應用於無線的異質傳感器網路的一個新的二元訓練協定

Abstract

最近，Barsi等人針對無線的異質傳感器網路提出了一個節省能源的位置訓練協定，傳感器可經由此協定被訓練並認知出它本身相對於附近之actor的粗略位置。在此協定中，傳感器是利用actor所傳輸之訊息以及二元搜尋法，來認知出自己的粗略位置，因此Barsi等人稱其協定為BINARY TRAINING PROTOCOL。在BINARY TRAINING PROTOCOL的運作中，actor交替廣播control broadcast及data broadcast，因此，若傳感器的清醒週期長度d為偶數，則在此清醒週期中，d/2個control broadcasts及d/2個data broadcasts被播放。我們因此猜測：如果在一個清醒週期中，actor所廣播的d個broadcasts中僅有一個是control broadcast，其餘(d − 1)個均是data broadcasts，則BINARY TRAINING PROTOCOL的效益能否提升？本論文之目的即在於驗證此一猜測。

Recently, Barsi et al. proposed an energy-e?cient location-training protocol for heterogeneous actor-centric sensor networks where the sensors acquire coarse-grained location awareness with respect to the actor in their vicinity. This training protocol is called BINARY TRAINING PROTOCOL according to its nature that the sensors perform a binary Recently, Barsi et al. proposed an energy-efficient location-training protocol for heterogeneous actor-centric sensor networks where the sensors acquire coarse-grained location awareness with respect to the actor in their vicinity. This training protocol is called BINARY TRAINING PROTOCOL according to its nature that the sensors perform a binary search among the actor transmissions to locate their correct corona. In BINARY TRAINING PROTOCOL, the actor alternates control broadcasts (at full power level) with data broadcasts (at decreasing power levels), transmitting corona identities in decreasing order. Therefore in this training protocol, if the awake period of a sensor is an even number d, then during this awake period, d/2 control broadcasts and d/2 data broadcasts are broadcasted by the actor. We therefore conjecture: if during an awake period, the actor broadcasts only one control broadcast and all of the other (d-1) broadcasts are data broadcasts, will the performance of BINARY TRAINING PROTOCOL be improved? The purpose of this thesis is to confirm this conjecture.