Context-aware Routing Schemes for WSNs
- 주제(키워드) context , WSN , sensor network , context-aware , routing , aggregation , cluster
- 발행기관 고려대학교 대학원
- 지도교수 안순신
- 발행년도 2009
- 학위수여년월 2009. 8
- 학위명 석사
- 학과 일반대학원 전자전기공학과
- 세부전공 전자전기컴퓨터공학전공
- 원문페이지 56 p
- 실제URI http://www.dcollection.net/handler/korea/000000009669
- 본문언어 영어
- 제출원본 000045551371
초록/요약
With the start of ubiquitous network era, researches on Wireless Sensor Networks (WSNs) have been widely and eagerly done. However, techniques considering data obtained from the environment as parameters that affect the behavior of WSNs are extremely rare. In this thesis, we define the sensed environment as context to propose two context-aware routing schemes: Context-aware Dynamic Clustering (CDC) and Context-aware Hybrid Routing (CHR). Both the schemes aim to enhance the performance of conventional WSNs with the new approach. One characteristic of the real environment is that it is sptio-temporary correlated. The characteristic can be used for efficient aggregation and routing in event based applications. For instance, in habitat monitoring, nodes may be requested to reply only when they are under user-defined context. (e.g., temperature >30) CDC enables nodes to spontaneously react to contextual changes, so that the nodes satisfying the condition form clusters by themselves. Due to spatio-temporal correlation mentioned earlier, the nodes tend to be located close to each other. Therefore large reduction in network traffic as well as in energy consumption can be achieved through clustering. When compared to Tiny AGgregation (TAG) [8] and ACE [13], traffic is reduced up to 27.8% and 28.1%, respectively. CHR proposes a different approach in handling contextual information. In most applications, sensing data are not equal in their importance. From the user’s point of view, unexpected data are more likely to contain important information. In other words, data collected during abrupt changes in the context carry more information than the others. The purpose of CHR is to provide more reliable routing method for important data, while the others use a conventional, single path routing scheme. To be specific, packets with high importance are diffused throughout the network, whereas the rest are delivered utilizing a popular routing algorithm, Ad hoc On-demand Distance Vector (AODV) [3]. Furthermore, AODV is revised to compensate with the increased traffic caused by diffusion. Revised version of AODV shows vastly reduced number of control packet transmissions, compared to the original protocol (a 76% decrease). When the probability of encountering critical data is assumed to be 0.01, CHR achieves 15.4% decrease in network traffic on average, while maintaining the delivery ratio of critical data at 100%.
more목차
CHAPTER 1. INTRODUCTION 1
1.1. OBJECTIVE OF THE THESIS 1
1.2. THESIS STRUCTURE 1
CHAPTER 2. CDC: CONTEXT-AWARE DYNAMIC CLUSTERING 3
2.1. INTRODUCTION TO CDC 3
2.2. QUERY MODEL 7
2.3. ALGORITHM OVERVIEW 8
2.4. PHASES OF CDC 10
2.4.1. Distribution Phase 10
2.4.2. Setup Phase 11
2.4.3. Reply Phase 15
2.5. PERFORMANCE EVALUATION 18
2.5.1. Simulation Results 22
CHAPTER 3. CHR: CONTEXT-AWARE HYBRID ROUTING 24
3.1. INTRODUCTION TO CHR 24
3.2. ALGORITHM OVERVIEW 25
3.3. HOW CHR WORKS 26
3.3.1. Query Model 26
3.3.2. Path Construction 27
3.3.3. Packet Forwarding 28
3.3.4. Path Restoration 35
3.3.5. Priority Decision 35
3.3.6. Design Objectives and Effects of CHR 41
3.4. PERFORMANCE EVALUATION 42
3.4.1. Simulation Results 43
CHAPTER 4. CONCLUSIONS AND FUTURE WORKS 47
REFERENCE 48
