J. Goldsmith and L. J. Greenstein, A measurement-based model for predicting coverage areas of urban microcells, IEEE J. Select. Areas Commun., vol. 11, pp , Sept [9] H. H. Xia, H. L. Bertoni, L. R. Maciel, A. L. Stewart, and R. Rowe, Microcellular propagation characteristics for personal communications in urban and suburban environments, IEEE Trans. Veh. Technol., vol. 43, pp , Aug [10] L. R. Maciel and H. L. Bertoni, Cell shape for microcellular systems in residential and commercial environments, IEEE Trans. Veh. Technol., vol. 43, pp , May [11] B. H. Fleury and P. E. Leuthold, Radiowave propagation in mobile communications: An overview of European research, IEEE Commun. Mag., vol. 34, pp , Feb [12] J. B. Anderson, T. S. Rappaport, and S. Yoshida, Propagation measurement and models for wireless communications channels, IEEE Commun. Mag., pp , Dec [13] L. J. Greenstein, N. Amitay, and T.-S. Chu et al., Microcells in personal communication systems, IEEE Commun. Mag., pp , Dec [14] V. Erceg, A. J. Rustako, and R. S. Raman, Diffraction around corners and its effects on the microcell coverage area in urban and suburban environments at 900MHz, 2GHz, and 6GHz, IEEE Trans. Veh. Technol., vol. 43, pp , Aug [15] R. Steele, J. Williams, D. Chandler, S. Dehghan, and A. Collard, Teletraffic performance of GSM900/DCS1800 in street microcells, IEEE Commun. Mag., vol. 33, pp , Mar [16] M. Benlib, GSM cell engineering and its performance, in Proc. 96 Wireless Commun. Workshop, Korea, 1996, pp [17] M. Frullone, C. Passerini, P. Grazioso, G. Riva, and G. Falciasecca, Advanced frequency planning criteria for second generation cellular radio systems, in Proc. ICT, Instanbul, Turkey, Apr. 1996, pp [18] M. Frullone, G. Riva, P. Grazioso, and G. Falciasecca, Advanced planning criteria for cellular systems, IEEE Personal Commun., vol. 3, pp , Dec [19] W. C. Y. Lee, Mobile Cellular Telecommunications Systems. New York: McGraw-Hill, [20] A. Hać, Wireless and cellular architecture and services, IEEE Commun. Mag., vol. 33, pp , Nov [21] A. R. Lopez and Hazeltine, Performance predictions for cellular switched-beam intelligent antenna systems, IEEE Commun. Mag., vol. 34, pp , Oct [22] M. J. Ho, G. L. Stüber, and M. D. Austin, Performance of switched-beam smart antennas for cellular radio systems, IEEE Trans. Veh. Technol., vol. 47, pp , Feb [23] W. C. Y. Lee, Applying the intelligent cell concept to PCS, IEEE Trans. Veh. Technol., vol. 43, pp , Aug [24] H. S. Cho, S. H. Kang, and D. K. Sung, A movable safety zone scheme in urban fiber-optic microcellular systems, IEEE Trans. Veh. Technol., vol. 48, pp , July [25] P. Hearley, Short distance attenuation measurements at 900MHz and 1.8GHz using low antenna heights for microcells, IEEE J. Select. Areas Commun., vol. 7, pp. 5 10, Jan [26] L. B. Milstein et al., On the feasibility of a CDMA overlay for personal communication networks, IEEE J. Select. Areas Commun., vol. 10, pp , May [27] H. H. Xia et al., Microcellular propagation characteristics for personal communications in urban and suburban environments, IEEE Trans. Veh. Technol., vol. 43, pp , Aug [28] M. J. Feuerstein et al., Path loss, delay spread, and outage models as functions of antenna height for microcellular system design, IEEE Trans. Veh. Technol., vol. 43, pp , Aug [29] V. Erceg, A. J. Rustako, and R. S. Roman, Urban/suburban out-of-sight propagation modeling, IEEE Commun. Mag., vol. 30, pp , June [30] A. J. Rustako, V. Erceg, R. S. Raman, T. M. Willis, and J. Ling, Measurements of microcellular propagation loss at 6GHz and 2GHz over nonline-of-sight paths in the city of boston, in Proc. GLOBECOM 95 Conf., pp [31] R. B. Cooper, Introduction to Queueing Theory. New York: Macmillan, [32] R. W. Wolff, Stochastic Modeling and the Theory of Queues. Englewood Cliffs, NJ: Prentice-Hall, [33] D. Hong and S. S. Rappaport, Traffic model and performance analysis for cellular mobile radio telephone systems with prioritized and non prioritized handoff procedures, IEEE Trans. Veh. Technol., vol. VT-35, pp , Aug [34] R. B. Cooper, Introduction to Queueing Theory, 2nd ed. New York: Elsevier Science, [35] R. L. Burden and J. D. Faires, Numerical Analysis. Boston, MA: Prindle, Weber Schmidt, [36] H. S. Cho, M. Y. Chung, S. H. Kang, and D. K. Sung, Performance analysis of cross- and cigar-shaped urban microcells considering user mobility characteristics, IEEE Trans. Veh. Technol., vol. 49, pp , Jan Ho-Shin Cho (S 92 M 99) received the B.S., M.S., and Ph.D. degrees in electrical engineering from the Korea Advanced Institute of Science and Technology in 1992, 1994, and 1999, respectively. From March 1999 to February 2001, he was a Senior Member of the Research Staff with the Electronics and Telecommunications Research Institute, where he was involved in developing a base station for IMT In March 2001, he joined the faculty of the Hankuk Aviation University, where he is currently a Lecturer in the School of Electronics, Telecommunications, and Computer Engineering. His current research interests include radio resource management, traffic modeling, radio propagation, and optical feeder and intelligent antenna schemes in wireless mobile communication systems.

13 604 IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, VOL. 50, NO. 2, MARCH 2001 Jae Hoon Chung (S 97) received the B.S. degree in electronics engineering from Yonsei University, Seoul, Korea, in 1997 and the M.S. degree in electrical engineering from the Korea Advanced Institute of Science and Technology (KAIST), Taejon, Korea, in Since March 1997, he has been a Teaching and Research Assistant in the Department of Electrical Engineering and Computer Science, KAIST. His research interests include mobility modeling, radio resource allocation schemes, indoor wireless communication systems, radio access networks of IMT-2000, and W-CDMA FDD/TDD systems. Dan Keun Sung (S 80 M 86 SM 00) received the B.S. degree in electronics engineering from Seoul National University, Seoul, Korea, in 1975 and the M.S. and Ph.D. degrees in electrical and computer engineering from the University of Texas at Austin in 1982 and 1986, respectively. From May 1977 to July 1980, he was a Research Engineer with the Electronics and Telecommunications Research Institute (ETRI), where he was engaged in various projects, including the development of an electronic switching system. In 1986, he joined the Faculty of the Korea Advanced Institute of Science and Technology (KAIST), where he is a Professor in the Department of Electrical Engineering and Computer Science. He was Director of the Satellite Technology Research Center(SaTReC) of KAIST from 1996 to He was Division Editor of the Journal of Communications and Networks from 1998 to 1999 and is currently an Editor. His research interests include ATM switching systems, traffic control, mobile communication networks, signaling networks, intelligent networks, performance and reliability of communication systems, and microsatellites. Prof. Sung is a member of IEICE (Japan), IEEK, KISS, KICS, Phi Kappa Phi, and Tau Beta Pi. He was Chairman of the Meeting and Conference Committee at the Asia Pacific Region of the IEEE COMSOC from 1998 to 1999.