A SOFT HANDOFF ALGORITHM FOR CELLULAR MOBILE SYSTEM

A SOFT HANDOFF ALGORITHM FOR CELLULAR MOBILE SYSTEM.

ABSTRACT 

 A soft Hand-off Algorithm for cellular mobile system is investigated in this paper. Since there can be only one carrier frequency being used in CDMA System, a handoff scheme with diversity, a so called “Soft Handoff”, was proposed for higher communication quality and capacity. During the studies on Hand-off algorithm, it was felt that the Hand-off procedure had to be evaluated for such system.  To understand the Soft Handoff, A simulation model, result reveals that for higher traffic blocking and drop out probability increases with respective traffic but the hand-off probability decreases with respective traffic. Hence to improve the performance of it is suggested that F1/F2 Scheme should be used. By allotting different frequency in the adjacent cell i.e. Eb/It will be very much less in the system.

1.INTRODUCTION

The 3G of Cellular Mobile Network promises to provide vide variety of services such as voice, Data & Multimedia to users on the move. Mobile customers can make a phone call as in wireless telephone, to make a connection to retrieve information message such as E-mail or Stock information or to make a connection to surf the internet or to do business over the internet. Due to the Cellular nature of mobile communication these networks have been modeled using queuing networks, a cell is modeled as a queue and channels as servers with limited number of channels representing the capacity restrictions on the number of customers in the queue. The Handoff concept in the cellular mobile communication is modeled as customers routing from one queue to another. The major difference between mobile network and queuing network is that in cellular communication network the requesting call holding time of call is related to call. Whereas the service time of customers in queuing network is related to the queue. Each bay station is responsible for providing service to mobile users in its area, called a cell. Within each cell, there are usually two classes of call traffic : New calls and Handoff calls. A New call is one which initiates the current cell, while a handoff call is the one which initiates in the another cell, and handed over to the current cell. When a new calls originates in the current cell, if an unused channel is available. It will be assigned for the communication between the mobile user and the bay station and the new call is accepted for services. If no channel can be assigned to the new call, which depends on the channel allocation scheme. The new may be blocked and may be cleared from the system and in this case the actual connection time for this call is Zero. If a new call is blocked in one cell, this does not mean that the new call is blocked in the network.

Recently, CDMA has become a most promising technology for the future cellular network, due to its various advantages. Because there can be only one frequency being used in CDMA system, a handoff scheme with diversity, a so called “Soft Handoff”, is proposed for higher communication quality. I brief a handoff process in which the mobile unit can commence communication with a target bay station without interrupting the communication with the current serving base station is called “Soft Handoff”, i.e. “make before break”. The traditional handoff scheme which requires the mobile unit to break the communication with the current base station before establishing a new communication with the target base station is called “Hard Handoff”, i.e. “break before make”.  The handoff region of the hard handoff scheme is in general, very narrow. This causes frequent hard handoffs i.e. frequent communication breaks, a so called “Ping Pong Effect” when a mobile unit drives in and out of the cells boundary. This also introduces larger interference to other users. In soft handoff, since the procedure is “make before break”, no matter how frequently mobile unit drives in and out of the handoff region communication “breaks” Seldomly occur. Moreover the adoption of diversity reception in soft handoff also leads to a better communication quality without as large  a transmitting power as that required in hard handoff to maintain the broader handoff region. In other words, by using the soft handoff scheme and a proper power control strategy as proposed in IS-95, the require transmitter powers as well as interference can be reduced to improve the communication quality and system capacity. Further more, larger handoff regions of soft handoff also introduces a longer mean queuing time to get a new channel from the target base station. In this paper handoff procedure without queue is presented.

2.PROBLEM FORMULATION:

In the simulation model we assume that one mobile unit carries one call only. It means that there is no bulk handoff arrival. We only consider that there are at most two different sources in diversity reception. It cell will reserve C_h channels out of a total of C available channels exclusively for handoff calls, a so called cut off priority, because a suddenly forced termination during a call session will be more upsetting than a failure to connect every handoff requirement is assumed to be perfectly detected in our simulation model and assignment of channel is instantaneous if channel is available.

 To evaluate the performance of soft handoff for CDMA System a simulation model is developed. In case of CDMA there are two schemes for the reuse of the channels. In first case all the cells will have the same frequency in a cluster. To improve over interference an F₁/F₂ Scheme is deployed. For our simulation we have considered reuse distance as 1 i.e. same frequency is used in adjacent cell. Since handoff is to be tested a two cell model is assumed which reflects the performance of entire system. For the simulation purpose a mobile assumed to be moving in a straight line only.

The handoff mechanism is to be based on Eb/It, which is computed on basis of signal strength. To compute the signal strength of the mobile we use an Eight ray model.

Theoretically the CDMA is a infinite capacity system, but in practice it is interference limited, hence number of users served by a cell at a cell at a single time is used to be 64 maximum for simulation purpose. We assume the call arrival ‘Poisson’s Process’ with mean call arrival. Call holding time is assumed to be 180 sec.

3. FLOW CHART

 

Fig. 2. Call Initialization              Fig.3. Updating call parameters and hand off

                                                                                         

PARAMETERS	VALUE
Specific Eb/It ratio (Eb/It)sp	5
Data rate (BPS)	9600
Band width of CDMA channel (MCPS)	1.23 x 106
Voice activity factor (VF)	0.5
Hysterisis for Handoff	2
Lower Threshold (LT)	-82
Call departure rate (l)	10
Mean call holding time (Sec)	180
Radius of the cell in meter (R)	1000
Number of calls (RN)	10000
Max. Velocity in m/sec (Vmax)	2

Table 1: Simulation Conditions

Following criteria shall be used to measure the performance of the handoff scheme

1. P_r (blocking)=

Where: C-no. of trunked channel offered by a trunked

radio system.

A- total offered traffic.

2. A_u =lH

Where : H-average duration of call.

l - the average no. of call requests per unit time for each user.

Fig. 4.Variation of blocking Probability

Fig. 5.Variation of Handoff Probability

Fig. 6.Variation of dropout Probability .

4.Result and discussion

From the simulation module for the traffic 1 to 2 Erlang.fig.4 shows blocking probability vs traffic. Here we observed that the blocking probability always increases linearly with respect to traffic. With reference to handoff this simulation result shown in fig.5 indicate that the hand off probability decreases with respect to traffic.

This is because of the no of calls generated are denied the service. Fig.6 shows the dropout probability for the handoff algorithm for lower traffic the dropout probability negligibly small, where as higher traffic the dropout probability is large. This shows that for higher traffic the blocking and dropout probability increases. Hence to improve the performance it is suggested that the f1/f2 scheme should be used.

5.Conclusion

From the simulation result we observed that the reused ratio of 1 the blocking probability and dropout probability is high but this performance can be improved by allotting different frequency in the adjacent cell i.e. Eb/It will be very much less in the system.

6.References

1. Fumihiro Urasawa, ken-ichi itoh, and laien-shang,’’Analysis and simulation result of multihof handoff scheme in an Ad- Hoc wireless network,’’ in IEEE Trans. Vol.8, PP. 1366-1369, 2002.
2. N.zhang and J. Holtzman,‘’Analysis of handoff algorithm using both absolute and relative measurement’’, in IEEE Trans. Veh. Technol, vol.45, no.1,PP.174-179, 1996,2.
3.        Szu-Lin, Jen-yeu chen, and Jana-Hwa Huang, ‘’ Performance Analysis of soft handoff in CDMA cellular networks’’, in IEEE journal,. Vol.14, No.9, PP.1762-1769, Dec.1996. 
4.  Wei Li,and Xiuli Chao,“modeling and performance evaluation of a cellular mobile networks”, in IEEE, Trans. Vol. 12, no.1, PP.131-145,Feb.2004.
5. Wei Li, Yuguang fang and Robert R, “ Actual call connection time characterization for wireless mobile network under a general channel allocation scheme”, in IEEE, Trans., Vol.1, no.4, PP.682-691, Oct.2003.