| With the deployment of 2G and 3G cellular | | | | normalization policy has been proposed when |
| networks and WLAN, WiMax world widely, the | | | | calculate the quality function value of different |
| next generation wireless networks are envisioned | | | | networks. The inner QoS demands of the |
| as heterogeneous networks with all sorts of | | | | application are considered in the marks of relative |
| wireless networks coexisting. For the diversity of | | | | quality parameters. This normalization means has |
| these networks, each network has its in point | | | | good fairness among all networks quality |
| applications and scenarios. How to fully utilize | | | | parameters. This article analyses handoff decision |
| ubiquitous heterogeneous networks' potential | | | | problem. Because the RSSs that are from |
| advantages has become many research | | | | different base stations vary greatly by distance |
| personnel's pursued goals. Vertical handoff among | | | | and fading, mobile terminal has to keep connected |
| heterogeneous networks and horizontal handoff | | | | and minimize the ping-pong times. Handoff decision |
| are important contents of mobility management in | | | | should decide when to trigger the handoff and |
| mobile communications. Next generation wireless | | | | which is the best objective base station among |
| networks have been envisioned with all-IP-based | | | | many available base stations . |
| loose-coupled architecture. Vertical handoff | | | | For the traditional handoff decision algorithms |
| decision manner has transformed from | | | | have poor performance, for instance, too many |
| network-controled to terminal-controled. | | | | ping-pong times or outage times. In |
| In this article, we have studied the best network | | | | heterogeneous networks, the overcome of |
| selection and handoff decision problems.Vertical | | | | vertical handoff is quite greater then horizontal |
| handoff problem is more complex then horizontal | | | | handoff, the performance problem seems more |
| handoff. Vertical handoff whole process can be | | | | crucial. This article proposes a coverage aware |
| divided into three main steps: best network | | | | handoff decision algorithm, which uses the precise |
| selection, handoff decision and handoff signalling | | | | location, moving speed and direction information |
| protocol interworking. The factors that contribute | | | | combined with downloaded coverage information |
| to best network selection are relatively long time | | | | opened by network operators. In order to |
| network characteristics, including money cost, | | | | implement the coverage aware strategy, we |
| bandwidth, end to end delay and security etc. | | | | analyse the propagation model and the revising of |
| These factors have low correlation with mobile | | | | propagation model parameters, and bring forward |
| terminal's velocity, position and environment's | | | | the algorithm to calculate the cell border. |
| short time variable. While the main considerations | | | | Simulation results prove the high performance |
| of handoff decision are the ping-pong handoff and | | | | compared with traditional RSS-based decision |
| possible outages caused by drastic change of | | | | algorithms. By measuring signal coverage |
| received signal strength (RSS) for the | | | | information in west area of Huazhong Univ. of Sci. |
| movements and fading.Network condition, terminal | | | | and Tech., the practicality of proposed algorithm |
| and user perferences must be considered in best | | | | has been proved. When the mobile terminal can |
| network selection decision. Any change in above | | | | not receive GPS (Globe Positioning System) signal, |
| three aspects may trigger one time network | | | | we put forward a distance and velocity assisted |
| reselection process. | | | | handoff decision algorithm. By transforming the |
| This article analyses the features and drawbacks | | | | signal strength into distance, and continuous |
| of published best network selection schemes and | | | | sampling the RSS, we gettwo estimated |
| proposes a cost sensitive best network selection | | | | parameters that are initial distance (d|^l) and |
| scheme. The RSS changes have not been added | | | | leaving velocity (v|^) from base station. The |
| into this selection scheme, but are considered in | | | | proposed algorithm uses these two estimated |
| handoff decision algorithm. According to this | | | | algorithms to assist the handoff decision with the |
| special money cost factor, we design a user | | | | help of previous downloaded base station |
| accepted maximal cost level parameter, this | | | | coverage information. Simulation results |
| parameter is inputted by user to accurately fit | | | | demonstrate the better perfonnance then |
| user's demand in cost. A fuzzy segmental | | | | traditional RSS-base handoff decision algorithms. |