Optimal Control of Hando s in Wireless Networks

A Dynamic Programming formulation is used to obtain an optimal strategy for the handoo problem in cellular radio systems. The formulation includes the modeling of the underlying randomness in received signal strengths and of the mobile's movements. The cost function is designed such that there is a cost associated with switching and a reward for improving the quality of the call. The optimum decision is characterized by a threshold on the diierence between the measured power that the mobile receives from the base stations. Also we study the problem of choosing the \best" xed threshold that minimizes the cost function. The performance of the optimal and suboptimal strategies are compared. I Introduction Wireless networks are experiencing rapid growth, a trend likely to continue into the foreseeable future. In both micro and macro cellular networks, a key issue for eecient operation is the problem of handoos. A call on a portable/mobile which leaves one cell (radio coverage area) and enters a neighboring cell must be transferred to the base station of this neighboring (new) cell. Each handoo involves a signaling cost. Because of statistical uctuations in signal strength due to fading, a call may get bounced back and forth between neighboring base stations before it is either successfully handed oo, or is forced to terminate as the signal strength falls below acceptable levels. An improperly designed handoo algorithm can result in an unacceptably high level of bouncing (resulting in high signaling costs) and/or a high probability of forced termination. We argue that approaching the handoo problem in a stochastic control framework is most appropriate. We use a Markov decision process formulation, and derive optimal handoo strategies via Dynamic Programming (DP). Typically, in a cellular mobile communication network (analog or digital), each cell is assigned a separate set of channels (frequencies, carriers, or time slots). The assigned set depends on the frequency planning strategy used for spatial reuse, and may be xed or changed dynamically. A successful handoo entails not only the availability of a channel in the new cell (to which the mobile enters) but also an acceptable level of signal strength on the available channel. To focus mainly on the handoo issue, we take a simple y The work of these authors was supported partially through NSF Grant NSFD CDR-88-03012 and through NASA Grant NAGW277S. model of two adjacent cells with one channel per cell, and analyze the optimal handoo …