通信工程基于WIFI的無線傳感器采集系統(tǒng)設(shè)計

1、長春理工大學(xué)畢業(yè)設(shè)計任務(wù)書題目名稱：基于WIFI的無線傳感器采集系統(tǒng)設(shè)計 學(xué)生姓名：華丹陽 起止日期：2016.3.32015.6.22題目要求(包括主要技術(shù)參數(shù)):1 題目內(nèi)容：設(shè)計基于WIFI技術(shù)的傳感器信息采集系統(tǒng)，實現(xiàn)數(shù)據(jù)信息的網(wǎng)絡(luò)發(fā)布2 具體要求及技術(shù)參數(shù)： 1.學(xué)習(xí)WIFI技術(shù)； 2.設(shè)計基于WIFI無線傳感器信息采集系統(tǒng)硬件； 3.實現(xiàn)節(jié)點信息無線傳輸并上網(wǎng)發(fā)布。 指導(dǎo)教師簽字 系主任簽字 年 月 日 開題報告1、本課題研究的目的、意義；2、國內(nèi)外研究現(xiàn)狀；3、擬采取的研究路線；4、進度安排；1.課題研究的目的及意義 社會信息化日新月異,新技術(shù)層出不窮。而傳感器技術(shù)在科技發(fā)展中卻

2、舉足輕重。微型化、多功能化、網(wǎng)絡(luò)化和智能化乃大勢所趨,無線傳感器網(wǎng)絡(luò)則詮釋了這些優(yōu)勢。將來人們將通過遍布周圍的傳感器網(wǎng)絡(luò)直接感知客觀世界，極大的改變?nèi)藗冋J(rèn)識世界、改造世界的能力。傳感器網(wǎng)絡(luò)覆蓋范圍大。能根據(jù)實際情況設(shè)計無線傳感器網(wǎng)絡(luò)的規(guī)模，有利于應(yīng)用范圍的擴展。傳感器網(wǎng)絡(luò)具有自組織功能。組網(wǎng)不需要任何固定的網(wǎng)絡(luò)設(shè)備，傳感器節(jié)點通過分布式網(wǎng)絡(luò)協(xié)議形成自組織網(wǎng)絡(luò)，能夠自動調(diào)整來適應(yīng)節(jié)點的變化，網(wǎng)絡(luò)中的節(jié)點可以快速、自動的組成一個獨立的網(wǎng)絡(luò)。可以動態(tài)拓?fù)?，無線傳感器網(wǎng)絡(luò)中可以隨時添加或減少節(jié)點而并不影響網(wǎng)絡(luò)其他節(jié)點數(shù)據(jù)的正常傳輸。功耗小，電池供電，網(wǎng)絡(luò)節(jié)點一般都能工作3年左右，甚至更長。本文根據(jù)傳感

3、器網(wǎng)絡(luò)發(fā)展?fàn)顩r，設(shè)計出基于WIFI的無線傳感器網(wǎng)絡(luò)，相比于傳統(tǒng)的無線傳感器網(wǎng)絡(luò)，能夠非常容易的與現(xiàn)有網(wǎng)絡(luò)進行無縫的連接，相對降低組網(wǎng)成本和功耗。WIFI 無線傳感器網(wǎng)絡(luò)具有傳輸速率快、組網(wǎng)便捷等優(yōu)點。2.國內(nèi)外研究現(xiàn)狀 1）國內(nèi)的研究現(xiàn)狀 山東省科學(xué)院與沈陽自動化研究所等研究單位及多所高校（如哈爾濱工業(yè)大學(xué)、北京郵電大學(xué)等）在無線傳感器網(wǎng)絡(luò)網(wǎng)絡(luò)協(xié)議的研究與優(yōu)化等方面也進行了大量的工作。除此之外，國家 863 計劃、973 計劃也對無線傳感器網(wǎng)絡(luò)的研究進行了相關(guān)規(guī)劃。比如中科院寧波軟件所和上海微系統(tǒng)所研究出自己的開發(fā)平臺,中國科技大學(xué),西北工業(yè)大學(xué)等院校都展 了路由層、數(shù)據(jù)鏈路層方面的研究。國家

4、對傳感器網(wǎng)絡(luò)的研究也非常重視,國家自然科學(xué)基金委員會從2003年起開始設(shè)立了無線傳感器網(wǎng)絡(luò)相關(guān)研究課題,國家的“863”項目、國家自然科學(xué)基金項目、各省區(qū)的自然科學(xué)基金項目的課題中都有相當(dāng)?shù)谋壤沁M行無線傳感器網(wǎng)絡(luò)研究的。 2）國外的研究現(xiàn)狀 在傳感器網(wǎng)絡(luò)方面,加州大學(xué)伯克利分校提出了應(yīng)用網(wǎng)絡(luò)連通性重構(gòu)傳感器位置的方法,并研究出一個專門用于傳感器網(wǎng)絡(luò)節(jié)點的操作系統(tǒng)TinyOS。加州大學(xué)洛杉肌分校開發(fā)了一個無線傳感器網(wǎng)絡(luò)和一個無線傳感器網(wǎng)絡(luò)模擬環(huán)境,用于考察傳感器網(wǎng)絡(luò)各方面的問題。南加州大學(xué)提出了在生疏環(huán)境部署移動傳感器的方法、傳感器網(wǎng)絡(luò)監(jiān)視結(jié)構(gòu)及其聚集函數(shù)計算方法、節(jié)省能源的計算、聚集的樹構(gòu)造

5、算法等。麻省理工學(xué)院已經(jīng)著手研究超低功耗無線傳感器網(wǎng)絡(luò)的問題,試圖解決超低功耗無線傳感器系統(tǒng)的方法和技術(shù)問題。在傳感器網(wǎng)絡(luò)通信協(xié)議方面,人們首先對已有的因特網(wǎng)和自組織無線網(wǎng)絡(luò)的通信協(xié)議進行了研究,發(fā)現(xiàn)這些協(xié)議不適用于傳感器網(wǎng)絡(luò)應(yīng)用場合。加州大學(xué)伯克利分校研究了傳感器網(wǎng)絡(luò)的數(shù)據(jù)查詢技術(shù),提出了實現(xiàn)可動態(tài)調(diào)整的連續(xù)查詢處理方法和管理傳感器網(wǎng)絡(luò)上多查詢方法,并研制了一個感知數(shù)據(jù)庫系統(tǒng)TinyDB。南加州大學(xué)研究了傳感器網(wǎng)絡(luò)上的聚集函數(shù)的計算方法,提出了節(jié)省能源的計算聚集的樹構(gòu)造算法,并通過實驗證明了無線通信機制對聚集計算的性能有很大的影響。 3）目前技術(shù)存在的問題 無線傳感器網(wǎng)絡(luò)即便節(jié)點靈活，可減硬

6、件成本，但依然受有限能量的制約，優(yōu)勢未能充分發(fā)揮。無線傳感器網(wǎng)絡(luò)壽命長短與節(jié)點功耗大小息息相關(guān)，應(yīng)致力于降低功耗。通信協(xié)議依然廣泛，網(wǎng)絡(luò)協(xié)議標(biāo)準(zhǔn)化較低。通信能力有限，傳輸距離不夠長，受環(huán)境變化干擾。3. 采取的研究路線首先查閱大量關(guān)于無線傳感器的相關(guān)文獻(xiàn)，選用WIFI技術(shù)作為無線傳感器網(wǎng)絡(luò)的通信技術(shù)。充分研究，并通過 WIFI 組網(wǎng)技術(shù)設(shè)計 WIFI 無線傳感器網(wǎng)絡(luò)的結(jié)構(gòu)。然后設(shè)計 WIFI 無線傳感器節(jié)點的硬件結(jié)構(gòu)。無線傳感器網(wǎng)絡(luò)節(jié)點分為核心控制模塊、外圍接口及電源管理模塊、數(shù)據(jù)采集模塊，針對各個模塊的功能進行硬件設(shè)計。 選用ARM芯片AT91SAM9G45作為處理器， 選用AD7492作為

7、A/D轉(zhuǎn)換器，選用FIFO CY7C4261作為緩存器，F(xiàn)PGA芯片選用XC3S500E， WIFI 芯片選用 RT3070。然后進行軟件設(shè)計，了解嵌入式Linus系統(tǒng)的開發(fā)環(huán)境，再進行基于FT245 USB芯片Linus驅(qū)動系統(tǒng)的設(shè)計，配置內(nèi)核，使系統(tǒng)支持 USB 接口的 WIFI 無線網(wǎng)卡。進行基于RT3070芯片的網(wǎng)卡驅(qū)動移植，最后設(shè)計WIFI的驅(qū)動程序，進行WIFI聯(lián)網(wǎng)。4.進度安排第1周第4周 資料收集，完成開題報告的撰寫，英文資料的翻譯。第5周第6周 擬定系統(tǒng)方案，進行系統(tǒng)總體設(shè)計。第7周第9周 電路設(shè)計 電路制作 程序編寫。第10周第12周 調(diào)試測試 電路調(diào)試 軟件調(diào)試。第13周

8、第15周 數(shù)據(jù)整理，撰寫論文。第16周 準(zhǔn)備答辯。5. 文獻(xiàn)綜述（2000字以上，列出主要參考文獻(xiàn)）5.1無線傳感器網(wǎng)絡(luò)概述 無線傳感器網(wǎng)絡(luò)的三個基本要素包括：無線傳感器網(wǎng)絡(luò)節(jié)點、覆蓋環(huán)境感知對象和接收數(shù)據(jù)觀察者。無線傳感器網(wǎng)絡(luò)節(jié)點是無線傳感器網(wǎng)絡(luò)最基本、最核心的組成部分，網(wǎng)絡(luò)節(jié)點主要集成相應(yīng)微型傳感器、數(shù)字信號處理器、無線通信模塊等功能單元。無線傳感器網(wǎng)絡(luò)節(jié)點按照執(zhí)行功能的不同又可劃分為傳感器節(jié)點和匯聚節(jié)點，傳感器節(jié)點完成數(shù)據(jù)的采集和通信鏈路的續(xù)傳，而匯聚節(jié)點只完成收發(fā)無線網(wǎng)絡(luò)數(shù)據(jù)和上傳給接收數(shù)據(jù)觀察者。覆蓋環(huán)境感知對象是指節(jié)點判定為價值有效的監(jiān)測目標(biāo)，可以是監(jiān)測區(qū)域的聲音、光線、溫度、震動

9、等等，節(jié)點傳感器通過數(shù)據(jù)采集、轉(zhuǎn)化為系統(tǒng)可以識別的信息資源，并最終上傳給接收數(shù)據(jù)觀察者。接收數(shù)據(jù)觀察者是無線傳感器網(wǎng)絡(luò)的終端用戶，完成采集數(shù)據(jù)的應(yīng)用。接收數(shù)據(jù)觀察者一般為終端計算器或者其它監(jiān)控設(shè)備，甚至是連接外部世界的萬維網(wǎng)，數(shù)據(jù)采集觀察者通過主動查詢或者被動接收的方式分析無線傳感器網(wǎng)絡(luò)的數(shù)據(jù)信息，并最終完成數(shù)據(jù)的分析、應(yīng)用。 圖1-1 無線傳感器網(wǎng)絡(luò)節(jié)點結(jié)構(gòu)示意圖 無線傳感器網(wǎng)絡(luò)的傳感器節(jié)點內(nèi)部結(jié)構(gòu)示意圖如圖 1-1 所示，內(nèi)部分為四個模塊：電源模塊、傳感器模塊、信息處理模塊和無線通信模塊。傳感器模塊通過傳感器觸頭感知外界信息，獲取傳感數(shù)據(jù)；無線通信模塊通過天線與其他節(jié)點通信完成數(shù)據(jù)交換。

10、無線傳感器網(wǎng)絡(luò)組成形式如圖 1-2 所示，其工作原理：首先分布于監(jiān)控區(qū)域的眾多傳感器節(jié)點通過無線通信的方式自組織成一個有傳播梯度的多跳網(wǎng)絡(luò)，接著某個傳感器節(jié)點采集接收到覆蓋范圍感知對象的有效數(shù)據(jù)，此節(jié)點將數(shù)據(jù)發(fā)送給周圍選擇的鄰居節(jié)點，鄰居節(jié)點再傳遞給自身周圍的鄰居節(jié)點，數(shù)據(jù)經(jīng)過多跳傳遞給匯聚節(jié)點（sink），匯聚節(jié)點最后再傳遞給接收數(shù)據(jù)觀察者，從而完成整個無線網(wǎng)絡(luò)的通信功能。 5.2無線傳感器網(wǎng)絡(luò)WIFI技術(shù)介紹WIFI 全稱 Wireless Fidelity14，又稱 802.11 標(biāo)準(zhǔn)，是由一個名為“無線以太網(wǎng)相容聯(lián)盟”（Wireless Ethernet Compatibility A

11、lliance, WECA）的組織所發(fā)布的業(yè)界術(shù)語，中文譯為“無線相容認(rèn)證”。它是一種短程無線傳輸技術(shù)，能夠在數(shù)百米范圍內(nèi)支持互聯(lián)網(wǎng)接入的無線電信號。WIFI技術(shù)傳輸速率快，采用直接序列擴頻技術(shù)，提供很高的傳輸速率，具有高移動性，在無線局域網(wǎng)覆蓋范圍內(nèi)，地理位置的限制進行任意移動，各個節(jié)點可以不受覆蓋范圍廣，WIFI 的覆蓋范圍半徑在 150m，但通過中繼能實現(xiàn)幾千米的通信距離。輻射小，IEE802.11 規(guī)定的發(fā)送功率是 100mW，而一般的WIFI 設(shè)備只要6070mW。而且易擴展，傳輸可靠，組網(wǎng)便捷。5.3無線傳感器網(wǎng)絡(luò)的硬件結(jié)構(gòu)在節(jié)點核心控制模塊硬件結(jié)構(gòu)中，ARM 作為一種嵌入式處理器

12、，具有高性能、低功耗、低成本、體積小等優(yōu)點。將 ARM 作為節(jié)點的主控制器可全面提高節(jié)點性能。本設(shè)計中，處理器選用 Ateml 公司的ARM芯片AT91SAM9G45，主頻達(dá)400MHZ。動態(tài)存儲器，選用 National Semiconductor 公司 64M DDR2 存儲器，工作溫度在系統(tǒng)中使用兩片，總?cè)萘窟_(dá) 128M，大幅提高 ARM 處理器的運算效率。在節(jié)點外接口與電源管理模塊中，電源管理芯片，選用 LM2596 和 LM1084，為系統(tǒng)提供 5V 和 3.3V 電壓。兩路 USB 接口控制芯片，選用雙 USB 電源開關(guān)芯片 SP2526A-2USB 和兩片 USB控制芯片 USB

13、LC6-2P6 為系統(tǒng)提供兩路 USB 接口，一路用于與無線模塊進行通信，一路用于測試數(shù)據(jù)的有限讀取。數(shù)據(jù)采集系統(tǒng)由傳感器、AD 轉(zhuǎn)換器、FPGA 組成，它的主要任務(wù)是把傳感器采集到的模擬信號轉(zhuǎn)換成數(shù)字信號。本文設(shè)計的系統(tǒng)所使用的 WIFI 無線網(wǎng)卡是 TOTOLINK 公司的 N200UA，這款 WIFI無線網(wǎng)卡的優(yōu)點在于外置天線，我們可以根據(jù)需要選用特殊形狀以及高增益的天線。無線 AP（AP，Access Point，無線接入節(jié)點）是一個包含很廣的名稱，它包含無線接入點（無線 AP）和無線路由器（含無線網(wǎng)關(guān)、無線網(wǎng)橋）等類設(shè)備的統(tǒng)稱。5.4無線傳感器采集系統(tǒng)的軟件設(shè)計5.4.1嵌入式Lin

14、us系統(tǒng)的介紹 嵌入式 Linux 是在 Linux 的基礎(chǔ)演變而成的，專門應(yīng)用于嵌入式設(shè)備中。Linux 是開放源代碼的，不存在黑箱技術(shù)，全球有眾多 Linux 愛好者，對 Linux 發(fā)展提供強大技術(shù)支持。Linux 的內(nèi)核小、執(zhí)行效率高，非常容易裁剪定制，其系統(tǒng)內(nèi)核最小只有約幾百 KB。Linux 是完全免費，與其它昂貴操作系統(tǒng)如 Vxworks 相比，容易普及。Linux 是一個跨平臺的操作，它適應(yīng)于多種處理器，到目前為止，它可以支持幾十種處理器，所以它的移植性非常好。Linux 內(nèi)核的結(jié)構(gòu)在網(wǎng)絡(luò)功能完善，支持包括百兆、千兆以太網(wǎng)絡(luò)以及無線網(wǎng)絡(luò)。5.4.2嵌入式Linus系統(tǒng)的組成及其

15、移植Linux 操作系統(tǒng)至少具有三部分：BootLoader（引導(dǎo)系統(tǒng)）、Kernel（內(nèi)核）、Rootfs（根文件系統(tǒng)）。這三部分需要寫到嵌入式系統(tǒng)的 NandFlash 中，不同的處理器，其燒寫方式有所不同。本文選用 Ateml 公司的工業(yè)級 ARM 芯片 AT91SAM9G45，該處理器 BootLoader 和 Kernel 需要使用 Ateml 公司的 SAM-BA 軟件通過 USB 口進行燒寫，而 Rootfs 是通過網(wǎng)口進行燒寫。 驅(qū)動程序設(shè)計好后，需要將其編譯生成二進制文件，驅(qū)動程序不同于應(yīng)用程序的編譯，由于驅(qū)動程序是 Linux 內(nèi)核的一部分，所以需要將驅(qū)動程序源碼放到 Li

16、nux 內(nèi)核源碼中。Linux 內(nèi)核是支持 WIFI 無線網(wǎng)絡(luò)通信的，但是需要對其進行配置，才能使用。所以要配置內(nèi)核，使內(nèi)核支持 USB2.0，對內(nèi)核進行相關(guān)的配置后，系統(tǒng)就完全支持 USB 接口的 WIFI 無線網(wǎng)卡了。參考文獻(xiàn)：1王亞超，寧濱，基于無線傳感器網(wǎng)絡(luò)的城軌列車運行能耗數(shù)據(jù)采集系統(tǒng)設(shè)計D，控制工程2015.6.2林一多，高德云. 基于 ARM 的無線傳感器網(wǎng)絡(luò) MAC 協(xié)議設(shè)計與實現(xiàn) J.計算機應(yīng)用，2010,30（5）:1145-1148.3林彬，基于 WIFI 的無線傳感器網(wǎng)絡(luò)檢測系統(tǒng)的設(shè)計D.西南交通大學(xué).2011.5.4黃茂芹，基于FPGA的實時無線傳感器網(wǎng)絡(luò)系統(tǒng)設(shè)計D，

17、電子科技大學(xué) 電子與通信工程，2013.5.5曾強，張志杰，WIFI無線傳感器網(wǎng)絡(luò)的設(shè)計與實現(xiàn)D，中北大學(xué)，2012.6.6王賽博，劉素凱，毛先柏，無線傳感器網(wǎng)絡(luò)綜述J，信息通信，2014.8.7秦邵華，無線傳感器多信道通信技術(shù)的研究D，山東大學(xué)，2014.6.8孫宇,基于嵌入式 Linux 的無線傳感器網(wǎng)絡(luò)基站軟件設(shè)計與實現(xiàn) D,吉林大學(xué),2009.4.9董云鵬.無線傳感器網(wǎng)絡(luò)節(jié)點的設(shè)計與實現(xiàn)D.北京:北京交通大學(xué),200810潘洋.基于 ARM 的無線測控系統(tǒng)J.微計算機信息.2009.25(4):156-157.11閻連龍.基于 ARM 的無線數(shù)據(jù)采集系統(tǒng) J.廣東技術(shù)師范學(xué)院學(xué)報，20

18、09.3:25-28.12 Camera calibration toolbox for matlab. bouguetj/calib doc/index.html#links.13 Free space optics:technology insight. .14 Irda. .15Mipav.16 Stan moore astronomy. size.html.17 A.Ashok, M.Gruteser, N. B. Mandayam, J. Silva, K. Dana, and M.Varga. Challenge: Mobile optical networks through v

19、isual mimo. In MobiCom 10: Proceedings of the sixteenth annual international conference on Mobile computing and networking, pages 105112, New York, NY, USA, 2010. ACM.外文文獻(xiàn)： Characterizing Multiplexing and Diversity in Visual MIMOAbstract Mobile optical wireless has so far been limited to very short

20、ranges for high data rate systems. It may be feasible to overcome the data rate limitations over large transmission range in optical wireless through camera receivers and light emitting transmitter arrays through a concept what we call ”visual MIMO”. In this concept multiple transmit elements of a l

21、ight emitting array (LEA) are used as transmitters to communicate to the individual pixel elements of the camera which act as multiple receive elements to create the visual MIMO channel. Multiplexing information over parallel data channels albeit be very similar to RF MIMO in concept, the visual MIM

22、O approach dramatically differs in its characterization. In visual MIMO since the received signal is essentially the image of the transmitting element, the perspective distortions in the visual channel dominate over some of the important properties of a RF wireless channel such as distance based att

23、enuation and multipath fading. Some of the prominent perspective distortions include the reduction in the size of the image with distance and skew/rotation in the image due to angular view. Further lens blur (typically due to focus imperfection or jerks while capturing the image) can also significan

24、tly depreciate the image quality. In this paper we will detail how MIMO techniques such as multiplexing and diversity are characterized based on the effect of perspective distortions. Based our visual MIMO channel model we will derive the analytical channel capacity of the visual MIMO channel and us

25、ing the same we illustrate the significance of parameters such as distance, viewing angle and blur in characterizing multiplexing and diversity in visual MIMO.I. INTRODUCTIONHigh data rate mobile optical wireless communications, has so far been limited to very short transmission ranges of less than

26、10m 3. To achieve transmission ranges greater than a few tens of meters in optical wireless requires highly directional light beams with very narrow angle-of-view 2. Optical wireless channels are characterized by large path loss and high background noise typically from sunlight or other ambient ligh

27、t sources in vicinity 16. Further the low transmit power levels in optical channels (due to output power regulations in optical sources such as LEDs and LASERs) limit the signal-to-noise ratios in these channels and thus the transmission range. In our recent work in 6, we have argued that it is now

28、becoming feasible to achieve high data rates over large transmission ranges in mobile optical wireless communications using camera receivers through a concept what we call ”visual MIMO”. In this concept, optical transmissions by an array of light emitting devices are received by an array of photodet

29、ector (pixels) elements of a camera. The pixels in a camera can essentially be viewed as an array of highly directional receive elements. Such a structure allows allows reducing interference and noise from other light sources in the channel. Such a system offers a degree of freedom in selecting and

30、combining a subset of the receiver elements that receive a strong signal from the transmitter and thus achieve large SNRs. This may be very similar to the antenna selection in RF-MIMO but will incur lesser overhead and non-complex processing at the camera receiver as the processing can be done in so

31、ftware using image processing and computer vision algorithms 6. However, the tradeoffs in the visual MIMO system, are a limited receiver sampling frequency and strong line-of-sight (LOS) requirements. We already showed in 6 that usingvisual MIMO it is possible to achieve considerable data rates over

32、 large transmission ranges with just a single transmitting element. Using MIMO techniques such as ”multiplexing” to send independent streams of bits using the multiple elements of the light transmitter array and recording over a group of camera pixels can further enhance the data rates. On the other

33、 hand the system could send the same information on all the transmit elements of the array and use diversity combining at the camera to achieve large transmission ranges due to the SNR gain. Though the multiplexing and diversity techniques are similar in concept to those in RF MIMO systems 11the vis

34、ual MIMO channel with very different characteristics attributes certain unique behavior to the MIMO gains in these systems.In visual MIMO the perspective distortions in the visual channel dominate over some of the important properties of a RF wireless channel such as distance based attenuation and m

35、ultipath fading. Though perspective distortions in visual channels are primarily distance dependent visual MIMO channels induce perspective distortions in the image even if the transmitter and receiver are aligned at an angle with respect to each other. Two images which are clearly separated in the

36、image plane may look overlapped when viewed from an angle. Such distortions can depreciate the signal quality and the detection capability leading to errors and thus reduction in the data rates. Further lens blur (typically due to focus imperfection or jerks while capturing the image) also can signi

37、ficantly depreciate the image quality and thus reduce the information capacity. In this paper we will detail how MIMO techniques such as multiplexing and diversity are characterized based on the effect of perspective distortions in the visual MIMO channel. Based our channel model we will derive the

38、analytical channel capacity of the visual MIMO channel and using the same we illustrate the significance of parameters such as distance, viewing angle and blur in characterizing multiplexing and diversity in visual MIMO.This paper is structured as follows; in section III we detail the visual MIMO ch

39、annel model followed by the perspective dependent MIMO characterizations in section IV-C. In section V we plot the analytical channel capacity in visual MIMO and follow up with key inferences about the multiplexing and diversity characterization in visual MIMO based on the capacity plots.II. RELATED

40、 WORKPrior work in optical wireless using visible light that use photodiode receivers or imaging receivers are either limited to short ranges or require complex processing at the receiver 17, 21, 22. Though photo diodes can convert pulses at very high rates, they suffer from large interference and b

41、ackground light noise. This results in very low SNRs and thus short communication ranges. We showed analytically in 6, based on the visual MIMO concept, that a camera receiver outperforms photodiode receivers in terms of its channel capacity at medium to long ranges. Recently, a few sporadic project

42、s have begun to investigate cameras as receivers, particularly for inter-vehicle communications 21 and traffic light to vehicle communications 8. Their analytical results show that communication distances of about 100m with a BER 106 are possible. Other work has investigated channel modeling 18 and

43、multiplexing 7. While earlier work has also used cameras to assist in steering of FSO transceivers 25, the visual MIMO approach differs by directly using cameras as receiver to design an adaptive visual MIMO system that uses multiplexing at short distances but still can achieve ranges of hundreds of

44、 meters in a diversity mode.Only a few projects till now have investigated MIMO techniques for optical wireless. For shorter range systems 15, 26 show a MIMO approach for indoor optical wireless communication, 13 studied the capacity of a optical MIMO system and 19 details some work on space-time co

45、des for optical MIMO. Earlier work by Kahn 23 investigates the use of multibeam transmitters and imaging receivers in Infra-Red systems very similar to MIMO in concept. Very recently the PixNet project 20 presents an implementation of an LCD - camera communication system that can deliver high data r

46、ates of the order of Mbps over distances of about 16m and wide view angles. PixNet uses OFDM to transmit between the LCD-camera pair similar to the pixelated - MIMO system proposed by Hranilovic and Kschischang 13. In this paper we will emphasize that regardless of any type of modulation and transmi

47、ssion scheme, visual MIMO can still achieve significantly high data rates by exploiting some of the uniquecharacteristics of the visual channel.III. VISUAL MIMO MODELIn the visual MIMO communications system, the optical transmit element generates a light beam (optical signal) whose output power is p

48、roportional to the electrical input power of the modulating signal, limited by the emitters peak transmission power 14, 18, 22. While RF channels are typically characterized by their impulse response that reflects the multipath environment, this aspect differs significantly for optical channels. Sin

49、ce the rate of change of the channel impulse response is very slow compared to the frequency of the optical signal, it is usually sufficient to use a static parameter (channel DC gain) 16 to represent the channel. For the same reason inter-symbol interference and multipath fading can be neglected in

50、 optical wireless channels. Similarly Doppler shift is negligible compared to the frequency as well. Consider the visual MIMO communication system model as shown in Fig. 1 where an optical transmitter consisting of an array of K transmitting elements communicates to a camera receiver with an array o

51、f I J pixels. The channel model for the visual MIMO system is given as, where Y 2 RIJ is the image current matrix with eachelement representing the received current y(i; j) in each pixel with image coordinates (i; j), xk 2 R represents the transmitted optical power from kth element of the LEA and Hk

52、 2 RIJ is the channel matrix of the kth transmit element of the LEA, with elements hk(i; j) representing the channel between the kth transmit element and pixel (i; j), and N is the noise matrix. Noise in optical wireless is dominated by shot noise from background light sources and typically modeled

53、as AWGN 16, 18. Each element n(i; j) of the noise matrix N representing the noise current at each pixel is given aswhere q is the electron charge, R is the responsitivity of the receiver characterized as the optical power to current conversion factor, Pn is the background shot noise power per unit a

54、rea, s is the square pixel side length and W is the sampling rate of the receiver (equates to the frame rate of the camera).The optical signal from the kth transmit element (k =1; 2; 3 : : :K) emitting a light beam of power Pin;k will be transmitted into the channel. At the receiver, depending on th

55、e focusing of the camera and the distance between the transmitting element and the camera, the transmitting elements image may strike a pixel or a group of pixels of the detector array. The signal current in each pixel will depend on the concentration of the received signal component on that pixel w

56、hich can be quantified as the ratio of the pixel area relative to the area spanned by the transmitting elements image on the detector. If ck(i; j) represents the concentration ratio of the kth transmit element of an LEA on pixel (i; j), the channel DC gain hk(i; j) from each transmit element k to th

57、e pixel(i; j) is given aswhere R is the responsitivity, Ro() is the Lambertian radiation pattern of the optical transmitting element 16 with half-power angle , Alens is the area of the camera lens, is the camera field-of-view (fov) and dk;i;j , k;i;j are the distance & viewing angle between each transmit element kand receiving pixel (i; j) respectively.Typically, since the pixel size is very small (order ofmicrons), the difference in distance dk;i;j and the viewing angle k;i;j between each element of the transmitter array and every pixel is negligible. Therefore we refer to the distance dk