published_or_final_version / abstract / Electrical and Electronic Engineering / Doctoral / Doctor of Philosophy

published_or_final_version / Electrical and Electronic Engineering / Master / Master of Philosophy

最近，幾種以Kodialam等人提出的RFID數量估計算法為基礎的新算法相繼出現。這些新算法不僅考慮到RFID讀器和RFID標簽之聞無線信道的不穩定性和不確定性，而且還可以保證其估算具有一定的準確性。然而，這些RFID數量估計算法的設計和性能評估都是基於一個比較簡化的信道模型。根據這個信道模型，封包在RFID讀器和標簽之聞傳送時遺失的概率是符合獨立分布的。然而，一些實證測量研究指出，在一般的室内環境，人的活動和設備的移動對信道的影響是不可忽視的。這些活動可以引起多普勒效應，從而使信號變成時域相關。因此，在真實的無線信道傳輸中，封包的遺失也是時域相關的。由於不同的RFID數量估計算法具體的設計和實現方法各有不同，封包遺失的時域相關性可能對這些RFID數量估計算法造成不同程度的影響。 / 在本論文中，我們評估了三系列的算法在更真實的時域相關的無線信道下的性能。具體來，我們重點研究和分析了時域相關性在哪些方面影響了現有的RFID數量估計算法的準確性。依據描述真實無線信道特性的實驗數據，我們改進了原來的信道模型，使其概括了RFID讀器和標簽之聞向前信道/向後信道的時域相關性。通過觀察這三系列算法在更真實的時域相關的無線信道下的表現，我們分析了造成估算值和真實值之間差距的原因。同時，提出了一些改善數量估計算法準確性的方法，並且通過模擬試驗證明這些改進可以提高算法在真實無線信道下的準確性。 / Recently, several new RFID counting algorithms have been proposed based on the probabilistic counting schemes introduced by Kodialam et. al.. These existing algorithms took into account the unreliable and non-deterministic nature of the radio communication channels between the RFID reader and the tags,and are capable of providing tag-count estimates that satisfy a desired level of accuracy. However, all algorithms were designed and evaluated based on a simplistic packet loss model. It assumes that the loss probability of a packet between the reader and the tag-set follows an independent, identical distribution. No characterizations of possible temporal correlations of the channels were performed. As presented by some empirical measurements, movements of personnel or equipments in a building can generate Doppler effect, which introduces time correlations to the fading signal. Thus, the realistic packet loss caused by the wireless channels is temporally correlated due to the frequent change of the nearby environment. Depending on specific implementation details of each individual algorithm,temporally correlated packet loss might have significant impact on the tag-set cardinality estimation. / In this thesis, we will evaluate the performance of the aforementioned RFID counting algorithms under a more sophisticated time-correlated channel fading model. In particular, we focus on investigating and analyzing how the temporal correlations would influence the accuracy of these existing counting algorithms. Based on the experimental statistics that characterized the realistic indoor channels, we refine the channel model to describe the time-correlation of the forward and backward channel between the RFID reader and the tags. We also modify the model to support implementations of different communication scenarios. Comparisons of the performance of the existing counting schemes under the simplistic uncorrelated packet loss channel model and the refined correlated channel model are conducted. We also propose extensions for these RFID counting schemes to mitigate the estimation inaccuracy generated by the correlated packet loss. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Deng, Yulin. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 99-107). / Abstracts also in Chinese. / Abstract --- p.i / Acknowledgement --- p.vi / Chapter 1 --- Introduction --- p.1 / Chapter 2 --- Background and related work --- p.7 / Chapter 3 --- Modeling time-correlated channels in RFID systems --- p.13 / Chapter 3.1 --- Memoryless channel model --- p.14 / Chapter 3.2 --- Backscattering channel model --- p.16 / Chapter 3.2.1 --- Discussions of channel differences with active tags --- p.18 / Chapter 3.3 --- Difficulties in extending mathematical analysis for Memoryless channel model to time-correlated Backscattering channel model --- p.19 / Chapter 3.4 --- Implementations of Backscattering channel model --- p.20 / Chapter 3.4.1 --- Model the one-way fading channel based on the Clarke and Gans model --- p.20 / Chapter 3.4.2 --- Other implementation aspects of the Twoway Backscattering channel model --- p.25 / Chapter 4 --- Performance evaluation of the First-two-moment implicit approach under time-correlated channels --- p.27 / Chapter 4.1 --- System model --- p.28 / Chapter 4.2 --- Estimator Accuracy --- p.30 / Chapter 4.3 --- Performance evaluation of the First-two-moment implicit RFID counting scheme over time-correlated channels --- p.32 / Chapter 4.3.1 --- Modeling differences --- p.34 / Chapter 4.3.2 --- Coherence time --- p.37 / Chapter 4.3.3 --- Rician fading channel --- p.40 / Chapter 4.3.4 --- Discussions for the case with active tags --- p.42 / Chapter 4.4 --- Correcting estimation errors under time-correlated channels --- p.44 / Chapter 4.4.1 --- Correcting estimation mean --- p.44 / Chapter 4.4.2 --- The impact of time-correlated channels on estimation variance --- p.47 / Chapter 4.5 --- Chapter summary --- p.49 / Chapter 5 --- Performance Evaluation of the Capture-Recapture approach under time-correlated channels --- p.50 / Chapter 5.1 --- System model --- p.51 / Chapter 5.2 --- Different estimation algorithms --- p.53 / Chapter 5.2.1 --- Union-based approach --- p.53 / Chapter 5.2.2 --- The Capture-Recapture approach --- p.55 / Chapter 5.3 --- Performance evaluation of the estimation schemes under time- correlated channels --- p.58 / Chapter 5.3.1 --- Simulation Setup --- p.60 / Chapter 5.3.2 --- Observations and Analysis --- p.65 / Chapter 5.4 --- Extensions of correcting the errors under timecorrelated channels --- p.69 / Chapter 5.5 --- Chapter Summary --- p.72 / Chapter 6 --- Performance evaluation of adaptive RFID counting algorithms under time-correlated channels --- p.74 / Chapter 6.1 --- System Model --- p.75 / Chapter 6.2 --- Adaptive RFID counting Algorithms --- p.76 / Chapter 6.2.1 --- Adaptive Union approach --- p.77 / Chapter 6.2.2 --- Probabilistic Estimation For Lossy Channels (PELOC) --- p.78 / Chapter 6.3 --- Performance evaluation of the adaptive counting algorithms under time-correlated channels --- p.80 / Chapter 6.3.1 --- Simulation Setup --- p.81 / Chapter 6.3.2 --- Observation and analysis --- p.82 / Chapter 6.4 --- Extensions to correct the inaccuracy under timecorrelated channels / Chapter 6.5 --- Chapter Summary --- p.93 / Chapter 7 --- Conclusion and Future work --- p.95 / Bibliography --- p.99

Bibliography: p. 235-243. / xxvi, 243 p. : ill. ; 30 cm. / Title page, contents and abstract only. The complete thesis in print form is available from the University Library. / Concerned with the mathematical characterisation and adaptive processing of narrowband high frequency signals received by a very wide aperture antenna array after reflection from the ionosphere. / Thesis (Ph.D.)--University of Adelaide, Dept. of Electrical and Electronic Engineering, 2000

本論文研究了無線通信系統中用戶之間相互干擾的模型. 在傳統的干擾模型中, 對於一個目標用戶, 其他共用信道的用戶的信號被認為是干擾, 且干擾被假設成是一個高斯隨機過程. 我們摒棄這種對干擾的分布的假設, 研究一個更實際的干擾模型. 我們稱之為非高斯假設的干擾模型. 在我們的模型中, 目標用戶和干擾用戶可以使用不同的發射功率和符號速率, 並且他們的信號不要求同步. 我們推導了非高斯假設的干擾模型下, 二進制移相鍵控 (BPSK) 的匹配濾波最佳接收的平均誤比特概率表達式. 並且我們評定了傳統的高斯干擾模型的有效性. / 對於非高斯干擾模型的研究, 我們先從時不變信道入手. 利用誤比特率作為系統性能的指標, 我們探討了兩類功率控制問題. 研究的結果顯示高斯干擾模型和非高斯假設的干擾模型有一些本質上的區別. 第一類功率控制問題是最小化所有用戶中的最大誤比特率. 我們發現在非高斯假設的干擾模型中, 當某些條件符合時, 最優化的誤比特率可以為零, 而在高斯干擾模型下, 最優化的誤比特率在任何條件下都不可能達到零. 同時, 我們發現在非高斯假設的干擾模型中, 在某些情況下, 有限的功率就能實現誤比特率的優化. 但在高斯干擾模型中, 無論什麼情況, 要實現最優的誤比特率就要使用無限的功率. 第二類功率控制問題是最小化所有用戶的發射功率總和, 且每個用戶滿足給定的誤比特率要求. 我們探究了非高斯假設的干擾模型下, 誤比特率函數的性能, 並且提出了尋找最優解的迭代算法. 通過仿真比較兩種干擾模型對功率控制的影響, 我們發現高斯干擾模型比非高斯假設的干擾模型要求更大的功率去實現相同的誤比特率要求. / 接著, 我們研究衰落信道下的非高斯假設的干擾模型 . 分析的重點集中在兩個用戶構成的系統: 一個目標用戶和一個干擾用戶. 我們分別探究了在瑞利(Rayleigh), Nakagami 和萊斯 (Rician) 衰落信道下的誤比特率性能. 首先我們從理論上分析了誤比特率隨著衰落嚴重程度的變化趨勢. 然後我們利用數值分析全面比較了高斯干擾模型和非高斯假設的干擾模型在衰落信道下的表現. 仿真結果表明, 高斯干擾模型准確預測誤比特率的能力非常有限, 它不能有效地反映誤比特率對應於信號噪聲功率比 (SNR), 信號干擾功率比 (SIR), 符號速率和衰落嚴重程度的變化. / This thesis studies the interference model of a wireless communication system. In the traditional Gaussian interference model (GIM), for a desired user, the combined interference from other simultaneous users is assumed to be a Gaussian process. We dispense with this Gaussian assumption on the interference and study a more realistic interference model. We call it the non-Gaussian interference model (NGIM). Our model allows for different transmission powers, symbol rates and symbol timing asynchronism between the desired user and interfering users. We derive precise expressions for the average bit error probability (BEP) of binary phase shift keying (BPSK) under the NGIM and access the validity of the GIM. / We start the study by first focusing on the NGIM for non-fading channels. We use the BEP as utility metric to investigate two types of power control problems under the new model and our work demonstrates some qualitative differences between the GIM and NGIM. The first power control problem is to minimize the maximal BEP of all users. We find that in the NGIM, the minimum of the maximal BEP of all users can be zero under certain conditions while in the GIM, the optimal value can never be zero. We also find that under the NGIM, in some cases, the optimal value is achieved by infinite power while under the GIM, the optimal value is always achieved by in&#xc9;IÆpg℗ / We then extend the study to the NGIM for fading channels. We analyze the BEP performance of a two-user system over the Rayleigh, Nakagami and Rician fading channels, respectively. We provide some theoretical results concerning the BEP behavior with respect to the fading severity under the NGIM. Comprehensive numerical study and comparison of the BEP performance between the GIM and the NGIM are also presented. The results show that the GIM has limitation in predicting the exact BEP performance in fading channels. It fails in accurately tracking the variation of the BEP with respect to the signal-to-noise ratio (SNR), signal-to-interference ratio (SIR), symbol rate and fading severity. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Chen, Yi. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 98-104). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese. / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Motivation --- p.1 / Chapter 1.2 --- Overview --- p.4 / Chapter 1.3 --- Outline --- p.7 / Chapter 2 --- Power Control for NGIM --- p.9 / Chapter 2.1 --- Introduction --- p.9 / Chapter 2.2 --- System Model and Error Probability Calculation --- p.11 / Chapter 2.2.1 --- BEP in the NGIM --- p.13 / Chapter 2.2.2 --- BEP in the GIM --- p.15 / Chapter 2.3 --- The Minimal BEP Problem --- p.16 / Chapter 2.4 --- The Minimal BEP for non-M-Matrix Character Matrix --- p.21 / Chapter 2.5 --- The Minimal Power Problem --- p.26 / Chapter 2.6 --- Simulation Results --- p.31 / Chapter 3 --- BEP of NGIM in fading channels --- p.35 / Chapter 3.1 --- Introduction --- p.36 / Chapter 3.2 --- System Model --- p.39 / Chapter 3.3 --- The moments of ξ --- p.43 / Chapter 3.3.1 --- T[subscript i] ≥ T[subscript j] --- p.43 / Chapter 3.3.2 --- T[subscript i] < T[subscript j] --- p.44 / Chapter 3.4 --- BEP under fading channels --- p.45 / Chapter 3.4.1 --- Rayleigh --- p.46 / Chapter 3.4.2 --- Nakagami --- p.48 / Chapter 3.4.3 --- Rician --- p.50 / Chapter 3.5 --- Numerical Results --- p.51 / Chapter 3.6 --- Discussion of multiple interferers --- p.55 / Chapter 4 --- Probability of M-matrix --- p.65 / Chapter 4.1 --- System model --- p.65 / Chapter 4.2 --- BEP Floor --- p.66 / Chapter 4.3 --- Probability of M-matrix in Rayleigh fading channels --- p.68 / Chapter 4.4 --- Discussion --- p.71 / Chapter 5 --- Summary --- p.75 / Appendices --- p.77 / Chapter A --- NGIM in non-fading channels --- p.77 / Chapter A.1 --- Derivation of the variance of W[subscript i subscript j] --- p.77 / Chapter A.2 --- Proof of Lemma 2.9 --- p.78 / Chapter A.3 --- Proof of Lemma 2.10 --- p.78 / Chapter A.4 --- Proof of Lemma 2.11 --- p.79 / Chapter A.5 --- Proof of Lemma 2.14 --- p.80 / Chapter A.6 --- Proof of Lemma 2.15 --- p.80 / Chapter A.7 --- Proof of Lemma 2.16 --- p.81 / Chapter B --- NGIM in fading channels --- p.84 / Chapter B.1 --- Calculation of the moments of ξ --- p.84 / Chapter B.2 --- BEP in Rayleigh fading --- p.86 / Chapter B.3 --- Lower bound of BEP difference in Rayleigh fading --- p.87 / Chapter B.4 --- BEP in Nakagami fading --- p.88 / Chapter B.5 --- Proof of Theorem 3.2 --- p.89 / Chapter B.6 --- Proof of Theorem 3.3 --- p.91 / Chapter B.7 --- BEP in Rician fading --- p.93 / Chapter B.8 --- Proof of Theorem 3.4 --- p.94 / Bibliography --- p.98

In this thesis we apply chaotic dynamic data analysis to the area of discrete time signal processing. A newly developed Hidden Filter Hidden Markov Model is introduced in detection of chaotic signals. Numerical experiments have verified that this novel nonlinear model outperforms linear AR model in detecting chaotic signals buried by noise having similar power spectra. A simple Histogram Model is proposed which can also be used to do detection on the data sets with chaotic behavior. Receiver Operating Characteristics for a variety of noise levels and model classes are reported.

published_or_final_version / Electrical and Electronic Engineering / Master / Master of Philosophy

"January 2003" Includes bibliographical references (leaves 120-129) Presents performance analyses of dynamic, stochastic task scheduling policies for a real- time-communications system where tasks lose value as they are delayed in the system.

"January 2003" / Includes bibliographical references (leaves 120-129) / x, 129 leaves : ill. ; 30 cm. / Title page, contents and abstract only. The complete thesis in print form is available from the University Library. / Presents performance analyses of dynamic, stochastic task scheduling policies for a real- time-communications system where tasks lose value as they are delayed in the system. / Thesis (Ph.D.)--University of Adelaide, Dept. of Electrical and Electronic Engineering, 2003

Exames de Comparação Forense de Locutores apresentam características complexas, demandando análises demoradas quando realizadas manualmente. Propõe-se um método para reconhecimento automático de vogais com extração de características para análises acústicas, objetivando-se contribuir com uma ferramenta de apoio nesses exames. A proposta baseia-se na medição dos formantes através de LPC (Linear Predictive Coding), seletivamente por detecção da frequência fundamental, taxa de passagem por zero, largura de banda e continuidade, sendo o agrupamento das amostras realizado por meio do método k-means. Experimentos realizados com amostras de três diferentes bases de dados trouxeram resultados promissores, com localização das regiões correspondentes a cinco das vogais do Português Brasileiro, propiciando a visualização do comportamento do trato vocal de um falante, assim como detecção de trechos correspondentes as vogais-alvo. / Forensic speaker comparison exams have complex characteristics, demanding a long time for manual analysis. A method for automatic recognition of vowels, providing feature extraction for acoustic analysis is proposed, aiming to contribute as a support tool in these exams. The proposal is based in formant measurements by LPC (Linear Predictive Coding), selectively by fundamental frequency detection, zero crossing rate, bandwidth and continuity, with the clustering being done by the k-means method. Experiments using samples from three different databases have shown promising results, in which the regions corresponding to five of the Brasilian Portuguese vowels were successfully located, providing visualization of a speaker’s vocal tract behavior, as well as the detection of segments corresponding to target vowels.