《基于熱釋電紅外傳感器的智能家居室內(nèi)感應(yīng)定位系》文獻(xiàn)及翻譯

1、 畢 業(yè) 設(shè) 計(jì)外 文 文 獻(xiàn) 譯 文 及 原 文學(xué) 生： 學(xué) 號(hào)： 200906030218 院 （系）： 電氣與信息工程學(xué)院 專 業(yè)： 電子信息工程 指導(dǎo)教師： 2013 年 6 月 16日基于熱釋電紅外傳感器的智能家居室內(nèi)感應(yīng)定位系統(tǒng)Suk Lee，電機(jī)及電子學(xué)工程師聯(lián)合會(huì)會(huì)員Kyoung Nam Ha, Kyung Chang Lee,電機(jī)及電子學(xué)工程師聯(lián)合會(huì)會(huì)員摘要：智能家居，是一種可以通過識(shí)別具有不同生活習(xí)慣和感覺的住戶來(lái)提供各種不同的智能服務(wù)。而實(shí)現(xiàn)這樣的功能其中最關(guān)鍵的問題之一就是如何確定住戶的位置。目前，研究工作只要集中于兩種方法：終端方式和非終端方式。終端方式需要一種住戶隨

2、身攜帶的設(shè)備，而非終端方式則不需要這樣的設(shè)備。本文提出一種使用可以探測(cè)到住戶的熱釋電紅外傳感器（紅外傳感器）的新的非終端方式。該系統(tǒng)的可行性已經(jīng)通過了測(cè)試平臺(tái)的實(shí)驗(yàn)性評(píng)估。關(guān)鍵詞：智能家居，定位服務(wù)，熱釋電紅外傳感器（紅外傳感器），定位識(shí)別算法1 簡(jiǎn)介現(xiàn)在由于人人都想有一個(gè)方便，舒適，安全的居住環(huán)境，因此大家對(duì)于智能家居表現(xiàn)的越來(lái)越感興趣1 2。一般來(lái)說(shuō)，智能家居旨在提供合適的智能服務(wù)來(lái)積極促進(jìn)住戶更好的生活，比如家務(wù)勞動(dòng)，娛樂，休息和睡眠。因此，為了提高住戶的便捷和安全，像家用電器，多媒體設(shè)備和互聯(lián)網(wǎng)設(shè)備應(yīng)通過家庭網(wǎng)絡(luò)系統(tǒng)連接在一起，如圖1所示。并且它們應(yīng)通過電視或個(gè)人數(shù)字助理（PDA）來(lái)控

3、制或遠(yuǎn)程監(jiān)控3 4。圖1 智能家居的家庭網(wǎng)絡(luò)體系結(jié)構(gòu)尤其要注意的是，作為一種提供高質(zhì)量的智能服務(wù)，目標(biāo)應(yīng)集中于定位服務(wù)，同時(shí)考慮人為因素，比如住戶的生活方式，健康狀況和居住感受57。也就是說(shuō)，如果智能家居能識(shí)別住戶的生活方式或健康狀況，那么家用電器應(yīng)該能預(yù)見住戶的需要，并能更主動(dòng)的提供適合的智能服務(wù)。例如，在一個(gè)被動(dòng)的服務(wù)環(huán)境下，需要住戶控制供熱通風(fēng)與空氣調(diào)節(jié)系統(tǒng)（供暖，通風(fēng)和空調(diào)），而智能家居將根據(jù)住戶情況自動(dòng)調(diào)節(jié)房間的溫濕度。智能家居或智能辦公室的各種室內(nèi)感應(yīng)定位系統(tǒng)的已經(jīng)研發(fā)到能夠識(shí)別住戶的位置。一般來(lái)說(shuō)，室內(nèi)定位感應(yīng)系統(tǒng)根據(jù)測(cè)量技術(shù)分為三種類型：三角測(cè)量，場(chǎng)景分析和接近方法8。三角測(cè)量

4、法是通過多個(gè)已知點(diǎn)來(lái)計(jì)算位置距離。運(yùn)用三角測(cè)量法的例子包括Active Badges9，Active Bats10和Easy Living11，它們分別運(yùn)用了紅外傳感器，超聲波傳感器和視覺傳感器來(lái)實(shí)現(xiàn)的。場(chǎng)景解析法是檢測(cè)一個(gè)場(chǎng)景內(nèi)的特定著眼點(diǎn)。場(chǎng)景解析法的典型例子是使用直流磁力跟蹤器的MotiveStar12，和使用無(wú)線局域網(wǎng)絡(luò)LAN標(biāo)準(zhǔn)IEEE 802,11的RADAR13。接近法則是以一組已知點(diǎn)中最接近的點(diǎn)近似作為定位點(diǎn)。接近法的例子有使用壓力傳感器的Smart Floor14。另外，室內(nèi)感應(yīng)定位系統(tǒng)可以根據(jù)是否需要住戶隨身攜帶一種設(shè)備來(lái)分類。終端方式，例如Active Bats，不需要直

5、接找到住戶位置，但是可以感應(yīng)到住戶隨身攜帶的設(shè)備位置，例如紅外收發(fā)器或者射頻識(shí)別技術(shù)（RFID）標(biāo)簽。因此，如果住戶沒有隨聲攜帶終端設(shè)備，那就不可能找到他。相反的，非終端方式如Easy Living和Smart Floor則不需要這種設(shè)備就能找到住戶位置。然而，人們認(rèn)為Easy Living侵犯了住戶隱私，Smart Floor則是擴(kuò)展和維護(hù)都比較困難。本文提出一種使用陣列熱釋電紅外（PIR）傳感器實(shí)現(xiàn)的基于非終端方式的室內(nèi)感應(yīng)定位系統(tǒng)15 16。紅外傳感器固定在天花板上，并使相鄰的傳感器的感應(yīng)范圍有重疊。當(dāng)它感應(yīng)到一名住戶時(shí)，通過多個(gè)紅外傳感器的綜合，能夠比較準(zhǔn)確的確定住戶的位置。該系統(tǒng)不僅

6、具有非終端方式的特有優(yōu)點(diǎn)，還避免了侵犯隱私，擴(kuò)展性不佳和維護(hù)困難的問題。為了證明其有效性，已經(jīng)在實(shí)驗(yàn)平臺(tái)上通過了各種不同測(cè)試環(huán)境下的實(shí)驗(yàn)性評(píng)估。包括此簡(jiǎn)介，本文共分為四個(gè)部分，第二部分介紹基于紅外傳感器的室內(nèi)定位感應(yīng)系統(tǒng)架構(gòu)（PILAS）以及定位識(shí)別算法。第三部分介紹了基于紅外傳感器的住戶檢測(cè)法和在實(shí)驗(yàn)測(cè)試平臺(tái)上的不同環(huán)境下評(píng)估系統(tǒng)的表現(xiàn)。最后一部分為總結(jié)和結(jié)論。2 基于熱釋電紅外傳感器的室內(nèi)感應(yīng)定位系統(tǒng)架構(gòu)2.1智能家居的結(jié)構(gòu)鑒于智能家居的室內(nèi)環(huán)境，室內(nèi)感應(yīng)定位系統(tǒng)必須滿足一下條件。第一，由于需要在各種大小不同的房間里安裝大量傳感器來(lái)感知智能家居中的住戶，因此定位感應(yīng)系統(tǒng)需保持較低的成本。第

7、二，傳感器的安裝必須是靈活可變的，因?yàn)楦鱾€(gè)房間的形狀結(jié)構(gòu)不同，并且還有各樣阻礙傳感器正常工作的家電和家具。第三，要求定位感應(yīng)系統(tǒng)使用的傳感器能夠抵御很強(qiáng)的噪聲，這是因?yàn)橹悄芗揖幽芾酶鞣N無(wú)線傳輸技術(shù)，比如無(wú)線局域網(wǎng)，射頻系統(tǒng)，它們都會(huì)產(chǎn)生電磁噪聲，并且光或溫度的巨大變化也會(huì)影響傳感器的正常工作。最后該系統(tǒng)的精度可以，根據(jù)房間類型作出最合適的調(diào)節(jié)。盡管基于熱釋電紅外傳感器的這個(gè)系統(tǒng)有諸多的優(yōu)點(diǎn)，但在眾多滿足要求的產(chǎn)品中并不能吸引人們更多的關(guān)注。它已應(yīng)用于感應(yīng)燈（當(dāng)它感應(yīng)到人體移動(dòng)時(shí)使燈自動(dòng)打開），并且成本低于許多其他種類的感應(yīng)器。另外，由于熱釋電紅外傳感器感應(yīng)的是人體發(fā)出的9.410.4微米波長(zhǎng)

8、的紅外線，從溫度、濕度和電磁噪聲來(lái)說(shuō)，這種波長(zhǎng)相對(duì)周圍環(huán)境較為明顯。而且，它可以通過調(diào)整感應(yīng)半徑來(lái)控制定位精度，并容易安裝在天花板上，這樣就不會(huì)受到房間結(jié)構(gòu)和障礙物的影響。圖2顯示的是為住戶提供基于位置的智能服務(wù)的PILAS智能家居框架。在這個(gè)框架下，包括熱釋電紅外傳感器、房屋終端、智能家居服務(wù)器和家用電器在內(nèi)的各種設(shè)備通過家庭網(wǎng)絡(luò)系統(tǒng)連接在一起。每個(gè)房間被視為一個(gè)單元，并在每個(gè)單元的天花板上安裝適當(dāng)數(shù)量的傳感器，為定位服務(wù)提供足夠的定位精度。每個(gè)紅外傳感器周期性的感應(yīng)住戶位置，然后將感應(yīng)信息通過家庭網(wǎng)絡(luò)系統(tǒng)傳輸?shù)椒课萁K端。因此，房屋終端通過集合來(lái)自同一個(gè)單元的傳感器信息來(lái)確定住戶的位置，再將

9、住戶位置傳輸?shù)街悄芗揖臃?wù)器，服務(wù)器就會(huì)控制家用電器為住戶提供基于位置的定位服務(wù)。圖2 PILAS智能家居框架在這個(gè)框架內(nèi)，智能家居服務(wù)器具有以下功能：（1）虛擬地圖發(fā)生器為智能家居提供虛擬地圖（生成虛擬地圖），并在虛擬地圖中標(biāo)出由房屋終端提供的住戶位置信息（標(biāo)注住戶位置）。然后，它通過連接住戶的連續(xù)定位點(diǎn)來(lái)繪制住戶的運(yùn)動(dòng)軌跡（追蹤住戶運(yùn)動(dòng)）。（2）家電控制器通過家庭網(wǎng)絡(luò)系統(tǒng)發(fā)送控制命令給家用電器為住戶提供智能服務(wù)。（3）運(yùn)動(dòng)模式預(yù)測(cè)器保存當(dāng)前的住戶運(yùn)動(dòng)軌跡、家電的動(dòng)作和反映居家環(huán)境的參數(shù)，比如時(shí)間、溫度、濕度、光照度。儲(chǔ)存足夠的信息后，它可能會(huì)使家電主動(dòng)提供滿足人們需要的人性化的智能服務(wù)。例

10、如，如果智能家居服務(wù)器“知道”住戶通常在早上7點(diǎn)醒來(lái)，之后要淋浴，它也許就會(huì)在那一時(shí)間打開燈并播放音樂。另外，住戶的淋浴水溫也會(huì)被自動(dòng)記錄。2.2定位識(shí)別算法為了確定住戶在房間里的位置，要使用一組熱釋電紅外傳感器，如圖3所示。在此圖中，每個(gè)傳感器的感應(yīng)面呈圓形并且相鄰的幾個(gè)傳感器有重疊的感應(yīng)范圍。因此，當(dāng)住戶進(jìn)入某一感應(yīng)區(qū)域后，系統(tǒng)根據(jù)從房間內(nèi)的所有傳感器收集到的感應(yīng)信息判斷他/她是否屬于這一感應(yīng)區(qū)。例如，當(dāng)一位住戶進(jìn)入B感應(yīng)區(qū)，a，b傳感器輸出“ON”信號(hào)，而c傳感器輸出“OFF”信號(hào)。收集輸出信號(hào)后，該算法可以推斷出住戶屬于B感應(yīng)區(qū)。根據(jù)傳感器的數(shù)量和傳感信號(hào)“ON”的排列，住戶的位置通常

11、有以下幾種情況。首先，如果只有一個(gè)傳感器輸出“ON”信號(hào)，那么認(rèn)為住戶處于該傳感器感應(yīng)區(qū)域的中心位置。其次，如果有兩個(gè)相鄰的傳感器輸出“ON”信號(hào)，那么認(rèn)為住戶位于兩傳感器的連線中心點(diǎn)處。最后，如果有三個(gè)或者更多的傳感器輸出“ON”信號(hào)，則認(rèn)為住戶位于所有這些傳感器的面心處。比如，假設(shè)住戶位于圖中的點(diǎn)1處，只有一個(gè)傳感器a輸出“ON”信號(hào)，而當(dāng)住戶位于點(diǎn)2處，傳感器a和b都輸出“ON”信號(hào)。圖3 熱釋電紅外傳感器的定位識(shí)別算法這個(gè)系統(tǒng)的定位精度定義為假設(shè)點(diǎn)和住戶之間的最大距離。例如當(dāng)住戶進(jìn)入A感應(yīng)區(qū)，住戶被假設(shè)在點(diǎn)1處。在此假設(shè)中，住戶可以代表一個(gè)點(diǎn)，熱釋電紅外傳感器的感應(yīng)半徑為1米，故定位精

12、確度是1米。因此，當(dāng)住戶位于A感應(yīng)區(qū)的邊緣時(shí)最有可能發(fā)生判斷錯(cuò)誤。另外，當(dāng)住戶位于B感應(yīng)區(qū)時(shí)，他被假設(shè)為在點(diǎn)2處，定位誤差最大時(shí)就是實(shí)際上住戶位于點(diǎn)3處。在這種情況下，定位誤差達(dá)1.5米，即在點(diǎn)2和點(diǎn)3之間距離是1.5米。因此，圖3顯示的這個(gè)系統(tǒng)定位精度被視為1米，這是每個(gè)區(qū)域定位精度的最大有效值。傳感器的數(shù)量和它們感應(yīng)區(qū)域的面積決定了PILAS的定位精度，必須合理安排傳感器，以保證系統(tǒng)的精確度。為了準(zhǔn)確判斷住戶位置，提高系統(tǒng)的精確度，這就需要給定數(shù)量的傳感器有更大的感應(yīng)范圍。圖4顯示的是一些傳感器不同排列產(chǎn)生的不同感應(yīng)范圍。圖4（a）和（b）顯示的是9個(gè)傳感器不同排列分別產(chǎn)生的40和21的感

13、應(yīng)范圍。如果從感應(yīng)范圍來(lái)看圖4（a） 中的排列比圖4（b）的好。但是，按照（a）中傳感器的排列，在某些區(qū)域?qū)o(wú)法感應(yīng)到住戶位置，并且定位精度低于（b）。圖4（c） 中顯示12個(gè)傳感器排列成28的感應(yīng)范圍內(nèi)沒有任何盲點(diǎn)。圖4 定位精度取決于熱釋電紅外傳感器的排列 （a）40感應(yīng)范圍 （b）21感應(yīng)范圍 （c） 12個(gè)傳感器構(gòu)成的28感應(yīng)范圍當(dāng)傳感器如圖4（c）一樣排列在房間邊緣，他有時(shí)會(huì)產(chǎn)生尷尬的結(jié)果。圖5就是一個(gè)例子。圖5（a）顯示的是住戶的運(yùn)動(dòng)路線。如果我們通過使用傳感器定位或相鄰傳感器的中點(diǎn)來(lái)標(biāo)記系統(tǒng)假設(shè)的住戶位置點(diǎn)，那么運(yùn)動(dòng)路線就會(huì)變?yōu)槿鐖D5（b）顯示的折形路線。為了緩解這一問題，我們認(rèn)

14、為處于房間邊緣的傳感器稍微向內(nèi)補(bǔ)償，結(jié)果就如圖5（c）。圖5 中心點(diǎn)外部傳感器的補(bǔ)償效果（a）住戶的運(yùn)動(dòng)路線 （b）外部傳感器補(bǔ)償之前 （c） 外部傳感器補(bǔ)償之后3 PILAS的性能評(píng)估3.1 使用熱釋電紅外傳感器的住戶檢測(cè)法結(jié)合來(lái)自同一單元內(nèi)的所有傳感器的輸出并判斷是否是單一的傳感器是“ON”或“OFF”來(lái)確定住戶位置，這直接影響定位精度?？偟膩?lái)說(shuō)，由于“ON/OFF”的值是通過預(yù)先設(shè)定的閾值和從紅外傳感器的模擬信號(hào)中抽樣獲得的數(shù)字輸出信號(hào)相比較來(lái)確定，故選擇一個(gè)合適的信號(hào)電平作為閾值是很有必要的。例如，使用非終端方式的Smart Floor，因?yàn)楫?dāng)住戶站在某一點(diǎn)不動(dòng)時(shí)，他的重量通過壓力傳感

15、器的輸出是一個(gè)恒定電壓，故比較閾值和傳感器的輸出值可以準(zhǔn)確的確定住戶位置。但是，紅外傳感器測(cè)量的是人體移動(dòng)產(chǎn)生的紅外線信號(hào)的變化，它輸出的是模擬信號(hào)，如圖6所示。也就是說(shuō)，當(dāng)住戶進(jìn)入一個(gè)紅外感應(yīng)區(qū)，住戶發(fā)出紅外輻射逐漸增強(qiáng)，紅外傳感器輸出一個(gè)增大的電壓信號(hào)。相反的，當(dāng)住戶離開感應(yīng)區(qū)，電壓信號(hào)變小。如果住戶站在感應(yīng)區(qū)不動(dòng)，紅外輻射不變，輸出電壓為0。因此，當(dāng)住戶一直停留在某一感應(yīng)區(qū)內(nèi)，只用紅外傳感器的電壓或電流閾值是很難判斷的。圖6 熱釋電紅外傳感器的輸出信號(hào)為了保證定位精度，住戶檢測(cè)法必須滿足下面幾個(gè)要求。首先，如果感應(yīng)區(qū)內(nèi)沒有住戶，傳感器不能輸出“ON”信號(hào)。也就是說(shuō)，傳感器不能被如移動(dòng)的寵

16、物、溫度變化、陽(yáng)光等干擾而誤報(bào)。其次，它應(yīng)該能精確確定住戶進(jìn)入和離開感應(yīng)區(qū)時(shí)的時(shí)間點(diǎn)。那么根據(jù)傳感器的變化特征、住戶的速度和高度應(yīng)該能完全確定時(shí)間點(diǎn)。最后，因?yàn)樽粼诟袘?yīng)區(qū)內(nèi)沒有移動(dòng)時(shí)，紅外傳感器的輸出電壓不能超過閾值電壓，故判定住戶是否停留在感應(yīng)區(qū)內(nèi)是很有必要的。為了滿足這些要求，下面介紹基于熱釋電紅外傳感器的住戶檢測(cè)法來(lái)實(shí)現(xiàn)。第一，使用菲涅爾透鏡以消除由于寵物或溫度變化引起的傳感器誤報(bào)。菲涅爾透鏡安裝在熱釋電紅外傳感器前部，它只允許人體的紅外線波長(zhǎng)通過而阻止其他波長(zhǎng)的紅外線。第二，當(dāng)傳感器輸出電壓超過正閾值電壓且保持?jǐn)?shù)個(gè)采樣周期，那么認(rèn)為住戶進(jìn)入了感應(yīng)區(qū)。這種方法中的閾值必須有足夠的能力區(qū)

17、別人體的紅外線變化和寵物或者溫度變化引起的環(huán)境紅外信號(hào)。此外，當(dāng)傳感器輸出電壓低于負(fù)閾值電壓且保持?jǐn)?shù)個(gè)采樣周期，則認(rèn)為住戶離開了感應(yīng)區(qū)。最后，當(dāng)輸出電壓處于正負(fù)閾值電壓之間，例如住戶站在某一感應(yīng)區(qū)內(nèi)不移動(dòng)，此區(qū)域的紅外傳感器輸出就從“ON”變?yōu)椤癘FF”。在這段時(shí)間內(nèi)，如果安裝在它附近的其它傳感器不輸出“ON”信號(hào)，此法認(rèn)為住戶保持在此區(qū)域。3.2 使用實(shí)驗(yàn)測(cè)試平臺(tái)進(jìn)行性能評(píng)估為了驗(yàn)證PILAS法的可行性，使用實(shí)驗(yàn)測(cè)試平臺(tái)進(jìn)行測(cè)試。智能家居中的基于位置的智能服務(wù)不要求有很高的定位精度，我們?cè)O(shè)計(jì)的系統(tǒng)定位精度為0.5米。圖7顯示的是房間中的實(shí)驗(yàn)測(cè)試平臺(tái)測(cè)量4×4×2.5米（寬

18、×長(zhǎng)×高）范圍。此實(shí)驗(yàn)中，12個(gè)傳感器如圖4（c）排列在天花板上。用美國(guó)Atmel公司的AT89C51CC001單片機(jī)17處理信號(hào)并判斷“ON/OFF”，用日本陶瓷公司RE431B型熱釋電紅外傳感器和NL-11型菲涅爾透鏡。特別注意，每個(gè)紅外傳感器套上一個(gè)圓環(huán)，只允許直徑2米的感應(yīng)范圍。圖8顯示的是帶圓環(huán)的實(shí)驗(yàn)結(jié)果。在圖中，當(dāng)住戶通過感應(yīng)圓時(shí)RE431B型傳感器輸出信號(hào)如（a）所示，而當(dāng)住戶在感應(yīng)圓區(qū)域內(nèi)移動(dòng)時(shí)輸出不規(guī)則的信號(hào)，如（b）所示。最后，當(dāng)住戶走出感應(yīng)圓，沒有信號(hào)輸出，如（c）所示。根據(jù)這些測(cè)試結(jié)果，證明熱釋電紅外傳感器只能感應(yīng)到在感應(yīng)區(qū)內(nèi)的住戶。此外，為了判斷信

19、號(hào)是“ON”還是“OFF”，并考慮到外部環(huán)境干擾，有必要為RE431B傳感器設(shè)置一個(gè)閾值。最初幾次實(shí)驗(yàn)用于確定閾值，確定時(shí)要考慮是由空調(diào)或電熱器或其他干擾（比如風(fēng)或陽(yáng)光）引起的內(nèi)部溫度變化。基于這些實(shí)驗(yàn)結(jié)果，當(dāng)RE431B傳感器的閾值設(shè)為±0.4V時(shí)，外部溫度變化不影響其感應(yīng)住戶的性能。此外，我們還驗(yàn)證了在同樣的閾值下，寵物也不會(huì)對(duì)感應(yīng)性能造成影響。圖7 PILAS實(shí)驗(yàn)測(cè)試平臺(tái)圖8 保證在邊角有準(zhǔn)確的感應(yīng)距離4 總結(jié)和結(jié)論本文提出了一種以熱釋電紅外傳感器為基礎(chǔ)的室內(nèi)感應(yīng)定位系統(tǒng)，此系統(tǒng)可以為智能家居中的基于位置的智能服務(wù)估計(jì)出住戶的位置。本文介紹了感應(yīng)定位系統(tǒng)中的智能家居框架，綜合紅

20、外傳感器收集的信息實(shí)現(xiàn)的定位識(shí)別算法。此外，本文提出了住戶檢測(cè)發(fā)放。最后，對(duì)PILAS進(jìn)行了性能評(píng)估。根據(jù)各種條件下的多次實(shí)驗(yàn)，我們可以證實(shí)，該P(yáng)ILAS法可以相當(dāng)準(zhǔn)確的估計(jì)出住戶的位置。此外，由于該系統(tǒng)的定位精度小于0.5米，不需要任何定位識(shí)別終端，因此非常實(shí)用。而且，通過增加傳感器的數(shù)量，或排列傳感器使它們的感應(yīng)范圍相等，或?qū)ν獠總鞲衅鞯闹行倪M(jìn)行補(bǔ)償，可以提高它的定位精度。由于該系統(tǒng)的定位精度根據(jù)傳感器的不同排列方法有所不同，我們需要找到最優(yōu)排列方法以提供最準(zhǔn)確的定位精度。為了提高定位精度，還需要對(duì)紅外傳感器的處理方法應(yīng)用更先進(jìn)的技術(shù)，最后，建議PILAS系統(tǒng)應(yīng)該擴(kuò)展到處理一個(gè)房間中的多人

21、定位。A Pyroelectric Infrared Sensor-based Indoor Location-Aware System for the Smart HomeSuk Lee, Member, IEEE, Kyoung Nam Ha, Kyung Chang Lee, Member, IEEEAbstract：Smart home is expected to offer various intelligent services by recognizing residents along with their life style and feelings. One of th

22、e key issues for realizing the smart home is how to detect the locations of residents. Currently, the research effort is focused on two approaches: terminal-based and non-terminal-based methods. The terminal -based method employs a type of device that should be carried by the resident while the non-

23、terminal-based method requires no such device. This paper presents a novel non-terminal-based approach using an array of pyroelectric infrared sensors (PIR sensors) that can detect residents. The feasibility of the system is evaluated experimentally on a test bed.Index Terms：smart home, location-bas

24、ed service, pyroelectric infrared sensor (PIR sensor), location-recognition algorithm1 IntroductionThere is a growing interest in smart home as a way to offer a convenient, comfortable, and safe residential environment 1, 2. In general, the smart home aims to offer appropriate intelligent services t

25、o actively assist in the residents life such as housework, amusement, rest, and sleep. Hence, in order to enhance the residents convenience and safety, devices such as home appliances, multimedia appliances, and internet appliances should be connected via ahome network system, as shown in Fig. 1, an

26、d they should be controlled or monitored remotely using a television (TV) or personal digital assistant (PDA) 3, 4. Fig. 1. Architecture of the home network system for smart homeEspecially, attention has been focused on location-based services as a way to offer high-quality intelligent services, whi

27、le considering human factors such as pattern of living, health, and feelings of a resident 5-7. That is, if the smart home can recognize the residents pattern of living or health, then home appliances should be able to anticipate the residents needs and offer appropriate intelligent service more act

28、ively. For example, in a passive service environment, the resident controls the operation of the HVAC (heating, ventilating, and air conditioning) system, while the smart home would control the temperature and humidity of a room according to the residents condition. Various indoor location-aware sys

29、tems have been developed to recognize the residents location in the smart home or smart office. In general, indoor location-aware systems have been classified into three types according to the measurement technology: triangulation, scene analysis, and proximity methods 8. The triangulation method us

30、es multiple distances from multiple known points. Examples include Active Badges 9, Active Bats 10, and Easy Living 11, which use infrared sensors, ultrasonic sensors, and vision sensors, respectively. The scene analysis method examines a view from a particular vantage point. Representative examples

31、 of the scene analysis method are MotionStar 12, which uses a DC magnetic tracker, and RADAR 13, which uses IEEE 802.11 wireless local area network (LAN). Finally, the proximity method measures nearness to a known set of points. An example of the proximity method is Smart Floor 14, which uses pressu

32、re sensors.Alternatively, indoor location-aware systems can be classified according to the need for a terminal that should be carried by the resident. Terminal-based methods, such as Active Bats, do not recognize the residents location directly, but perceive the location of a device carried by the r

33、esident, such as an infrared transceiver or radio frequency identification (RFID) tag. Therefore, it is impossible to recognize the residents location if he or she is not carrying the device. In contrast, non-terminal methods such as Easy Living and Smart Floor can find the residents location withou

34、t such devices. However, Easy Living can be regarded to invade the residents privacy while the Smart Floor has difficulty with extendibility and maintenance.This paper presents a non-terminal based location-aware system that uses an array of pyroelectric infrared (PIR) sensors 15, 16. The PIR sensor

35、s on the ceiling detect the presence of a resident and are laid out so that detection areas of adjacent sensors overlap. By combining the outputs of multiple PIR sensors, the system is able to locate a resident with a reasonable degree of accuracy. This system has inherent advantage of non-terminal

36、based methods whileavoiding privacy and extendibility, maintenance issues. In order to demonstrate its efficacy, an experimental test bed has been constructed, and the proposed system has been evaluated experimentally under various experimental conditions. This paper is organized into four sections,

37、 including this introduction. Section II presents the architecture of the PIR sensor-based indoor location-aware system (PILAS), and the location-recognition algorithm. Section III describes a resident-detection method using PIR sensors, and evaluates the performance of the system under various cond

38、itions using an experimental test bed. Finally, a summary and the conclusions are presented in Section IV.2 Architecture of the pirsensor-based indoor location-aware system2.1 Framework of the smart homeGiven the indoor environment of the smart home, an indoor location-aware system must satisfy the

39、following requirements. First, the location-aware system should be implemented at arelatively low cost because many sensors have to be installed in rooms of different sizes to detect the resident in the smart home. Second, sensor installation must be flexible because the shape of each room is differ

40、ent and there are obstacles such as home appliances and furniture, which prevent the normal operation of sensors. The third requirement is that the sensors for the location-aware system have to be robust to noise, and should not be affected by their surroundings. This is because the smart home can m

41、ake use of various wireless communication methods such as wireless LAN or radio-frequency (RF) systems, which produce electromagnetic noise, or there may be significant changes in light or temperature that can affect sensor performance. Finally, it is desirable that the systems accuracy is adjustabl

42、e according to room types.Among many systems that satisfy the requirement, the PIR sensor-based system has not attracted much attention even though the system has several advantages. The PIR sensors,which have been used to turn on a light when it detects human movement, are less expensive than many

43、other sensors. In addition, because PIR sensors detect the infrared wavelengthemitted from humans between 9.410.4 m, they are reasonably robust to their surroundings, in terms of temperature, humidity, and electromagnetic noise. Moreover, it ispossible to control the location accuracy of the system

44、by adjusting the sensing radius of a PIR sensor, and PIR sensors are easily installed on the ceiling, where they are not affected by the structure of a room or any obstacles. Figure 2 shows the framework for the PILAS in a smart home that offers location-based intelligent services to a resident. Wit

45、hin this framework, various devices are connected via a home network system, including PIR sensors, room terminals, a smart home server, and home appliances. Here, each room is regarded as a cell, and the appropriate number of PIR sensors is installed on the ceiling of each cell to provide sufficien

46、t location accuracy for the location-based services. Each PIR sensor attempts to detect the resident at a constant period, and transmits its sensing information to a room terminal via the home network system. Fig. 2. Framework of smart home for the PILAS.Consequently, the room terminal recognizes th

47、e residents location by integrating the sensor information received from all of the sensors belonging to one cell, and transmits the residents location to the smart home server that controls the home appliances to offer location-based intelligent services to the resident.Within this framework, the s

48、mart home server has the following functions. 1) The virtual map generator makes a virtual map of the smart home (generating a virtual map), and writes the location information of the resident, which is received from a room terminal, on the virtual map (writing the residents location). Then, it make

49、s a moving trajectory of the resident by connecting the successive locations of the resident (tracking the residents movement). 2) The home appliance controller transmits control commands to home appliances via the home network system to provide intelligent services to the resident. 3) The moving pa

50、ttern predictor saves the current movement trajectory of the resident, the current action of home appliances, and parameters reflecting the current home environment such as the time, temperature, humidity, and illumination. After storing sufficient information, it may be possible to offer human-orie

51、nted intelligent services in which the home appliances spontaneously provide services to satisfy human needs. For example, if the smart home server “knows” that the resident normally wakes up at 7:00 A.M. and takes a shower, it may be possible to turn on the lamps and some music. In addition, the te

52、mperature of the shower water can be set automatically for the resident.2.2 Location-recognition algorithmIn order to determine the location of a resident within a room, an array of PIR sensors are used as shown in Fig. 3. In the figure, the sensing area of each PIR sensor is shown as a circle, and

53、the sensing areas of two or more sensors overlap. Consequently, when a resident enters one of the sensing areas, the system decides whether he/she belongs to any sensing area by integrating the sensing information collected from all of the PIR sensors in the room. For example, when a resident enters

54、 the sensing area B, sensors a and b output ON signals, while sensor c outputs OFF signal. After collecting outputs, the algorithm can infer that the resident belongs to the sensing area B. According to the number of sensors and the arrangement of the sensors signaling ON, the residents location is

55、deter-mined in the following manner. First, if only one sensor outputs ON signal, the resident is regarded to be at the center of the sensing area of the corresponding sensor. If the outputs of two adjacent sensors are ON, the residents location is assumed to be at the point midway between the two s

56、ensors. Finally, if three or more sensors signal ON, the resident is located at the centroid of the centers of the corresponding sensors. For example, it is assumed that the resident is located at point 1 in the figure when only sensor a signals ON, while the resident is located at point 2 when sens

57、ors a and b both output ON signals. The location accuracy of this system can be defined the maximum distance between the estimated points and the resident. For example, when a resident enters sensing area A, the resident is assumed to be at point 1. On the assumption that a resident can be represent

58、ed by a point and the radius of the sensing area of a PIR sensor is 1 m, we know that the location accuracy is 1 m because the maximum error occurs when the resident is on the boundary of sensing area A. Alternatively, when the resident is in sensing area B, the resident is assumed to be at point 2, and the maximum location error occurs when the resident is actually at point 3. In this case, the error is 3 / 2 m which is the distance between points 2 and 3. Therefore, the location accuracy of the total system shown in Fig. 3 can be regarded as 1