黃慶東,孫 晴,閆喬喬
?
基于CDS的分布式協(xié)作共識頻譜感知方法
黃慶東,孫 晴,閆喬喬
(西安郵電大學(xué)通信與信息工程學(xué)院,信息與通信技術(shù)國家級實驗教學(xué)中心 西安 710121)
針對原有全網(wǎng)絡(luò)分布式協(xié)作共識方法信息交互量大、收斂速度慢、共識收斂結(jié)果不穩(wěn)定的問題,本文在分布式協(xié)作共識頻譜感知方法的基礎(chǔ)上,提出了基于連通支配集的分布式協(xié)作共識頻譜感知方法。該方法通過網(wǎng)絡(luò)連通支配子集進行網(wǎng)絡(luò)頻譜感知信息的收集和共識計算,獲得穩(wěn)定的共識結(jié)果,再將共識結(jié)果分享給網(wǎng)絡(luò)其他非支配集節(jié)點,實現(xiàn)全網(wǎng)絡(luò)快速共識收斂。與原有分布式協(xié)作共識方法相比,降低了網(wǎng)絡(luò)節(jié)點間的信息交互量,并且能快速收斂到穩(wěn)定、精確的共識結(jié)果,仿真結(jié)果驗證了算法的優(yōu)良特性。
認(rèn)知無線電; 連通支配集; 協(xié)作頻譜感知; 共識
認(rèn)知無線電(cognitive radios, CRs)主要用于克服無線通信頻譜資源的短缺問題,使次用戶在不干擾主用戶的情況下獲得對空閑頻譜資源的使用。CRs中譜感知方式可通過非協(xié)作或協(xié)作的方式實施,本文研究的是一群次用戶通過協(xié)作的方式實施頻譜感知的情況。協(xié)作方式可以避免和改善影深和遮掩衰落對檢測結(jié)果的影響,另外空間分布的用戶具有更好的區(qū)域覆蓋性,有助于提高準(zhǔn)確度。
CRs中頻譜感知[1]主要有3種方式:匹配濾波理論最優(yōu),但需要主用戶先驗信息;能量檢測次優(yōu),它簡單易實施,對主用戶沒有太多要求;循環(huán)平穩(wěn)特征檢測方法能夠檢測信噪比(signal to noise ratio, SNR)很低的信號,但它也需要主用戶的先驗知識[2]。文獻[3]提出了分布協(xié)作感知,通過分布的次用戶協(xié)作達到共識并最終判決,特別適應(yīng)網(wǎng)絡(luò)中無線衰落嚴(yán)重的用戶檢測,但是并沒有考慮和利用網(wǎng)絡(luò)拓?fù)滟Y源。隨著無線通信的發(fā)展,CRs技術(shù)和應(yīng)用得到廣泛關(guān)注,文獻[4]研究了時變衰落信道下認(rèn)知無線電技術(shù);文獻[5]利用CRs技術(shù)改善工業(yè)環(huán)境中干擾的影響;文獻[6]研究了CRs協(xié)作信息傳遞時的信息碰撞問題。近年來,網(wǎng)絡(luò)拓?fù)淇芍碚撚糜诟纳凭W(wǎng)絡(luò)性能,得到關(guān)注和研究[7-9]。連通支配集(connected dominating set, CDS)是網(wǎng)絡(luò)拓?fù)涞闹匾矫?。文獻[9]研究了無線ad hoc網(wǎng)絡(luò)高效路由機制,提出了連通支配集的簡單有效計算方法,以及拓?fù)涓淖儠r的動態(tài)更新/重算方法。
本文將文獻[3]與文獻[9]中的方法相結(jié)合,在原算法中考慮了網(wǎng)絡(luò)拓?fù)涞挠欣蛩?,提出基于連通支配集的分布式協(xié)作共識頻譜感知方法。
圖1 能量檢測方塊圖
1) 對于固定連通網(wǎng)絡(luò)模型,網(wǎng)絡(luò)拓?fù)洳话l(fā)生改變的情況
其中,
2) 對于隨機網(wǎng)絡(luò)模型,網(wǎng)絡(luò)拓?fù)浒l(fā)生隨機改變的情況
分布式協(xié)作共識雖然能夠達到一致收斂,但沒有借助拓?fù)浣Y(jié)構(gòu)提升共識速度。實際上可以采用全網(wǎng)絡(luò)節(jié)點分布式協(xié)作,并在連通支配集上快速形成穩(wěn)定共識,然后將結(jié)果分享給其他非支配集節(jié)點來提升網(wǎng)絡(luò)共識狀態(tài)的收斂速度,且最終一致收斂結(jié)果不變。
圖2 網(wǎng)絡(luò)拓?fù)鋱D
連通支配集由連通的支配節(jié)點構(gòu)成,網(wǎng)絡(luò)所有節(jié)點要么在當(dāng)前連通支配集中,要么是連通支配集的鄰居節(jié)點[9]。如圖2所示,12個節(jié)點隨機分布于長寬各為1的正方形區(qū)域,節(jié)點間的邊用連線表示,圖中黑色方塊是連通支配集節(jié)點{6,8,9,10,11,12}。連通支配集經(jīng)過單步可以到達網(wǎng)絡(luò)中的任意節(jié)點,故網(wǎng)絡(luò)信息通過它可以高效傳遞。因此將文獻[3]的方法與連通支配集相結(jié)合,可以借助網(wǎng)絡(luò)拓?fù)浣Y(jié)構(gòu),達到快速達成共識的效果。對于包含鏈路誤差的隨機網(wǎng)絡(luò)模型,文獻[9]提出了相應(yīng)的更新和重算措施。
另:
不同于文獻[3]中的算法,這里各節(jié)點按照連通支配集分享的收斂值進行判決,進而有效地利用拓?fù)滟Y源提升算法性能。連通支配集收斂值作為全網(wǎng)絡(luò)的收斂值分享給周圍非支配集的鄰居節(jié)點,最后根據(jù)式(7)進行判決。
圖3 12個節(jié)點固定網(wǎng)絡(luò)
圖4 20個節(jié)點隨機網(wǎng)絡(luò)
本文研究了無向網(wǎng)絡(luò)中無中心分布式共識策略認(rèn)知無線電頻譜感知方法,對固定網(wǎng)絡(luò)和隨機網(wǎng)絡(luò)中分布式感知收斂性進行了全新的推證,并將原有算法與支配集特性結(jié)合,利用支配集收斂結(jié)果能夠更快獲得收斂值,可將收斂分享給其他非支配集節(jié)點實現(xiàn)各個節(jié)點的判決。支配集與分布式算法結(jié)合使用體現(xiàn)了拓?fù)湮粗姆植际剿惴ㄏ蛲負(fù)淇芍植际接嬎愕倪^渡和邁進,這使得分布式算法與網(wǎng)絡(luò)實際拓?fù)淠軌蚋玫钠ヅ?,算法性能得到更好的?yōu)化。當(dāng)采用支配集與分布式結(jié)合的算法后,網(wǎng)絡(luò)中其他節(jié)點通過支配集節(jié)點連接傳輸網(wǎng)絡(luò)節(jié)點采集的信息,同時節(jié)點可以分享支配集的共識結(jié)果;另外因為支配集與其鄰居節(jié)點覆蓋了整個網(wǎng)絡(luò)所有節(jié)點,所以可以避免信息的冗余傳輸、信道的資源浪費,同時可以節(jié)約能耗、提升信息的傳遞效率和分布式計算的收斂速度。
[1] CABRIC D, MISHRA S, BRODERSEN R. Implementation issues in spectrum sensing for cognitive radios[C]//The 38th Asilomar Conference on Signals, Systems, and Computers. California: IEEE, 2004: 772-776.
[2] SUN C, ZHANG W, LETAIEF K B. Cluster-based cooperative spectrum sensing in cognitive radio systems[C]//The 2007 IEEE International Conference on Communications. Glasgow, Scotland: IEEE, 2007: 2511-2515.
[3] LI Zhi-qiang, RICHARD Y U F, HUANG Min-yi. A distributed consensus-based cooperative spectrum-sensing scheme in cognitive radios[J]. IEEE Transactions on Vehicular Technology, 2010, 59(1): 383-392.
[4] LI B, SUN M, LI X, et al. Energy detection based spectrum sensing for cognitive radios over time-frequency doubly selective fading channels[J]. IEEE Transactions on Signal Processing, 2015, 63(2): 402-417.
[5] CHIWEWE T M, MBUYA C F, HANCKE G P. Using cognitive radio for interference-resistant industrial wireless sensor networks: an overview[J]. IEEE Transactions on Industrial Informatics, 2015, 11(6): 1466-1481.
[6] LUNDéN J, MOTANI M, POOR H V. Distributed algorithms for sharing spectrum sensing information in cognitive radio networks[J]. IEEE Transactions on Wireless Communications, 2015, 14(8): 4667-4678.
[7] BERTRAND A, MOONEN M. Seeing the bigger picture: How nodes can learn their place within a complex ad hoc network topology[J]. IEEE Signal Processing Magazine, 2013, 30(3): 71-82.
[8] BERTRAND A, MOONEN M. Distributed computation of the fiedler vector with application to topology inference in ad hoc networks[J]. Signal Processing, 2013, 93(5): 1106-1117.
[9] JIE W U, HAILAN L I. A dominating-set-based routing scheme in ad hoc wireless networks[J]. Telecommunication Systems, 2001, 18(1-3): 13-36.
[10] URKOWITZ H. Energy detection of unknown deterministic signals[J]. Proceedings of the IEEE, 1967, 55(4): 523-531.
[11] KOSTYLEV V. Energy detection of a signal with random amplitude[C]//The 2002 IEEE International Conference on Communications. New York: IEEE, 2002: 1606-1610.
[12] HUANG M, MANTON J H. Coordination and consensus of networked agents with noisy measurements: Stochastic algorithms and asymptotic behavior[J]. SIAM Journal on Control and Optimization, 2009, 48(1): 134-161.
編 輯 葉 芳
Distributed Consensus Cooperative Spectrum Sensing Method Based on Connected-Dominating-Set
HUANG Qing-dong, SUN Qing, and YAN Qiao-qiao
(Informations and Communications Technology of National Experimental Teaching Center, School of Communication and Information Engineering, Xi’an University of Posts and Telecommunications Xi’an 710121)
Aiming at the problems of large amount of information exchange, slow convergence and unstable convergence result of the original whole-network distributed cooperative consensus method, a distributed consensus cooperative spectrum sensing method based on connected-dominating-set is proposed. In this paper, the connected-dominating-set method collects information of network spectrum sensing and does consensus calculation through the network connected dominating subset, which gets a stable consensus result. Then, the consensus results are shared to other non-dominated nodes in the network, so rapid convergence is achieved. There are two advantages of the proposed method compared with the original network distributed cooperative consensus method. One is that the amount of information exchanged among network nodes reduces, the other is that the network can quickly converge to a stable and accurate consensus result. The proof of convergence theorem of distribution consensus is given in this paper. Lastly, the simulation results show the excellent characteristics of this new algorithm.
cognitive radios; connected dominating set; cooperative spectrum sensing; consensus
TN911.23
A
10.3969/j.issn.1001-0548.2017.05.004
2016-04-12;
2017-06-02
國家重大專項(2017ZX03001012-005)
黃慶東(1976-),男,副教授,博士,主要從事分布式信號處理、陣列信號處理、低復(fù)雜度算法等方面研究.