本文選題：煤礦安全　切入點：物聯(lián)網(wǎng)　出處：《中國礦業(yè)大學(xué)》2016年博士論文　論文類型：學(xué)位論文

【摘要】：隨著以無線通信技術(shù)為主要支撐的物聯(lián)網(wǎng)在煤礦井下的廣泛應(yīng)用,通過無線傳輸?shù)母鞣N煤礦安全監(jiān)測和控制數(shù)據(jù)呈現(xiàn)出幾何級增長的趨勢。被用于數(shù)據(jù)傳輸?shù)臒o線通信信道上的電磁干擾則是寬帶無線吞吐量(信道容量)的最大制約因素。特別是在煤礦井下特殊的封閉環(huán)境,具有多徑效應(yīng)的信道干擾將對無線通信系統(tǒng)產(chǎn)生更加嚴重的影響。強烈的同鄰信道干擾將嚴重降低無線通信質(zhì)量并大幅增加安全監(jiān)控數(shù)據(jù)的丟失和傳輸延遲,直接危害煤礦安全生產(chǎn)。因此,研究煤礦井下應(yīng)用背景下最小化全網(wǎng)無線通信干擾的信道分配策略以及時變信道狀態(tài)下為躲避嚴重電磁干擾的智能無線信道切換算法具有十分重要和深遠的意義。本文首先研究通信質(zhì)量平穩(wěn)不隨時間變化(非時變)無線信道狀態(tài)下煤礦物聯(lián)網(wǎng)分布式自組織無線信道的分配問題。在深入分析現(xiàn)有分布式信道分配算法的基礎(chǔ)上結(jié)合煤礦物聯(lián)網(wǎng)無線感知節(jié)點分布特點和網(wǎng)絡(luò)架構(gòu)等特征,創(chuàng)新地提出了滿足無同信道干擾條件下的基于網(wǎng)絡(luò)拓撲最大鏈路邊數(shù)的分布式正交無線信道分配算法CALENT。同時確定了和物理層密切相關(guān)的煤礦物聯(lián)網(wǎng)無線感知節(jié)點的最小發(fā)射功率以及具體的算法實現(xiàn)步驟。通過與現(xiàn)有的無同信道干擾分布式無線信道分配算法Dis-Link的深入分析和對比研究表明,CALENT算法具有更好的同信道干擾指數(shù)性能和更快的算法收斂速度。在深入研究分析現(xiàn)有物理干擾模型和各協(xié)議干擾模型的基礎(chǔ)上,創(chuàng)新地提出了具有數(shù)據(jù)回傳和控制功能的雙向通信模式下無線信道協(xié)議干擾模型:收發(fā)者協(xié)議干擾模型SRPIM,同時提出了適合該通信模式的分布式正交無線信道分配算法CALENT-BE。因具有異步的無線信道分配機制,CALENT-BE算法克服了原CALENT(參考算法CALENT-RB)在擁有雙向數(shù)據(jù)傳輸鏈路的對等通信模式下算法收斂速度緩慢的問題。為提高無線信道的空間復(fù)用率,在消除部分(頻譜)重疊信道的同信道和鄰間信道干擾的基礎(chǔ)上,創(chuàng)新地提出了當前已被廣泛應(yīng)用的Wi-Fi無線通信非正交信道集下的分布式無線信道分配算法CALENT-POC。通過有效的無線信道分配約束機制,避免信道分配過程中實現(xiàn)全網(wǎng)零干擾所需最小信道總數(shù)過度擴散的同時,又保證了信道分配結(jié)果的多樣性,為尋求無線信道資源分配的全局最優(yōu)解提供了充分必要條件。本文其次研究的是在煤礦井下采煤工作面,掘進和機車巷道等特殊生產(chǎn)環(huán)境中通信質(zhì)量隨時間變化(時變)信道狀態(tài)下煤礦物聯(lián)網(wǎng)無線感知智能通信終端動態(tài)選擇電磁干擾最小,通信質(zhì)量最好(通信中斷和重傳次數(shù)最少)的智能無線信道切換問題。在深入研究傳統(tǒng)的單一固定信道通信和Q學(xué)習(xí)信道切換策略的基礎(chǔ)上,創(chuàng)新地提出了基于EWA學(xué)習(xí)的無線信道切換算法。通過引入信道反饋機制,構(gòu)建了多級報酬反饋體系。在深入分析報酬結(jié)構(gòu)的基礎(chǔ)上,細化和優(yōu)化了各報酬成分要素,使不同層次的報酬反饋更加細致和合理。系統(tǒng)能夠通過在線學(xué)習(xí)煤礦井下時變的無線環(huán)境和通信信道特性,積累歷史信道切換的經(jīng)驗,推測、選擇并更改和切換到當前通信質(zhì)量最好的無線信道,實時動態(tài)地保證了通信鏈路的質(zhì)量,降低了系統(tǒng)通信中斷的概率。在最大限度地保證信道最優(yōu)選擇和切換精度的前提下,創(chuàng)新地提出了無線信道切換簡化和快速算法EWAS,用以解決原EWA算法存在的計算復(fù)雜,能耗高等問題。EWAS無線信道快速切換算法成功地避免了指數(shù)運算的存在,降低了計算復(fù)雜度。同時,EWAS無線信道快速切換算法直接對目標函數(shù)進行優(yōu)化和更新,簡化了算法步驟。最后本文作研究工作總結(jié),分析了當前研究的缺點和不足并對未來研究的內(nèi)容和方向進行了進一步的研究展望和規(guī)劃。
[Abstract]:Along with the wireless communication technology as the main support of the Internet is widely used in underground coal mine, coal mine safety monitoring and control through a variety of wireless data transmission showed a geometric growth trend. Be used for electromagnetic interference in wireless communication channel for data transmission on wireless broadband is the maximum throughput (channel capacity) special constraints. That is closed in the special environment of coal mine, the channel interference has multipath effect will have a more serious impact on the wireless communication system. With strong adjacent channel interference will seriously reduce the quality of wireless communications and increase safety monitoring data loss and transmission delay, directly endanger the safety production of coal mine. Therefore, channel allocation strategy of coal mine the application background of the whole network to minimize interference in wireless communication and time-varying channel conditions for intelligent wireless channel to avoid serious electromagnetic interference Is a very important and far-reaching significance. This paper firstly studies the channel switching algorithm stable communication quality does not change with time (time invariant) wireless channel state coal network distributed self distribution organization of wireless channel. Based on the in-depth analysis of the existing distributed channel allocation algorithm based on coal mineral network of wireless sensor nodes and network distribution the characteristics of architecture, creatively put forward to meet the conditions of co channel interference under the maximum number of edges of the network topology link distributed orthogonal wireless channel allocation algorithm based on CALENT. and to determine the minimum transmit power and physical layer is closely related to the coal mine internet wireless sensor nodes and the specific algorithm implementation steps. Through in-depth analysis and comparison research and no existing cochannel interference in distributed wireless channel allocation algorithm of Dis-Link show that CALENT algorithm has better with Channel interference index performance and faster convergence speed. Based on the research of the existing physical interference model and the protocol interference model, proposes the two-way communication mode with data transmission and control function of the wireless channel under the protocol interference model: transceiver protocol interference model SRPIM, and puts forward the mode of communication distributed wireless channel allocation algorithm for orthogonal CALENT-BE. wireless channel allocation mechanism is asynchronous, CALENT-BE algorithm overcomes the original CALENT (reference CALENT-RB algorithm) in a peer-to-peer communication mode with bidirectional data transmission link algorithm under the problem of slow convergence. In order to improve the spatial reuse of wireless channel in the elimination rate, part (spectrum) based channel overlap interference the co channel and adjacent channel between, creatively put forward the current Wi-Fi wireless communication has been widely used in non orthogonal channel In the distributed wireless channel allocation algorithm CALENT-POC. through the wireless channel allocation of effective restraint mechanism, avoid the required minimum number of channel over diffusion to achieve full network zero interference channel allocation process at the same time, but also ensure the diversity results of distribution channel, provides the necessary and sufficient conditions for the global optimal solution for wireless channel resource allocation secondly. This study is in the coal working face, the communication quality of the special production environment in roadway tunneling and locomotive changes with time (time-varying) coal channel state network wireless sensor intelligent communication terminal dynamically selects the minimum electromagnetic interference, the best quality of communication (communication interrupt and retransmission times at least) intelligent wireless channel the handover problem based learning. Channel switching strategy in a single fixed channel communication and in-depth study of traditional Q, puts forward the innovation based on EWA Wireless channel switching algorithm learning. By introducing the channel feedback mechanism, construct a multi-level feedback compensation system. Based on analyzing the compensation structure, refinement and optimization of the compensation components, so that different levels of compensation feedback more detailed and reasonable. The system can learn online through the wireless environment of coal mine and time-varying communication channel the characteristics of the accumulation of historical experience, channel switching, speculation, select and change and switch to the wireless channel the best communication quality, guarantee the quality of real-time communication link, reduces the probability of breaking the system communication. To maximize the premise of optimal channel selection and handoff accuracy, proposes a wireless a simplified and fast channel switching algorithm EWAS, to solve the calculation of the original EWA algorithm is complex, the problem of higher energy consumption of fast handover algorithm in.EWAS wireless channel successfully 閬垮厤浜嗘寚鏁拌繍綆楃殑瀛樺湪,闄嶄綆浜嗚綆楀鏉傚害.鍚屾椂,EWAS鏃犵嚎淇￠亾蹇，

本文編號：1598931