Performance Analysis of an Adaptive Soft Handoff Algorithm for Mobile Cellular Systems Huamin Zhu and Kyungsup Kwak Graduate School of Information Technology and Telecommunications Room 428, High-Tech Center, Inha University #253 Yonghyun-dong, Nam-gu, Incheon, 402-751, Korea zhu@inhaian.net, kskwak@inha.ac.kr Abstract. In this paper, an adaptive soft handoff algorithm, which dynamically calculates the threshold values for soft handoff based on the received signal strength, is proposed for mobile cellular communication systems. An analytical model is developed to study the performance of soft handoff algorithms. Per- formance is evaluated in terms of the average number active set updates, the mean size of the active set, the mean time of soft handoff, the average signal quality and link degradation probability, where a link degradation is defined to be the event that the signal strength falls below a level required for satisfactory communication. The adaptive soft handoff is shown to yield a significantly bet- ter tradeoff than the handoff algorithm with static thresholds. 1 Introduction Handoff is an essential component of mobile cellular communication systems [1] [2]. It is the process whereby a mobile station (MS) communicating with one base station (BS) is switched to another BS when the MS moves across a cell boundary during a call. A call in progress could be forced to abort during handoff if sufficient resources cannot be allocated in the new wireless cell. A properly designed handoff algorithm is essential in reducing the switching load of the system while maintaining the quality of service (QoS). The design of reliable handoff algorithms is crucial to the operation of a cellular communication system and is especially important in microcellular systems, where the MS may traverse several cells during a call. The decision to initiate a hand- off may be based on different measurements, such as the received signal strength (RSS) from the serving BS and neighboring BSs, the distance between the MS and the surrounding BSs, signal to noise ration (SNR), and bit error rate (BER). The RSS measurement is one of the most common criteria. There are two types of handoff: hard handoff and soft handoff. Hard handoff is a break-before-make method, where a new link is set up after the release of the old link. A certain amount of margin is introduced to eliminate the ping-pong effect, which is the scenario of repeated handoff between two adjacent BSs caused by rapid fluctua- tions in the RSS from both of the BSs. Soft handoff is a make-before-break method [3]-[6]. With soft handoff, an active set is maintained, which is the set of all the BSs