網(wǎng)絡(luò)體系結(jié)構(gòu)無線傳感器網(wǎng)絡(luò)技術(shù)

1、第二章 網(wǎng)絡(luò)體系架構(gòu),WSN節(jié)點(diǎn)概述和分類WSN節(jié)點(diǎn)結(jié)構(gòu)WSN節(jié)點(diǎn)體系結(jié)構(gòu)WSN網(wǎng)絡(luò)結(jié)構(gòu)無線傳感器網(wǎng)絡(luò)分類和特點(diǎn)WSN操作系統(tǒng),,傳感器概述：功用：一感二傳,即感受被測信息,并傳送出去。,傳感器定義我國國標(biāo)（GB7665-2005）對傳感器的定義是：“能感受被測量并按照一定的規(guī)律轉(zhuǎn)換成可用輸出信號的器件或裝置”。 傳感器的作用主要是感受和響應(yīng)規(guī)定的被測量，并按一定規(guī)律將其轉(zhuǎn)換成有用輸出，特別是完成非電量到電量的轉(zhuǎn)換。,

2、2/28,,,,,傳感器的組成,傳感器的組成，并無嚴(yán)格的規(guī)定。一般說來，可以把傳感器看作由敏感元件（有時(shí)又稱為預(yù)變換器）和變換元件（有時(shí)又稱為變換器）兩部分組成。,qzm_mail@nuaa.edu.cn,3/28,傳感器的分類,根據(jù)被測量與輸出電量的轉(zhuǎn)換原理分類能量轉(zhuǎn)換型、能量控制型根據(jù)傳感器感知的對象分類溫度傳感器、力傳感器、等根據(jù)輸出信號的性質(zhì)不同分類二值型、數(shù)字型、模擬型根據(jù)傳感器測量原理分類物理和化學(xué)….,4

3、,,常用物理傳感器 與化學(xué)傳感器,5,,分類-物理傳感器,1. 物理傳感器物理傳感器的原理是利用力、熱、聲、光、電、磁、射線等物理效應(yīng)，將被測信號量的微小變化轉(zhuǎn)換成電信號物理傳感器可以進(jìn)一步分為：力傳感器、熱傳感器、聲傳感器、光傳感器、電傳感器、磁傳感器與射線傳感器等7類,6,（1）力傳感器,力傳感器是能感受外力并將其轉(zhuǎn)換成可用輸出信號的傳感器。力傳感器的種類繁多，常用的力與壓力傳感器有電阻應(yīng)變式、半導(dǎo)體應(yīng)變式、壓阻式、電

4、感式、電容式、諧振式壓力傳感器，以及光纖壓力傳感器等用金屬應(yīng)變絲作為敏感元件的壓力傳感器原理示意圖,7,,不同用途的 力傳感器,8,,（2）溫度傳感器（3）聲傳感器,9,,（4）光傳感器圖像傳感器光纖傳感器,10,,分布式光纖傳感系統(tǒng)分布式光纖傳感系統(tǒng)利用光纖作為傳感敏感元件和傳輸信號介質(zhì)，探測出沿著光纖不同位置的溫度和應(yīng)變的變化，實(shí)現(xiàn)分布、自動(dòng)、實(shí)時(shí)、連續(xù)、精確的測量。分布式光纖傳感系統(tǒng)應(yīng)用領(lǐng)

5、域包括： —智能電網(wǎng)的電力電纜表面溫度檢測、事故點(diǎn)定位 —發(fā)電廠和變電站的溫度監(jiān)測、故障點(diǎn)檢測和報(bào)警 —水庫大壩、河堤安全與滲漏監(jiān)測 —橋梁與高層建筑結(jié)構(gòu)安全性監(jiān)測 —公路、地鐵、隧道地質(zhì)狀況的監(jiān)測分布光纖溫度傳感系統(tǒng)可以在易燃、易爆的環(huán)境下同時(shí)測量上萬個(gè)點(diǎn)，可以對每個(gè)溫度測量點(diǎn)進(jìn)行實(shí)時(shí)測量與定位,11,,（5）電傳感器電傳感器可以分為：電阻式、電容式、電感式傳感器電阻式傳感器利用

6、變阻器將非電量轉(zhuǎn)換成電阻信號的原理制成的。電阻式傳感器主要用于位移、壓力、應(yīng)變、力矩、氣流流速、液面與液體流量等參數(shù)的測量電容式是利用改變電容器的幾何尺寸或介質(zhì)參數(shù)，來使電容量變化的原理制成的。電容式傳感器主要用于壓力、位移、液面、厚度、水分含量等參數(shù)的測量電感式是利用改變電感磁路的幾何尺寸或磁體位置，來使電感或互感量變化的原理制成的，主要用于壓力、位移、力、振動(dòng)、加速度等參數(shù)的測量。,12,（6）磁傳感器,磁傳感器是最古老的傳感器

7、，指南針是磁傳感器的最早的一種應(yīng)用磁傳感器將磁信號轉(zhuǎn)化成為電信號輸出磁電式傳感器目前已經(jīng)被高性能磁敏感材料的新型磁傳感器所替代,13,（7）射線傳感器,射線傳感器是將射線強(qiáng)度轉(zhuǎn)換出可輸出的電信號的傳感器射線傳感器可以分為：X射線傳感器、γ射線傳感器、β射線傳感器、輻射劑量傳感器射線傳感器已經(jīng)在環(huán)境保護(hù)、醫(yī)療衛(wèi)生、科學(xué)研究與安全保護(hù)領(lǐng)域廣泛使用,14,分類-化學(xué)傳感器,化學(xué)傳感器可以將化學(xué)吸附、電化學(xué)反應(yīng)過程中被測信號的微小變化轉(zhuǎn)

8、換成電信號的一類傳感器 按傳感方式的不同，可分為： —接觸式化學(xué)傳感器 —非接觸式化學(xué)傳感器按結(jié)構(gòu)形式的不同，可以分為： —分離型化學(xué)傳感器 —組裝一體化化學(xué)傳感器按檢測對象的不同，可以分為： —?dú)怏w傳感器 —離子傳感器 —濕度傳感器,15,分類-生物傳感器,生物傳感器是由生物敏感元件和信號傳導(dǎo)器組成生物敏感元件可以是生物體、組織、細(xì)胞、酶、核酸或有

9、機(jī)物分子不同的生物元件對于光強(qiáng)度、熱量、聲強(qiáng)度、壓力有不同的感應(yīng)特性,16,生物傳感器的分類,,常用傳感器,qzm_mail@nuaa.edu.cn,17/28,,,,,智能傳感器,qzm_mail@nuaa.edu.cn,18/28,,,定義：傳統(tǒng)傳感器加上處理器具有智能功能的傳感器。所謂智能式傳感器，就是一種帶有微處理機(jī)的，兼有信息檢測、信號處理、信息記憶、邏輯分析與判斷功能的傳感器。 組成：傳感器+微處理器+信號處理電路。微處

10、理器為核心。,,智能傳感器的特點(diǎn),具有自動(dòng)調(diào)零和自動(dòng)校準(zhǔn)功能具有判斷和信息處理功能、對測量值進(jìn)行修正和誤差補(bǔ)償實(shí)現(xiàn)多慘了綜合測量（復(fù)合感知能力）自動(dòng)診斷故障靈活的通信能力智能傳感器的發(fā)展為傳感器技術(shù)的研究提出了很多富有挑戰(zhàn)性的課題,qzm_mail@nuaa.edu.cn,19/28,微機(jī)電系統(tǒng)對智能傳感器發(fā)展的影響,微機(jī)電系統(tǒng)（MEMS）是指集微型機(jī)構(gòu)、微型傳感器、微型執(zhí)行器以及信號處理和控制電路，直至接口、通信和電源等

11、于一體的微型器件或系統(tǒng),qzm_mail@nuaa.edu.cn,20/28,性能指標(biāo),靜態(tài)指標(biāo)線性范圍：輸出量與輸入量呈線性關(guān)系的測量范圍線性度：傳感器實(shí)際輸入輸出量曲線與擬合曲線的重疊程度靈敏度：輸出變化量與輸入變化量的比值穩(wěn)定性：表示傳感器經(jīng)過長期使用后，輸出特性不發(fā)生變化的性能分辨率：能檢測到的被測量的最小變化量,21/28,（具體定義公式見教材）,性能指標(biāo),遲滯：傳感器在輸入量從小到大正向變化與反向變化，其輸入輸出特

12、性曲線不重合的程度重復(fù)性：傳感器在輸入量按同一方向做全量程連續(xù)多次變化時(shí)所得特性曲線一致的程度漂移：輸入量不變的情況下，輸出量隨時(shí)間變化的現(xiàn)象測量范圍：最小/最大輸入量的范圍精度：測量結(jié)果的可靠程度，以給定的準(zhǔn)確度來表示重復(fù)某個(gè)讀數(shù)的能力動(dòng)態(tài)指標(biāo)輸出信號對應(yīng)隨時(shí)間變化的輸入量的響應(yīng)特性，反映測量精度、重復(fù)性和可靠性,第二章 網(wǎng)絡(luò)體系架構(gòu),WSN節(jié)點(diǎn)概述和分類WSN節(jié)點(diǎn)結(jié)構(gòu)WSN節(jié)點(diǎn)體系結(jié)構(gòu)WSN網(wǎng)絡(luò)結(jié)構(gòu)無線傳感器網(wǎng)絡(luò)

13、分類和特點(diǎn)WSN操作系統(tǒng),,24,無線傳感器節(jié)點(diǎn)結(jié)構(gòu)（1）,無線傳感器節(jié)點(diǎn)通常由4個(gè)功能模塊組成：感知模塊、處理模塊、通信模塊和電源模塊,,,25,無線傳感器節(jié)點(diǎn)結(jié)構(gòu)（2）,感知模塊：由一個(gè)或多個(gè)傳感器以及模數(shù)轉(zhuǎn)換器組成。傳感器負(fù)責(zé)感知監(jiān)測目標(biāo)的物理特征和現(xiàn)象，并產(chǎn)生相應(yīng)的模擬信號。感知聲、光、電、震動(dòng)、磁、氣體等信號模數(shù)轉(zhuǎn)換器負(fù)責(zé)將模擬信號轉(zhuǎn)換為數(shù)字信號，并將數(shù)字信號送往處理模塊進(jìn)行處理。處理模塊：由一個(gè)微處理器及相應(yīng)的內(nèi)存組

14、成，負(fù)責(zé)對數(shù)據(jù)進(jìn)行處理、濾波、決策、執(zhí)行、資源分配等判斷，并對傳感器節(jié)點(diǎn)進(jìn)行控制通信模塊：由數(shù)模轉(zhuǎn)換器和無線收發(fā)送器（Zigbee, WiFi, GPRS,短波等）組成，負(fù)責(zé)發(fā)送和接收數(shù)據(jù)和控制信息,,,26,無線傳感器節(jié)點(diǎn)結(jié)構(gòu)（3）,電源模塊：負(fù)責(zé)節(jié)點(diǎn)的供電（電池供電或從自然界采集能量）其他模塊位置模塊：節(jié)點(diǎn)可以配置全球定位系統(tǒng)（Global Positioning System, GPS）或北斗定位系統(tǒng)接收器移動(dòng)模塊：在某些

15、需要移動(dòng)傳感器節(jié)點(diǎn)的應(yīng)用中，可以配置馬達(dá)來驅(qū)動(dòng)傳感器節(jié)點(diǎn),,,https://en.wikipedia.org/wiki/List_of_wireless_sensor_nodes,一個(gè)簡單的應(yīng)用開發(fā)需要：,,一個(gè)簡單的應(yīng)用開發(fā)需要：,,Programming Board (MIB510),29,Mica 2 Motes,MICA 2 MOTE,https://www.eol.ucar.edu/isf/facilities/isa/in

16、ternal/CrossBow/DataSheets/mica2.pdf,30,,Sensor Board - MTS310,32,第二章 網(wǎng)絡(luò)體系架構(gòu),WSN節(jié)點(diǎn)概述和分類WSN節(jié)點(diǎn)結(jié)構(gòu)WSN節(jié)點(diǎn)體系結(jié)構(gòu)WSN網(wǎng)絡(luò)結(jié)構(gòu)無線傳感器網(wǎng)絡(luò)分類和特點(diǎn)WSN操作系統(tǒng),,,跨層設(shè)計(jì) 聯(lián)合優(yōu)化,無線傳感器網(wǎng)絡(luò)協(xié)議棧,WSN的體系架構(gòu),35,WSN的通信協(xié)議,物理層(主要是電磁波)負(fù)責(zé)頻率選擇、頻率產(chǎn)生、信號檢測、調(diào)制解調(diào)和數(shù)據(jù)加密,,

17、36,WSN的通信協(xié)議（續(xù)）,,鏈路層負(fù)責(zé)介質(zhì)訪問和差錯(cuò)控制確保通信中點(diǎn)-點(diǎn) /點(diǎn)-多點(diǎn)的連接介質(zhì)訪問控制：創(chuàng)建網(wǎng)絡(luò)架構(gòu)在節(jié)點(diǎn)間公平有效的共享通信資源差錯(cuò)控制前向糾錯(cuò)FEC自動(dòng)重發(fā)檢錯(cuò)ARQ,,37,WSN的通信協(xié)議（續(xù),,網(wǎng)絡(luò)層（路由層）節(jié)能總是最重要的考慮傳感網(wǎng)絡(luò)多以數(shù)據(jù)為中心數(shù)據(jù)聚合不能影響傳感節(jié)點(diǎn)的協(xié)同理想的傳感器網(wǎng)絡(luò)基于屬性尋址并能感知位置,,38,WSN的通信協(xié)議（續(xù),,傳輸層負(fù)責(zé)數(shù)據(jù)流的傳輸控制

18、可靠性和擁塞控制當(dāng)傳感系統(tǒng)需要被Internet或者其它外部網(wǎng)絡(luò)訪問時(shí)傳輸層設(shè)計(jì)與實(shí)現(xiàn)至關(guān)重要研究不成熟,,39,WSN的通信協(xié)議（續(xù)）,,跨層設(shè)計(jì) 聯(lián)合優(yōu)化,應(yīng)用層負(fù)責(zé)任務(wù)調(diào)度和數(shù)據(jù)分發(fā)等具體業(yè)務(wù)，使得底層硬件和軟件對傳感應(yīng)用是透明的 Sensor management protocol (SMP) Task assignment and data advertisement protocol (TADAP) Sensor

19、 query and data dissemination protocol (SQDDP),,網(wǎng)絡(luò)管理平臺(tái),主要是對傳感器節(jié)點(diǎn)自身的管理和用戶對wsn的管理，包括：拓?fù)淇刂品?wù)質(zhì)量管理能力管理安全管理移動(dòng)管理網(wǎng)絡(luò)管理,,應(yīng)用支撐平臺(tái),包括一系列基于監(jiān)測任務(wù)的應(yīng)用層軟件，通過應(yīng)用服務(wù)借口和網(wǎng)絡(luò)管理接口為用戶提供各種應(yīng)用的支持。時(shí)間同步定位應(yīng)用服務(wù)接口網(wǎng)絡(luò)管理接口,,第二章 網(wǎng)絡(luò)體系架構(gòu),WSN節(jié)點(diǎn)概述和分類WSN節(jié)

20、點(diǎn)結(jié)構(gòu)WSN節(jié)點(diǎn)體系結(jié)構(gòu)WSN網(wǎng)絡(luò)結(jié)構(gòu)無線傳感器網(wǎng)絡(luò)分類和特點(diǎn)WSN操作系統(tǒng),,43,無線傳感網(wǎng)的布設(shè)形式,隨機(jī)布設(shè)網(wǎng),戰(zhàn)場：炮射、空投應(yīng)急：火災(zāi)、毒氣,固定布設(shè)網(wǎng),工控：自動(dòng)控制監(jiān)測：遠(yuǎn)程監(jiān)測,44,無線傳感器網(wǎng)絡(luò)的網(wǎng)絡(luò)結(jié)構(gòu)(1),傳感器網(wǎng)絡(luò)基本組成：傳感器節(jié)點(diǎn)(sensor node) 匯聚節(jié)點(diǎn)(sink) 管理節(jié)點(diǎn)（Management node）匯聚節(jié)點(diǎn)→傳感器節(jié)點(diǎn): 查詢消息或指令匯聚節(jié)點(diǎn)還作為連接外部傳輸網(wǎng)絡(luò)

21、（如互聯(lián)網(wǎng)、衛(wèi)星網(wǎng)等）的網(wǎng)關(guān)，對收集到的數(shù)據(jù)進(jìn)行簡單的處理，然后將數(shù)據(jù)通過互聯(lián)網(wǎng)或其它傳輸網(wǎng)絡(luò)，傳送給監(jiān)控中心和需要使用這些數(shù)據(jù)的終端用戶傳感器節(jié)點(diǎn)→匯聚節(jié)點(diǎn): 監(jiān)測數(shù)據(jù),,,sink,Internet/Satellite,,,,Sensor node,45,無線傳感器網(wǎng)絡(luò)的網(wǎng)絡(luò)結(jié)構(gòu)(2),單跳結(jié)構(gòu)各傳感器節(jié)點(diǎn)可以采用單跳的方式將各自的數(shù)據(jù)直接發(fā)送給匯聚節(jié)點(diǎn)基于蜂窩的窄帶物聯(lián)網(wǎng)（Narrow Band Internet of T

22、hings, NB-IoT）,,,,Sensor node,,Ref: NB-IoT: A sustainable technology for connecting billons of devices Ericssion Technology Review April 22, 2016,,NB-IoT的應(yīng)用,無線傳感器網(wǎng)絡(luò)的網(wǎng)絡(luò)結(jié)構(gòu)(3),多跳結(jié)構(gòu)傳感器節(jié)點(diǎn)通過一個(gè)或多個(gè)網(wǎng)絡(luò)中間節(jié)點(diǎn)將所采集到的數(shù)據(jù)傳送給匯聚節(jié)點(diǎn)，從而有效地降低

23、通信所需的能耗,,,,,,,平面結(jié)構(gòu),多跳分簇結(jié)構(gòu),,多層分簇結(jié)構(gòu),第二章 網(wǎng)絡(luò)體系架構(gòu),WSN節(jié)點(diǎn)概述和分類WSN節(jié)點(diǎn)結(jié)構(gòu)WSN節(jié)點(diǎn)體系結(jié)構(gòu)WSN網(wǎng)絡(luò)結(jié)構(gòu)無線傳感器網(wǎng)絡(luò)分類和特點(diǎn)WSN操作系統(tǒng),,50,無線傳感器網(wǎng)絡(luò)的分類（1）,單跳網(wǎng)絡(luò)與多跳網(wǎng)絡(luò) single-hop/multi-hop靜止網(wǎng)絡(luò)與移動(dòng)網(wǎng)絡(luò) static/mobile靜止傳感網(wǎng)中，所有傳感器節(jié)點(diǎn)都是靜止不動(dòng)的。靜止傳感網(wǎng)實(shí)現(xiàn)簡單，但靜止WSN存在瓶頸；

24、移動(dòng)傳感網(wǎng)中，某些節(jié)點(diǎn)可以移動(dòng)。移動(dòng)可以平衡網(wǎng)絡(luò)流量和能耗分布，但動(dòng)態(tài)性使得協(xié)議實(shí)現(xiàn)復(fù)雜。靜止匯聚節(jié)點(diǎn)網(wǎng)絡(luò)與移動(dòng)匯聚節(jié)點(diǎn)網(wǎng)絡(luò) 靜止匯聚節(jié)點(diǎn)網(wǎng)絡(luò)中，匯聚節(jié)點(diǎn)的位置靜止固定在監(jiān)測區(qū)域附近或內(nèi)部；移動(dòng)匯聚節(jié)點(diǎn)網(wǎng)絡(luò)中，匯聚節(jié)點(diǎn)在監(jiān)測區(qū)域移動(dòng)，收集傳感器節(jié)點(diǎn)的監(jiān)測數(shù)據(jù)，可以平衡各網(wǎng)絡(luò)負(fù)載，減小熱點(diǎn)效應(yīng)。,,,,51,無線傳感器網(wǎng)絡(luò)的分類（2）,確定性網(wǎng)絡(luò)與非確定性網(wǎng)絡(luò)根據(jù)傳感器節(jié)點(diǎn)部署的情況不同確定性傳感器網(wǎng)絡(luò)中，傳感器節(jié)點(diǎn)的位置是預(yù)先

25、規(guī)劃的，且一旦部署不再變化；非確定性傳感器網(wǎng)絡(luò)中，節(jié)點(diǎn)隨機(jī)布設(shè)，甚至可以移動(dòng)，適合自然條件惡劣或敵對的應(yīng)用場合，難以預(yù)先規(guī)劃。單匯聚節(jié)點(diǎn)網(wǎng)絡(luò)與多匯聚節(jié)點(diǎn)網(wǎng)絡(luò)根據(jù)匯聚節(jié)點(diǎn)的數(shù)量單匯聚節(jié)點(diǎn)網(wǎng)絡(luò)中，只有一個(gè)匯聚節(jié)點(diǎn)位于監(jiān)測區(qū)域的附近或內(nèi)部；多匯聚節(jié)點(diǎn)網(wǎng)絡(luò)中，多個(gè)匯聚節(jié)點(diǎn)布設(shè)于不同位置，各傳感器節(jié)點(diǎn)可以將其監(jiān)測數(shù)據(jù)發(fā)送給離自己最近的匯聚節(jié)點(diǎn)。,,,,52,無線傳感器網(wǎng)絡(luò)的分類（3）,同構(gòu)網(wǎng)絡(luò)與異構(gòu)網(wǎng)絡(luò)根據(jù)傳感器節(jié)點(diǎn)是否具有相同的能力

26、（通信能力、感知能力、計(jì)算能力等的不同）同構(gòu)傳感器網(wǎng)絡(luò)中，所有傳感器節(jié)點(diǎn)能力完全相同；異構(gòu)傳感器網(wǎng)絡(luò)中，傳感器節(jié)點(diǎn)（或部分）能力不同，可以配備較強(qiáng)的通信、處理模塊或傳感器。先應(yīng)式網(wǎng)絡(luò)和反應(yīng)式網(wǎng)絡(luò)先應(yīng)式網(wǎng)絡(luò)采用連續(xù)操作模式，節(jié)點(diǎn)定期打開傳感器和發(fā)送器、感知環(huán)境并進(jìn)行數(shù)據(jù)匯包 適用于定期監(jiān)控的應(yīng)用反應(yīng)式網(wǎng)絡(luò)采用查詢-響應(yīng)的操作模式，節(jié)點(diǎn)按需的進(jìn)行感知和數(shù)據(jù)匯報(bào),,,,WSN的特征,四大受限：能量受限 通信能力受限計(jì)算能力受限

27、存儲(chǔ)能力受限組網(wǎng)特征：自組織性以數(shù)據(jù)為中心應(yīng)用相關(guān)性事件驅(qū)動(dòng)拓?fù)鋭?dòng)態(tài)性大規(guī)模需要考慮可靠性和魯棒性,54,無線傳感器網(wǎng)絡(luò)的設(shè)計(jì)目標(biāo),節(jié)點(diǎn)體積小型化節(jié)點(diǎn)成本低廉節(jié)點(diǎn)功耗超低節(jié)點(diǎn)可自組成網(wǎng)可擴(kuò)展性強(qiáng)自適應(yīng)性強(qiáng)可靠性高滿足一定的安全性和服務(wù)質(zhì)量要求,55,傳感網(wǎng)絡(luò)面臨的問題和挑戰(zhàn),,56,影響傳感器網(wǎng)絡(luò)的因素(1),57,影響傳感器網(wǎng)絡(luò)的因素(2),,58,影響傳感器網(wǎng)絡(luò)的因素(3),制造成本單個(gè)節(jié)點(diǎn)的成本是

28、衡量整個(gè)網(wǎng)絡(luò)成本的關(guān)鍵如何以低于一美元的成本制造一定功能的傳感節(jié)點(diǎn)是很大的挑戰(zhàn)傳輸介質(zhì)（選擇的傳輸介質(zhì)必須是全球可用的）無線電紅外其他光波,59,影響傳感器網(wǎng)絡(luò)的因素(4),傳感器的環(huán)境敵方戰(zhàn)場居家/大型建筑物大型倉庫動(dòng)物身上快速移動(dòng)的車輛有水流動(dòng)的排水管/河流,第二章 網(wǎng)絡(luò)體系架構(gòu),WSN節(jié)點(diǎn)概述和分類WSN節(jié)點(diǎn)結(jié)構(gòu)WSN節(jié)點(diǎn)體系結(jié)構(gòu)WSN網(wǎng)絡(luò)結(jié)構(gòu)WSN分類WSN操作系統(tǒng),,Major Concern

29、s in WSN OS Design,能在有限的資源上運(yùn)行:WSN節(jié)點(diǎn)資源非常有限(通信帶寬資源/能量資源 /計(jì)算資源),操作系統(tǒng)必須能夠高效地使用各種資源; 允許高度的并發(fā)性:要求執(zhí)行模式能對事件作出快速的直接響應(yīng); 適應(yīng)硬件升級 支持一系列平臺(tái):要求組件和執(zhí)行模式能夠應(yīng)對硬件/ 軟件的替換，具有可移植性; 支持多樣化的應(yīng)用程序:良好的模塊化設(shè)計(jì),使應(yīng)用/協(xié)議/服務(wù)與硬件資源之間可以隨意搭配，要求能夠根據(jù)實(shí)際需要, 裁減操作系統(tǒng)

30、的服務(wù); 魯棒性強(qiáng):要求通過組件間有限的交互渠道,就能應(yīng)對各種復(fù)雜情況; 提供高效的組網(wǎng)機(jī)制：WSN是一個(gè)網(wǎng)絡(luò)系統(tǒng),其操作系統(tǒng)也必然是面向 網(wǎng)絡(luò)化開發(fā)的。網(wǎng)絡(luò)化系統(tǒng)要求操作系統(tǒng)必須為 應(yīng)用提供高效的組網(wǎng)和通信機(jī)制,WSNOS Outline,Functional AspectsData TypesSchedulingStacksSystem CallsHandling InterruptsMultithreadingT

31、hread-based vs. Event-based ProgrammingMemory AllocationNon-Functional AspectsSeparation of ConcernSystem OverheadPortabilityDynamic ReprogrammingPrototypesTinyOSContikiEvaluation,Operating Systems,An operatin

32、g System isa thin software layerresides between the hardware and the application layerprovides basic programming abstractions to application developersIts main task is to enable applications to interact with hardware

33、 resources,Operating Systems,Operating systems are classified as: single-task/multitasking and single-user/multiuser operating systemsmulti-tasking OS - the overhead of concurrent processing because of the limited resou

34、rcessingle task OS - tasks should have a short durationThe choice of a particular OS depends on several factors; typically functional and non-functional aspects,Outline,Functional AspectsData TypesSchedulingStacksS

35、ystem CallsHandling InterruptsMultithreadingThread-based vs. Event-based ProgrammingMemory AllocationNon-Functional AspectsSeparation of ConcernSystem OverheadPortabilityDynamic ReprogrammingPrototypesTinyOSC

36、ontikiEvaluation,Data Types,Interactions between the different subsystems take place through:well-formulated protocolsdata typesComplex data types have strong expression power but consume resources - struct and enum

37、Simple data types are resource efficient but have limited expression capability - C programming language,Scheduling,Two scheduling mechanisms: queuing-based scheduling FIFO - the simplest and has minimum system overhea

38、d, but treats tasks unfairlysorted queue - e.g., shortest job first (SJF) - incurs system overhead (to estimate execution duration)round-robin schedulinga time sharing scheduling techniqueseveral tasks can be process

39、ed concurrently,Scheduling,Regardless of how tasks are executed, a scheduler can be eithera non-preemptive scheduler - a task is executed to the end, may not be interrupted by another taskor preemptive scheduler - a t

40、ask of higher priority may interrupt a task of low priority,Stacks & System Calls,Stacksa data structure that temporarily stores data objects in memory by piling one upon anotherobjects are accessed using last-in-f

41、irst-out (LIFO)System Calls：is the programmatic way in which a computer program requests a service from the kernel of the operating system it is executed ondecouple the concern of accessing hardware resources from imp

42、lementation detailswhenever users wish to access a hardware resource, they invoke these operations without the need to concern themselves how the hardware is accessed,Handling Interrupts,An interrupt is an asynchronous

43、signal generated by a hardware deviceseveral system eventsOS itselfAn interrupt causes:the processor to interrupt executing the present instructionto call for an appropriate interrupt handlerInterrupt signals can

44、have different priority levels, a high priority interrupt can interrupt a low level interruptInterrupt mask: let programs choose whether or not they wish to be interrupted,Multi-threading,A thread is the path taken by a

45、 processor or a program during its executionMulti-threading - a task is divided into several logical piecesscheduled independent from each otherexecuted concurrentlyTwo advantages of a multi-threaded OS:tasks do not

46、 block other tasksshort-duration tasks can be executed along with long-duration tasks,Multi-threading,Threads cannot be created endlessly the creation of threads slows down the processorno sufficient resources to divi

47、deThe OS can keep the number of threads to a manageable size using a thread pool,Thread-based vs. Event-based Programming,Decision whether to use threads or events programming:need for separate stacksneed to estimate

48、maximum size for saving context informationThread-based programs use multiple threads of control within:a single program a single address space,Thread-based vs. Event-based Programming,Advantage:a thread blocked can

49、be suspended while other tasks are executed in different threadsDisadvantages:must carefully protect shared data structures with locksuse condition variables to coordinate the execution of threads,Thread-based vs. Eve

50、nt-based Programming,In event-based programming: use events and event handlersevent-handlers register with the OS scheduler to be notified when a named event occursa loop function: polls for eventscalls the appropri

51、ate event-handlers when events occurAn event is processed to completionunless its handler reaches at a blocking operation (callback and returns control to the scheduler),Memory Allocation,The memory unit is a precious

52、 resourceReading and writing to memory is costly How and for how long a memory is allocated for a piece of program determines the speed of task execution,Memory Allocation,Memory can be allocated to a program:statical

53、ly - a frugal approach, but the requirement of memory must be known in advancememory is used efficientlyruntime adaptation is not alloweddynamically - the requirement of memory is not known in advance (on a transient

54、basis)enables flexibility in programmingbut produces a considerable management overhead,Outline,Functional AspectsData TypesSchedulingStacksSystem CallsHandling InterruptsMultithreadingThread-based vs. Event-bas

55、ed ProgrammingMemory AllocationNon-Functional AspectsSeparation of ConcernSystem OverheadPortabilityDynamic ReprogrammingPrototypesTinyOSContikiEvaluation,Separation of Concern,SoC is a design principle for se

56、parating a computer program into distinct sections, such that each section addresses a separate concernIn general, separation between the operating system and the applications layerThe operation systems can provide:a

57、number of lightweight modules - “wired” together, oran indivisible system kernel + a set of library components for building an application, ora kernel + a set of reconfigurable low-level servicesSeparation of concern

58、enables:flexible and efficient reprogramming and reconfiguration,Portability,Ideally, operating systems should be able to co-exist and collaborate with each otherHowever, existing operating systems do not provide this

59、type of supportIn order to accommodate unforeseen requirements, operating systems should be portable and extensible,System Overhead,An operating system executes program code - requires its own share of resourcesThe res

60、ources consumed by the OS are the system’s overhead, it depends onthe size of the operating system the type of services that the OS provides to the higher-level services and applications,System Overhead,The resources o

61、f wireless sensor nodes have to be shared by programs that carry out:sensingdata aggregationself-organizationnetwork managementnetwork communication,Dynamic Reprogramming,Once a wireless sensor network is deployed,

62、 it may be necessary to reprogram some part of the application or the operating system for the following reasons:the network may not perform optimallyboth the application requirements and the network’s operating enviro

63、nment can change over timemay be necessary to detect and fix bugs,Dynamic Reprogramming,Manual replacement may not be feasible - develop an operating system to provide dynamic reprogramming support, which depends oncle

64、ar separation between the application and the OSthe OS can receive software updates and assemble and store it in memoryOS should make sure that this is indeed an updated versionOS can remove the piece of software that

65、 should be updated and install and configure the new versionall these consume resources and may cause their own bugs,Dynamic Reprogramming,Software reprogramming (update) requires robust code dissemination protocols:sp

66、litting and compressing the codeensuring code consistency and version controllingproviding a robust dissemination strategy to deliver the code over a wireless link,Outline,Functional AspectsData TypesSchedulingStack

67、sSystem CallsHandling InterruptsMultithreadingThread-based vs. Event-based ProgrammingMemory AllocationNon-Functional AspectsSeparation of ConcernSystem OverheadPortabilityDynamic ReprogrammingPrototypesTinyO

68、SContikiEvaluation,TinyOS (Gay et al. 2007),TinyOS is the most widely used, richly documented, and tool-assisted runtime environment in WSN http://www.tinyos.netstatic memory allocation event-based systemTinyO

69、S’s architecture consists ofa scheduler a set of components, which are classified intoconfiguration components - "wiring" (how models are connected with each other) modules - the basic building blocks of a