Research Group tc at Electrical and Information Engineering, Sydney University

A Belief Propagation Based Smart Grid Communication Infrastructure for Load Prediction of Dynamic Power Market

Speakers	Masud Rana
Seminar Date	Thursday, 23 May, 2013
Location	Room 712 EE Building
Abstract	The huge amount of electrical energy cannot be stored. So, a smart load prediction control system is needed to ensure electric generation closely matches the demand. We propose a belief propagation (BP) based smart grid recursive systematic convolutional (RSC) coded communication infrastructure for load prediction of dynamic power market. The observation information is transmitted to a load prediction program for taking action to the next time power generation. The simulation results show that the BP based load prediction algorithm achieves significant performance gain compared with the traditional Kalman filter (KF) based technique in terms of the mean squared error (MSE) and bit error rate (BER).

A Novel Hybrid CDL Based Multipath Propagation Model for the High-Speed Railway at 2.35GHz

Speakers	Jiahui Qiu
Seminar Date	Thursday, 16 May, 2013
Location	Room 712 EE Building
Abstract	Based on the actual channel measurement on the High-Speed Railway (HSR) in the viaduct scenario at 2.35GHz, a hybrid channel modeling technology combining the statistical model and theoretical model is proposed. This novel channel-characterization approach is used to reveal the statistical properties for “irregular” scattering, and to provide a geometric mechanism for computing the time-variant delay tap for “regular” scattering. By using a modified clustering algorithm in the delay domain, inter-cluster and intra-cluster characteristics against the distance , between the transmitter and the receiver are extracted and analyzed. Additionally, the measurement results are compared with some conclusions from other measurement campaigns, and the proposed hybrid Cluster Delay Line (CDL) based channel model is validated by comparing the simulated results with the measurement data.

Transceiver Design for Multi-Pair Two-Way Relay Networks with Analogue Network Coding

Speakers	Miao Wang
Seminar Date	Thursday, 9 May, 2013
Location	Room 712 EE Building
Abstract	We consider multi-pair two-way relay networks where multiple user pairs simultaneously exchange information via a common multi-antenna relay station. The relay precoding matrix, power allocation and user-pair selection are jointly optimized to maximize the system sum rate. Numerical results show that the network sum rate can be significantly increased after joint optimization.

How to Construct Multiple Unique Quantization Codebooks for MU-MIMO

Speakers	Malcolm Egan
Seminar Date	Thursday, 11 April, 2013
Location	Room 712 EE Building
Abstract	A key component of multiuser MIMO using zero-forcing precoding is the feedback of quantized channel state information to the base station. A problem arises when each user has a common codebook as the quantized channels can form a singular matrix, which results in a reduced sum-rate. In this seminar, we propose two new structured constructions to generate different codebooks at each user via transformations of the base codebook. The first construction is based on the Householder transform, which is used to construct a different codebook at each user with no additional storage in addition to the base codebook. Although our first construction is applicable to many common base codebooks, it cannot be used for the Fourier base codebook. To overcome this problem, we propose a second construction that is based on the unitary representation theory of groups. We show that both our constructions significantly reduce storage requirements compared with the intuitive but impractical random construction, while obtaining the same sum-rate performance.

Belief Propagation Method in Intercell Interference Management

Speakers	Nur Ilyana, Anwar Apandi
Seminar Date	Thursday, 28 March, 2013
Location	Room 712 EE Building
Abstract	Due to increased cell density, downlink Inter-cell interference (ICI) between users in adjacent cells and its management are major challenges in the future wireless network. We address the downlink ICI problem by controlling the base station transmit power for cells. We first formulate the optimization problem that maximizes the capacity with transmit power as variables, subject to minimum signal to interference plus noise ratio (SINR) requirement for each user transmission. A factor graph representation is used to decompose the optimization problem into multiple single cell optimizations and to develop a distributed ICI management technique by using a Belief Propagation (BP) technique where the belief on the base station transmit power and the value of optimization function are exchanged between base stations. In this presentation, we present our preliminary work where we develop a factor graph representation for a simple three adjacent cells scenario and the BP methods for ICI management. The work on numerical simulation is still under progress.

Adaptive Analog Fountain for Wireless Channels

Speakers	Mahyar Shirvanimoghaddam
Seminar Date	Thursday, 21 March, 2013
Location	Room 712 EE Building
Abstract	We propose an analog rateless code to achieve high spectral-efficient adaptive transmission and increase the system throughput in AWGN channels. In the proposed analog rateless coding scheme, each coded symbol is generated from a number of information bits that are selected uniformly at random and multiplied by some real values obtained randomly from a predetermined probability distribution function, called weight distribution. The analog rateless codes can be described by a weighted bipartite graph. However, unlike the conventional bipartite graph, where the combining coefficients are the binary symbols, the combining coefficients in the weighted bipartite graph of analog rateless codes are real numbers selected from a finite set. As a result, the conventional sum-product decoder cannot be directly applied. We have developed a simple decoding algorithm, called 2-Sum verification decoder, for the proposed analog rateless codes. Its performance is evaluated by using Sum-Or tree analysis. The code degree and weight distributions are optimized to maximize the error recovery probability of the 2- Sum verification decoder. Simulation results shows the proposed code can approach the channel capacity within one bit across a wide range of SNRs.

Triangular MIMO Relay Channels: Simultaneous Signal and Interference Alignment

Speakers	Keov Kolyan Teav
Seminar Date	Thursday, 22 November, 2012
Location	Room 712 EE Building. J03
Abstract	In this paper, we propose a new network information flow, called triangular multiple-input multiple-output (MIMO) relay (TMR) channel, which consists of three users and three relays equipped with nU and nR antennas, respectively. For this channel, each user wants to send two independent messages to the other two users via the selected relay based on location criteria in two time slots, the multiple access (MA) stage and the broadcast (BC) stage. The optimal transmit beamforming is proposed based on the pseudo inverse problem resulting from the simultaneous signal and interference alignment. An efficient gradient projection algorithm is designed to optimize the pseudo inverse solution which is used to determine beamforming vectors by maximizing the weighted sum rate as the objective function. Moreover, we apply a quasi-Newton method called BFGS, by approximating the Hessian matrix of the pure Newton's method in order to ensure the global convergence, obtain the rapid convergence speed and reduce the computational complexity. Finally, numerical results are presented to validate the performance of our proposed system model.

Reliability of lossy wireless networks with Network Coding

Speakers	li ma
Seminar Date	Thursday, 15 November, 2012
Location	Room 712 EE Building. J03
Abstract	Wireless communications between devices can be lossy owing to a number of issues, such as channel fading, interference or mobility of devices. In some scenarios, such unreliability of wireless communications can be random hence better characterized from a stochastic perspective. In view of this, lossy wireless networks have been studied recently, where the transmission between each pair of nodes is successful with a certain probability. Network coding can be used to improve the reliability of wireless communications in these networks. In this work, a neighbor network coding scheme is proposed and network reliability using the coding scheme is investigated analytically, where the reliability is measured by the probability that every node in the network receives the packets from every other node. It is shown that reliability of networks can be improved using the proposed neighbor coding scheme. Further, closed-form upper and lower bounds on the network reliability are presented. Moreover, an optimal neighbor coding scheme that maximizes the probability that the packet broadcast from a designated source node can be received by all nodes in the network is discussed.

Instantaneous Load Pricing Based Demand Side Management Framework for Future Smart Grid: A Variational Inequality Approach

Speakers	He Chen (Henry)
Seminar Date	Thursday, 8 November, 2012
Location	Room 712 EE Building. J03
Abstract	In this paper, we investigate a new polynomial pricing function based demand side management scheme where the consumers are charged based on their instantaneous load. A non-cooperative game is formulated, where each consumer aims to minimize their total cost based on this pricing function. By casting this game in the variational inequality framework, we present sufficient conditions for the uniqueness of the optimal solution. We then propose a distributed and simultaneous algorithm to achieve the optimal solution, and prove sufficient conditions for the geometrical convergence of our algorithm. Our numerical results reveal that our proposed algorithm converges very quickly, and is effective in encouraging consumers to shift their energy usage from peak to non-peak times.

Uncoordinated cooperative communications via multi hops

Speakers	Xiaofeng Lang
Seminar Date	Thursday, 1 November, 2012
Location	Room 712 EE Building. J03
Abstract	Cooperative communication techniques offer significant performance benefits over traditional methods that do not exploit the broadcast nature of wireless transmissions. Such techniques generally require advance coordination among the participating nodes to discover available neighbors and negotiate the cooperation strategy. However, the associated discovery and negotiation overheads may negate much of the cooperation benefit in mobile networks with highly dynamic or unstable topologies (e.g. vehicular networks). This presentation discusses uncoordinated cooperation strategies, where each node overhearing a packet decides independently whether to retransmit it, without any coordination with the transmitter, intended receiver, or other neighbors in the vicinity. Existing works are limited to one-hop case; multi-hop scenario, however, has not been studied. The aim of my project is to find the optimal retransmission strategy to fulfil the pre-determined overall success probability.

Reliability-Constrained Broadcasting over Unreliable Wireless Links with Network Coding

Speakers	Peng Wang
Seminar Date	Thursday, 1 November, 2012
Location	Room 712 EE Building. J03
Abstract	One major challenge for broadcast traffic in wireless networks is that wireless links are usually unreliable, and their qualities differ from node to node. One common solution to guarantee reliable deliveries over unreliable links is Automatic Repeat reQuest (ARQ), where nodes provide feedback information to the base station after each transmission using either acknowledgements (ACKs) or negative acknowledgements (NACKs). However, as the overhead of gathering feedback information increases with the size of the network, using ARQ for wireless broadcasting can incur significant delay and is not scalable. Thus, a solution for wireless broadcasting that does not require feedback information is needed. Recent work has shown that the capacity of wireless broadcasting over unreliable links can be increased by employing network coding. In this work, an optimal coding scheme is planned to be designed to minimum the number of broadcast the system required to meet the system reliability requirement.

Kalman Filtering Based Dynamic State Estimation for Smart Grid

Speakers	Md Masud Rana
Seminar Date	Thursday, 25 October, 2012
Location	Room 712 EE Building. J03
Abstract	A communications infrastructure in a smart grid will link together the meters, utility mobile workforce, sensors, built into every element of the grid and control centers, into a single integrated widely distributed automated energy delivery network. In order to design such smart control and monitoring centre, control centre will need to know node voltage, branch current and power follow of a smart grid. State estimation helps to calculate the states (voltage, current and power flow) as well as estimated values of the measurements in presence of error. In this work, the communication for the control of distributed energy generations (DEGs) in SG is discussed. We used Kalman filter for linear dynamic state estimation for smart grid. A practical dynamic system for the DEGs is use for numerical simulation.

Novel Nested Convolutional Lattice Codes for Multi-Way Relaying Systems over Fading Channels

Speakers	yuanye ma
Seminar Date	Thursday, 18 October, 2012
Location	Room 712 EE Building. J03
Abstract	In this presentation, we focus on the realization of multiple interpretations (MI) in multi-way relay channels (MWRC) with fading, where multiple sources communicate with each other with the help of a relay. We first propose a novel nested convolutional lattice codes (NCLC) over the finite field, which can achieve the MI for each source in two time slots. Then we derive a theoretical upper bound for the codeword error rate (WER) of the NCLC. We further optimize our NCLC by developing a code design criterion which minimizes the derived WER. In simulations, we construct a specific NCLC based on our code design criterion. Simulation results show that our code can realize MI for each source in two time slots, and validate the derived upper bound in the high normalized signal-to-effective-noise ratio region.

Traffic Modeling for SG communications Networks

Speakers	Obada Al Khatib
Seminar Date	Thursday, 20 September, 2012
Location	Room 712 EE Building. J03
Abstract	A dramatic increase of interest in the research and deployment of Smart Grid in both academic organizations and industrial fields has been witnessed in the last few years. In order to realize the Smart Grid vision, it is necessary to develop a traffic model for the communication technology that will be used in Smart Grid access networks. Modeling will provide a basis for predicting performance of the network before costly equipment procurement and deployment takes place. Utilities can build their own private network or they can use a public network for their SG communications needs. Two traffic models for these two options via a queuing theory approach that is primarily based on a multi-server, multi-priority, non-preemptive queue will be presented .We assume there are two classes of traffic in Smart Grid access networks, namely fixed scheduling traffic and event-driven traffic. Formulae are developed for evaluation of some performance metrics such as mean waiting time, and mean queue size. The analytical model is validated simulations.

Energy-Efficient Wireless Communications-Interference Free and Limited Scenarios

Speakers	Guowang Miao
Seminar Date	Thursday, 13 September, 2012
Location	Room 712 EE Building. J03
Abstract	The future success of communication networks hinges on the ability to overcome the mismatch between requested quality of service (QoS) and limited network resources. Energy efficiency (EE) is becoming increasingly important as battery technology and energy industry has not kept up with the growing requirements stemming from ubiquitous mobile multimedia applications. This presentation introduces cross-layer technologies to improve energy efficiency from different perspectives of wireless systems. We will first discuss technology trend and the methodologies needed to enable highly energy-efficient wireless networks. Then we will introduce state-of-art cross-layer wireless communication technologies that enable high energy efficiency for both individual users (interference free) and multi-user (interference limited) networks. The presentation will be concluded by the discussion of the trade off between spectral and energy efficiency in both interference-free and -limited wireless networks.

A Necessary Condition for Connected Wireless CSMA Multi-hop Networks

Speakers	Tao yang
Seminar Date	Thursday, 6 September, 2012
Location	Room 712 EE Building. J03
Abstract	Connectivity is one of the most fundamental properties of wireless multi-hop networks. In a wireless network with many concurrent transmissions, signals transmitted at the same time may mutually interfere with each other. In this work we consider the impact of interference on the connectivity of CSMA networks using the SINR model. On the basis of our earlier work in which we give a sufficient condition, i.e. an upper bound, on the critical transmission power required for CSMA networks with n nodes i.i.d. on a square of area n to be a.a.s. connected, in this work we study the necessary condition for CSMA networks to be a.a.s. connected, i.e., a lower bound is obtained on the critical transmission power required for CSMA networks to be a.a.s. connected under any scheduling scheme satisfying the carrier-sensing constraint. The lower bound differs from the upper bound by a constant factor only. Compared with previous literature assuming a unit disk model, it is shown that the critical transmission power for a CSMA network under the SINR model to be a.a.s. connected is within a constant factor of that required for a network under the unit disk model, which does not consider the impact of interference, to be a.a.s. connected. That is, transmission power only needs to be increased by a constant factor to combat interference and maintain connectivity. This result is also in sharp contrast with previous results considering the connectivity of ALOHA networks under the SINR model.

Distributed Rateless coding: Partially Coded Cooperation

Speakers	Mahyar Shirvanimoghaddam
Seminar Date	Thursday, 30 August, 2012
Location	Room 712 EE Building. J03
Abstract	In this talk, we present a new rateless coded cooperation (CC) scheme for the two-user cooperative multiple access channel (CMAC), where two users cooperatively communicate with a common destination. We consider two rateless CC strategies, a fully coded cooperation (FCC) scheme used in the conventional rateless cooperative schemes and a new partially coded cooperation (PCC) scheme. In FCC, each user starts coded cooperation process only after the whole block of the other user’s information symbols are fully recovered. In contrast, in PCC, each user starts cooperation as soon as it receives a fraction of new message sent from the other user. The degree distribution for the PCC scheme is designed to maximize the overall system throughput. Simulation results show that the proposed PCC scheme achieves a considerably higher throughput than the conventional scheme in various scenarios.

Coordinated Multipoint Scheduler Design for Base Stations with Finite Buffer Queues

Speakers	Malcolm Egan
Seminar Date	Thursday, 16 August, 2012
Location	Room 712 EE Building. J03
Abstract	Abstract: Coordinated multipoint (CoMP) is a technology standardized in LTE for cell-edge user interference mitigation. One form of CoMP is coordinated scheduling, where base stations communicate via backhaul links to jointly schedule the base station to transmit. In this seminar, we propose a fixed weight signal-to-noise ratio (SNR) CoMP scheduling policy. We derive a new bound for the packet loss probability (PLP) due to buffer overflow of each user with Poisson packet arrivals. We then design the weights and maximum transmission time for our scheduling policy to ensure that PLP targets are met. Finally, we compare our fixed weight policy with an adaptive weight scheduling policy. Surprisingly, we show that our fixed weight scheduling policy can achieve a comparable PLP to the adaptive weight scheduling policy, with reduced backhaul overhead.

Delay-Doppler channel estimation in almost linear complexity

Speakers	Alexander Fish
Seminar Date	Thursday, 2 August, 2012
Location	Room 712 EE Building. J03
Abstract	A fundamental task in wireless communication is channel estimation: Compute the channel parameters a signal undergoes while travelling from a transmitter to a receiver. In the case of delay-Doppler channel, a widely used method is the matched filter algorithm. It uses a pseudo-random waveform of length N; and, in case of non-trivial relative velocity between transmitter and receiver, its arithmetic complexity is O(N^2 log(N)). In this lecture we introduce a novel approach of designing waveforms that allow much faster channel estimation. Using group representation techniques we construct waveforms, which enable us to introduce a new algorithm, called the flag method, that significantly improves the matched filter algorithm. The flag method finds the channel parameters in O(mN log(N)) operations (In certain applications N >> 1000), for channel of sparsity m. We discuss applications of the flag method to mobile communication of fast moving users, and GPS.

Random Network Coding for Wireless Multimedia Delivery

Speakers	Dejan Vukobratovic
Seminar Date	Thursday, 19 July, 2012
Location	Room 712 EE Building. J03
Abstract	In this talk, we discuss the design of Random Network coding (RNC) schemes for multimedia delivery over wireless networks. We provide a construction method and performance analysis of a class of Unequal Error Protection (UEP) RNC suitable for matching with layered multimedia sources. As a case study, we discuss a potential impact of RNC integration within the MAC layer of the LTE Radio Access Network (RAN).

A network coding approach to improve connectivity

Speakers	Li Ma
Seminar Date	Thursday, 28 June, 2012
Location	Room 712 EE Building
Abstract	This work considers a wireless network where the existence of wireless links between nodes is determined by a given probability connectivity matrix. Each node holds a distinct packet to broadcast to every other node in the network. The probability of all nodes receiving packets from every other node is considered under the scenarios where either network coding is applied or not. The gain in the aforementioned probability that introduced by network coding and its relationship with the probabilistic connectivity matrix will be investigated. This study shed insights on network configuration, such as network coding scheme, maximizes the probability of successfully receiving packets.

A General Price Function based Real-time Pricing Framework for Residential Demand Side Management in Future Smart Grid

Speakers	He Chen
Seminar Date	Thursday, 28 June, 2012
Location	Room 712 EE Building
Abstract	Recently, demand side management (DSM) has been regarded as one of the key techniques to transform today’s aging power grid into a more efficient and more reliable operated smart grid. Among different approaches for the implementation of DSM, real-time pricing is one of the most commonly used tools that can convince the users to consume more wisely and efficiently. In most of current works related to residential DSM via real-time pricing, a cost function based framework was adopted, where an increasing and convex cost function was defined to model the electricity cost. Although the definition of the cost function in current literature has several advantages, it may meet several problems when it is implemented in practical power grid. To overcome the problems existing in the current energy cost model, we build a general price function based framework, where the real-time electricity price (i.e., the cost of one unit electricity) is defined as a function of the total demand during the considered time interval. We firstly formulate the problem in a centralized manner and analyze its convexity. Then, game theory is adopted to make a decentralized design. In addition, the existence and uniqueness analysis for the Nash equilibrium (NE) of the formulated game is also presented.

1-D density estimation in vehicular networks using hop count statistics

Speakers	Ruixue Mao
Seminar Date	Thursday, 21 June, 2012
Location	Room 712 EE Building
Abstract	Density estimation is a basic question and intermediate step to many practical problems, such as redeployment, management of large scale sensor networks, real-time routing algorithm adapting. Vehicle density is one of the main metrics used for assessing road traffic condition. 1-D density estimation in vehicular networks can be used to help setting up traffic control system to estimate vehicles’ speed, density of certain sections of highway. In this research work, we investigate an ideal scenario where the positions of the vehicles are randomly and Poisson distributed with the average density ρ veh/m. Our goal is estimating the average density ρ veh/m using the Maximum Likelihood Estimation (MLE) based on observing numbers of neighboring vehicles in different hop count orders within the vicinity of arbitrary chosen probe vehicle. And evaluate the accuracy of density estimate using expected value, variance and entropy of estimated ρ.

Network Distributed Network-Channel Coding for Multi-Hop Networks with Multiple Sources and Destinations

Speakers	Jing Yue
Seminar Date	Thursday, 14 June, 2012
Location	Room 712 EE Building
Abstract	In this research work, we investigate the performance of a multi-access relay interference channel (MARIC) model, in which multiple groups of source nodes communicate with their respective destination nodes (MSMD) via a shared wireless multi-hop relay network. A multiple destinations with code nulling (MDCN) process is proposed to remove the intergroup interference at each destination. However, the process simultaneously creates some unstable zero elements and extra noise, which degrade the formed code’s bit error rate (BER) performance. Based on the analysis of the MDCN process, we investigate the relationship among the equivalent received signal vector, the number of source groups and the column weight of the generator matrix. A code design criteria for the generator matrix construction method is given. We further evaluate the proposed MDCN scheme in terms of throughput and complexity. Our evaluation demonstrates that the proposed scheme not only has a linear time complexity, but also show a considerable throughput improvement compared with a multi-access relay channel (MARC) scheme, where a single group of source nodes communicates with a single destination.

The Triangle MIMO Relay Channels: Signal Space Alignment and Beamforming Optimization

Speakers	Keov Kolyan Teav
Seminar Date	Thursday, 7 June, 2012
Location	Room 712 EE Building
Abstract	In this research work, we propose a new network information flow for a multiple-input multiple-output (MIMO) two-way relay channels with three users and three intermediate relays equipped with nU and nR antennas, respectively. We called this proposed model as a “Triangle MIMO relay channel (TMRC).” For this channel, each user wants to send two independent messages to the other two users via the selected relay based on distance criteria in two time slots, the multiple access (MA) stage and the broadcast (BC) stage. The iterative beamforming optimization algorithm based on orthogonal projection with signal subspace alignment and power allocation is applied in order to maximize the effective signal-to-noise ratios (SNRs). Moreover, a cooperative network coding protocol named denoise-demodulate-and-forward (DDF) is applied in the purpose of reducing the computational complexity at the relay and improving the system throughput. Performances evaluation showed that the proposed TMRC with beamforming optimization produce significant improvement over the existing MIMO Y channels.

Self-adaptive Backoff Algorithm in Smart Grid Home Area Network

Speakers	Qiong Zhao
Seminar Date	Thursday, 7 June, 2012
Location	Room 712 EE Buiding
Abstract	In the Home Area Networks (HAN) of Smart Grid communication system, the backoff algorithm employed in ZigBee is unable to meet the HAN requirements, due to its incapability in terms of contention resolution, transmission reliability and efficiency in high load networks. In this paper, we propose a self-adaptive backoff algorithm, which derives the analytical expression of contention window (CW) without assuming that all nodes always have packet to transmit. We first derive an probabilistic analytical expression of the optimal channel access probability of each node based on the criteria of maximizing throughput and minimizing collision. The optimal channel access probability is then used to derive the expression of CW, which is dynamically adjusted by each node and adapted to real-time measurement of channel congestion condition. Simulation results show the proposed algorithm significantly reduce collision and energy consumption while achieving high throughput and low latency. Since the proposed self-adaptive backoff algorithm is based on CSMA/CA, it can be widely employed in any communication protocol running on top of CSMA/CA, such as, ZigBee and WiFi.

Distributed Rateless Coding with Cooperative Sources

Speakers	Mahyar Shirvanimoghaddam
Seminar Date	Thursday, 31 May, 2012
Abstract	We propose a distributed rateless coding (DRC) scheme for a two-user cooperative system. In DRC, the overall transmission is divided into two phases, a broadcast phase and a cooperation phase. In the broadcast phase, each user keeps transmitting its rateless coded symbols to the other user and the destination until its message has been successfully decoded by the destination or the other user. In the cooperation phase, each user encodes both users’ messages by using a rateless code and transmits them to the destination. A linear programming optimization problem is then formulated to find the optimal degree distribution for the proposed distributed rateless code. The performance of the proposed code is analysed using AND-OR tree analysis and validated by simulations.

A Physical-layer Rateless Code for Wireless Channels

Speakers	Shuang Tian
Seminar Date	Thursday, 24 May, 2012
Abstract	In this talk, a physical-layer rateless code for wireless channels is introduced. A novel rateless encoding scheme is developed to overcome the high error floor problem caused by the low density generator matrix (LDGM)-like encoding scheme in conventional rateless codes. This is achieved by providing each symbol with approximately equal protection in the encoding process. An extrinsic information transfer (EXIT) chart based optimization approach is proposed to obtain a universal check node degree distribution, which can achieve near-capacity performances for a wide range of signal to noise ratios (SNR). Simulation results show that, under the same channel conditions and transmission overheads, the bit-error-rate (BER) performance of the proposed scheme considerably outperforms the existing rateless codes in additive white Gaussian noise (AWGN) channels, particularly at low BER regions.

A Near Optimal Routing Scheme for Multi-hop Relay Networks Based on Viterbi Algorithm

Speakers	Qimin You
Seminar Date	Thursday, 17 May, 2012
Location	Room 712 EE Building
Abstract	In a wireless multi-hop relay network, the optimal routing scheme with exhaustive path search entails high computational complexity and large storage requirement, and is impractical for a large number of hops. In this paper, we propose a suboptimal path selection scheme, based on amplify-and-forward (AF) protocol, that has outage performance close to the optimal routing scheme, but with much less complexity. The proposed scheme draws on the analogy between the node distribution of a commonly used relay network model and the trellis of a convolutional code, and applies the Viterbi algorithm in selecting a path to maximize the end-to-end signal-to-noise ratio (SNR). In specific, the relay network topology is first mapped to the trellis diagram of a convolutional code. In the trellis, the branch metric is defined as the inverse of the instantaneous SNR of the channel connecting two relays in two adjacent clusters. Consequently, the path metric is equal to the inverse of the equivalent SNR of the path. Then, the sliding window Viterbi algorithm is used to select a path from the source to the destination. Simulation results show that when the window size is five times the total encoder memory or more, the proposed routing scheme achieves near optimal outage performance. The proposed scheme has a polynomial complexity and low communication overhead. Therefore, it is very efficient for relay networks with a large number of hops.

A New Cross-Layer User Scheduler for Wireless Multimedia Relay Networks

Speakers	Malcolm Egan
Seminar Date	Thursday, 10 May, 2012
Location	Room 712 EE Building
Abstract	We propose a new scheduler for wireless multimedia relay networks. Our scheduler is designed to account for delay, symbol error probability (SEP), and packet loss probability (PLP) due to buffer overflow. We develop a cross-layer scheduling approach for the downlink to balance these system metrics. Our scheduler is based on a new metric we call the delay in packet scheduling (DPS). The user with the largest weighted signal-to-noise ratio is scheduled, where the weight is a function of the DPS. We derive analytical expressions for the probability mass function of the DPS and the SEP of the scheduled user in Rayleigh fading. We then derive an analytical expression for the PLP due to buffer overflow. We show that the probability a target DPS is met is 30% higher for our new scheme compared to the standard opportunistic equal weight scheduler, with negligible degradation in the SEP of the scheduled user. We show that this can lead to a 85% improvement in the PLP.

Multiple Interpretations for Multi-source Multi-destination Wireless Relay Network Coded Systems

Speakers	Yuanye Ma
Seminar Date	Thursday, 10 May, 2012
Location	Room 712 EE Building
Abstract	Multi-source multi-destination wireless relay network coded systems will be presented in this seminar. To achieve multiple interpretations at different receivers, we employ nested codes in our proposed system. Besides, an opportunistic scheduling (OS) technique is adopted at the relay to maximize the system capacity. The proposed system model combines the merits of both nested codes and OS. First, we present the detailed coding process of the proposed scheme. Then, we derive the upper bounds on the bit error probability of the schemes with and without OS. Finally, we investigate good codes for our system and carry out simulations to validate the theoretical analysis.

Traffic Modeling for Smart Grid Applications

Speakers	Obada AL-Khatib
Seminar Date	Thursday, 26 April, 2012
Location	Room 712 EE Building
Abstract	A dramatic increase of interest in the research and deployment of Smart Grid in both academic organizations and industrial fields has been witnessed in the last few years. In order to realize the Smart Grid vision, it is necessary to develop a traffic model for the communication technology that will be used in Smart Grid access networks. Modeling will provide a basis for predicting performance of the network before costly equipment procurement and deployment takes place. A traffic model via a queuing theory approach that is primarily based on a multi-server, multi-priority, non-preemptive queue will be presented . We assume there are two classes of traffic in Smart Grid access networks, namely fixed scheduling traffic and event-driven traffic. Formulae are developed for evaluation of some performance metrics such as mean waiting time, and mean queue size. The analytical model is validated simulations.

Transmission capacity of wireless CSMA networks

Speakers	Tao Yang
Seminar Date	Thursday, 15 March, 2012
Abstract	Outage probability (OP) and transmission capacity (TC) are two metrics that are often used together to quantify the achievable capacity of decentralized wireless networks, where the OP measures the probability that a direct transmission fails and the TC measures the maximum spatial density of successful concurrent transmissions, subject to a constraint on the OP. In CSMA networks, spatial correlations between concurrent transmitters makes the analysis of the OP and the TC a challenging task. In this talk, we show how to derive an upper bound on the OP in CSMA networks subject to Rayleigh fading, which is applicable for any node distribution. On that basis, we obtain an upper bound on the TC in CSMA networks satisfying a pre-determined OP constraint.

Distributed Matching Algorithm for Physical Layer Security with Multiple Source-Destination Pairs and Jammer Nodes

Speakers	Siavash Bayat
Seminar Date	Thursday, 1 December, 2011
Location	Room 712 EE Building
Abstract	Physical layer security is an emerging security research area that explores the possibilities of achieving high secrecy data transmission between source-destination nodes, while malicious eavesdroppers will obtain zero information from the source nodes. In this paper we consider enhancing the security level of a network comprising of multiple source-destination pairs, multiple friendly jammers and a malicious eavesdropper. We propose a distributed algorithm to facilitate secure data transmission from each source to its corresponding destination, through the help of friendly jammers. The key idea behind the proposed algorithm is that the friendly jammers help the source nodes to increase their secrecy rates and in exchange, the source nodes provide monetary compensation to the friendly jammers. We prove that after a limited number of iterations, the proposed algorithm converges to an optimal stable matching. Numerical analysis also reveals that the distributed algorithm can achieve a performance comparable to an optimal centralized solution, but with a significantly less overhead and complexity.

Throughput Optimization for MIMO Y Channels with Physical Network Coding and Adaptive Modulation

Speakers	Keov Kolyan Teav
Seminar Date	Thursday, 17 November, 2011
Location	Room 712 EE Building
Abstract	The multiple-input multiple-output (MIMO) Y channel, where three users simultaneously exchange independent messages with each other via a single relay within two time slots, is considered in this paper. We first propose a cooperative network coding protocol, which is called denoise-demodulate-and-forward (DDF), with the design of transmit beamforming and combining schemes to increase network throughput. More importantly, we formulate an optimization problem by using the newly derived bit error rate (BER) expression of adaptive M-ary quadrature amplitude modulation (M-QAM). As the result, the modulation types for both time slots can be chosen to maximize the total throughput of the proposed system under the BER constraint. Performance evaluations show that the proposed scheme can significantly improve the total throughput by comparing to the existing MIMO Y channel and the solution of the optimization problem is validated.

Information Propagation in Mobile Ad-hoc Networks Using Epidemic Routing

Speakers	Zijie Zhang
Seminar Date	Thursday, 3 November, 2011
Location	Room 712 EE Building
Abstract	A mobile ad-hoc network (MANET) is a self-organizing network composed of mobile devices with wireless communication capabilities, such as PDAs, smart phones or intelligent vehicles. The applications of MANET are prospering in various areas, e.g. mobile social networks for news/advertisement delivery or emergency communication networks in the event of natural disasters. In a MANET, the information dissemination relies on the local ad-hoc connections between mobile users that emerge as these mobile users move and meet each other. The wireless link between two devices is intermittent due to the dynamics of the network topology and the limits of the radio transmission power of mobile devices. Therefore the information propagation process in a MANET is usually different from that in a static network. In this talk, we consider the dissemination of a piece of information from several base stations to mobile nodes in a MANET, using a Susceptible-Infectious-Recovered (SIR) epidemic routing algorithm. A distinguishing feature of the SIR algorithm, which leverages the mobility of mobile users, is that a relay node carries and forwards a piece of information for a specified amount of time. We study the probability that a non-vanishingly-small fraction of nodes receive the information, as well as the expected fraction of nodes that receive the information.

Limited Feedback for Multiuser MIMO: Structure for High Resolution

Speakers	Malcolm Egan
Seminar Date	Thursday, 20 October, 2011
Location	Room 712 EE Building
Abstract	The finite rate MIMO broadcast channel has been employed in many current and future standards such as LTE. We propose a new design criterion for the codebook at each user employing vector quantization to feedback channel state information to the base station. We show that the design criterion has a globally optimal solution which is achieved by a codebook that is a tight frame and the orbit of a finite group of unitary matrices on an initial vector. As a result, the codebooks are structured and can be used at high SNR where a large number of codewords are required to achieve high resolution. We then construct a new codebook based on the proposed design criterion using the group $SL_2(F_5)$. We show via simulations that the proposed codebook outperforms the structured codebooks in the current literature. In particular the codebook based on $SL_2(F_5)$ outperforms structured codebooks in the current literature by 60% at a SNR of 30dB for a system with a two antenna base station and two users.

A New Iterative Doppler-assisted Channel Estimation and parallel Interference Cancellation for High Mobility MIMO-OFDM Systems

Speakers	Neda Aboutorab
Seminar Date	Thursday, 6 October, 2011
Location	2pm, Room 712, EE Building
Abstract	The new wireless standard, Long-Term-Evolution (LTE), needs to support high data rate multiple input multiple output (MIMO)-orthogonal frequency division multiplexing (OFDM) transmission for highly mobile users. Due to users' mobility, the wireless channel becomes time-variant and frequency-selective. The symbol transmission is thus impaired by Doppler spread. As a consequence, the known channel estimation methods do not give satisfactory performances. In this talk, we propose an iterative channel estimation and inter-carrier interference (ICI) cancellation method for highly mobile users in LTE systems. The proposed scheme estimates the wireless channel by using pilot symbols, estimates of the data symbols and Doppler spread information at the receiver. The wireless channel is expressed by a weighted time-domain channel interpolation, where the interpolation weights are designed based on the Doppler spread and time-domain channel correlations. The channel estimates are obtained by employing a least square (LS) method. A simplified parallel interference cancellation (PIC) scheme coupled with decision statistical combining (DSC) is used to cancel the ICI and to improve the data symbols detection. These data symbols are then utilized to refine the channel estimation further, iteratively. The simulation results show that the performance degradation of the proposed iterative channel estimation and ICI cancellation scheme in a high mobility environment, compared to the performance of a system when users are static and perfect channel state information (CSI) is available at the receiver, is minimal.

User cooperation via rateless coding

Speakers	Mahyar Shirvanimoghaddam
Seminar Date	Thursday, 29 September, 2011
Location	2pm, Room 712, EE Building
Abstract	In this work we present a new rateless coded cooperation (CC) scheme for the two-user cooperative multiple access channel (MAC), where two users cooperatively communicate with a common base station (BS). In this scheme, each user uses a rateless code to encode a block of information symbols into a codeword and broadcasts it to the partner and destination. We consider two CC strategies, a fully coded cooperation (FCC) scheme used in the conventional cooperative schemes and a newly proposed partially coded cooperation (PCC) scheme. In FCC, each user starts coded cooperation process only when it can fully recover the whole block of the other user's information symbols. In contrast, in PCC, each user starts cooperation as soon as it receives a fraction of new messages sent from the other user. The degree distributions for both FCC and PCC schemes are designed to maximize the system throughput. Simulation results show that the proposed schemes achieve a higher throughput than the conventional schemes in various scenarios.

Capacity of Interference-limited Three Dimensional CSMA Networks

Speakers	Tao Yang
Seminar Date	Thursday, 22 September, 2011
Location	2pm, Room 712, EE Building
Abstract	We study the throughput of interference-limited three dimensional CSMA networks. Specifically, consider a network with a total of n nodes uniformly and i.i.d. in a cube with edge length n^{1/3}. Further, CSMA random access scheme is employed and a direct transmission between a pair of nodes is successful if and only if the associated SINR, determined by the widely used SINR model, is greater than or equal to a given threshold. First, we give a sufficient condition on the transmit power required for the above CSMA network to be asymptotically almost surely connected as n\rightarrow\infty under the SINR model. Then, we show that the throughput obtainable by each node for an arbitrarily chosen destination is lower bounded by \Theta(1/(n\log^{2}n)^{1/3}) bits per second. The result shows that a throughput of \Theta(1/(n log^{2}n)^{1/3}) is achievable with a simple routing scheme even when CSMA random access scheme is used and a minimum SINR requirement is imposed for each successful transmission.

Low complexity semi-blind channel estimation algorithm in two-way relay networks

Speakers	Qiong Zhao
Seminar Date	Thursday, 8 September, 2011
Location	2pm, Room 712, EE Building
Abstract	In this work, we propose a low complexity semi-blind channel estimation algorithm, referred to as the variance squared maximum likelihood (VSML) estimator, which employs only one training symbol in each channel estimation, to estimate general non-reciprocal flat-fading channels in amplify-and-forward (AF) two-way relay networks (TWRNs). We formulate a non-convex objective function and obtain closed-form channel estimates by minimizing its approximate expression. Theoretical analysis proves that the derived channel estimation is asymptotically optimal in large sample size scenarios. Monte-Carlo simulation results show that the VSML estimator outperforms the existing relaxed maximum likelihood (RML) estimator in terms of mean squared error (MSE) performance and remarkably reduces the computational complexity by completely avoiding the grid-search algorithm under M-ary phase-shift-keying (MPSK) modulation.

A Near Optimal Distributed Routing Scheme Based on Viterbi Algorithm

Speakers	Qimin You
Seminar Date	Thursday, 4 August, 2011
Location	2pm, Rm 712, EE Building
Abstract	In this paper, a novel path selection algorithm which aims to maximize the end-to-end (E2E) SNR in an amplify-and-forward (AF) multi-hop relay network is investigated. Due to the high computational complexity and large storage requirements, the optimal routing scheme which searches all the possible paths and selects the best one is impractical for real life applications. To facilitate and simplify the path selection process, we propose a distributed routing scheme based on Viterbi algorithm with truncated decoding memory. In the proposed algorithm, we map the network topology to a trellis diagram of a convolutional code. The link metric of the trellis is the inverse of the instantaneous SNR of the corresponding hop. Consequently, the path metric, which is the accumulated link metrics along the path, is equal to the inverse of the equivalent E2E SNR of the path. In the proposed algorithm, information on the most recent w (window size) hops of the network is stored and a relay is selected on the first hop within the window. The proposed routing algorithm has a polynomial complexity and low communication overhead. Therefore, it is very efficient for large relay networks with hundreds of hops and up to dozens of relays per hop.

Cooperative Spectrum Sharing in Cognitive Radio Networks with Multiple Antennas

Speakers	Raed Manna
Seminar Date	Thursday, 26 May, 2011
Location	2pm, Room 712, EE Building
Abstract	In this talk, we consider a cognitive radio network consisting of a primary transmitter-primary receiver pair, and a secondary base station-secondary receiver pair. To improve the performance of both the primary and secondary pair, we propose an overlay spectrum sharing scheme where the primary user (PU) leases half of its time slots to the secondary user (SU) in exchange for the SU cooperatively relaying the PU's data using the amplify and forward scheme. The proposed scheme will involve the design of antenna weights and power allocation to meet a certain error or rate design criteria for both the PU and SU. To analyze the performance of the proposed scheme, we derive new closed form expressions for the rate and bit error rate for arbitrary signal-to-noise ratio (SNR). In addition, we carry out an asymptotic analysis in the high SNR regime to obtain the diversity order. These expressions, along with numerical analysis, reveal that the proposed cooperative overlay scheme can achieve significant performance gains, for both the PU and the SU, compared to a conventional non-cooperative underlay scheme, which gives both users the incentive to cooperate.

Creative Inspiration in Research

Speakers	Dr Ehssan Sakhaee
Seminar Date	Thursday, 12 May, 2011
Location	2pm, Room 712, EE Building
Abstract	Creative Inspiration in Research looks at the art of creative thinking in a research context. It explores how to think outside the box and come up with new ideas. Creative thinking could lead to invention and innovation, new products or technologies that have not been thought of before. Creativity requires lateral thinking, a skill that can be developed through exercises and building confidence. It also involves willing to make mistakes and exploring different methods of solving the same problem. Since creative thinking is often in oppose to logical thinking patterns, engineers are sometimes less prone exercise their creative side due to the rigid logical and often routine nature of current engineering practices. However in engineering research creativity is essential to come up with new ideas for not just success in your research degree but also to open new doors for your future research and career prospects. Due to the fact that much of routine engineering jobs are increasingly being outsourced from Asia, creativity becomes an essential skill for the future engineer in developed nations. This seminar will run through how to unlock your creativity in engineering research.

On the Information Propagation in Mobile Ad-hoc Networks Using Epidemic Routing

Speakers	Zijie Zhang
Seminar Date	Thursday, 28 April, 2011
Location	11am, Room 712, EE Building
Abstract	A mobile ad-hoc network (MANET) is a self-organizing network composed of mobile devices with wireless communication capabilities, such as PDA or smart phones. In a MANET, the information dissemination relies on the local ad-hoc connections between mobile users that emerge as these mobile users move and meet each other. The wireless link between two devices is intermittent due to the dynamics of the network topology and the limits of the radio transmission power of mobile devices. Therefore the information propagation process in a MANET is different from that in a static network. In this talk, we consider the case that a piece of information is broadcast from a source node to all other nodes in the network, using a Susceptible-Infectious-Recovered (SIR) epidemic routing algorithm. A distinguishing feature of the SIR algorithm, which leverages the mobility of mobile users, is that a relay node carries and forwards a piece of information for a specified amount of time. We study the probability that a non-zero fraction of nodes receive the information in the limit of large network size, as well as the expected fraction of nodes that receive the information.

Distributed TAS/MRC in MIMO Relay Networks

Speakers	Phee Lep Yeoh
Seminar Date	Thursday, 21 April, 2011
Location	11am, Room 712, EE Building
Abstract	We propose distributed transmit antenna selection with receiver maximal-ratio combining (TAS/MRC) for use in a two-hop multiple-input multiple-output (MIMO) relay network. The network under consideration is equipped with NS, NR, and ND antennas at the source, relay, and destination, respectively. First, we derive a new closed-form expression for the exact cumulative distribution function (cdf) of the end-to-end SNR. Based on this, we present a new closed-form expression for the exact symbol error rate (SER). Our analytical results are further evaluated in the high SNR regime, leading to practical design insights. Our asymptotic expressions are concise and have the added advantage of explicitly characterizing the diversity order and the array gain of the network. Our exact and asymptotic results are valid for general operating scenarios with distinct average received SNRs in each hop.

Cooperative Spectrum Sharing with multiple primary users and secondary users with incomplete information

Speakers	Siavash Bayat
Seminar Date	Thursday, 14 April, 2011
Location	2pm, Room 720, EE Building
Abstract	Spectrum sharing is a technique which enables licensed (primary) and unlicensed (secondary) users use the same spectrum. In this work we use a “property-rights model” in which the all users know about the presence of each other and are negotiating with each other to reach to an agreement. We choose a realistic channel model which includes the fading and path loss. In many cases considering the fading makes the spectrum sharing model completely untraceable and normal games cannot be used to analyze the model. In this model buyers and sellers are trying to increase their quality of service by cooperation with each other. So we will talk about a mechanism which is distributed and optimized in terms of sum of data rates and also brings in to account the dynamic features of spectrum at any time. We will find a lower bound and upper bound for the PUs and SUs and will introduce a mechanism which can adjust the data rate of both PUs and SUs at any arbitrary value.

Codebook Design for the Finite Rate MIMO Broadcast Channel with Zero-Forcing Precoding

Speakers	Malcolm Egan
Seminar Date	Monday, 28 March, 2011
Location	11am, Room 720, EE Building
Abstract	We present a novel codebook design criterion for the limited feedback MIMO broadcast channel with zero-forcing precoding. To reduce system implementation complexity, each user has the same codebook. We derive a new sum-rate bound, show the bound maximization problem is invex and explicitly solve it using the Karush-Kuhn-Tucker (KKT) conditions. We show that the solution has the same structure as Grassmannian frames. We also derive a new lower bound on the outage probability. We show via simulations that our codebooks designed using Grassmannian frames outperform random vector quantization with different codebooks at each user for codebook sizes greater than 9 bits.

A New Iterative Doppler-assisted Channel Estimation with ICI Cancellation for High Mobility OFDM Systems

Speakers	Neda Aboutorab
Seminar Date	Monday, 14 March, 2011
Location	11am, Room 720, EE Building
Abstract	The new wireless standard, Long-Term-Evolution (LTE), needs to support high data rate orthogonal frequency-division multiplexing (OFDM) transmission for highly mobile users. Due to users’ mobility, the wireless channel becomes time-variant-frequency-selective and the symbol transmission is impaired by Doppler spread. As a consequence, the channel estimation process required for reliable data detection is no longer accurate. In this paper, we propose a joint iterative channel estimation and inter-carrier interference (ICI) cancellation method that estimates wireless channel by using the pilot symbols, the estimate of the data symbols and users’ velocity at the receiver. The wireless channel is estimated by using a weighted time-domain interpolation of the pilot channels. The interpolation weights are designed based on Doppler spread that is calculated using the knowledge of users’ velocity at the receiver. The pilot channels is estimated using a least squares (LS) method. The estimate of data symbols is then used to improve the pilot channel estimates in the next iterations. The simulation results show that the performance degradation of the proposed iterative channel estimation with ICI cancellation when users move at the speed of up to 324 Km/h compared to the performance of a system when users are static and perfect channel state information (CSI) is available at the receiver, is minimal.

Intelligent Soft Frequency Reuse and Inter-Cell Interference Management for LTE Networks

Speakers	Manli Qian
Seminar Date	Monday, 7 March, 2011
Location	11am, Room 720, EE Building
Abstract	In existing literature, transmit power or the number of subcarriers allocated to each cell in fractional frequency reuse (FFR) and soft frequency reuse (SFR) schemes are fixed. This limits the potential performance of these frequency reuse schemes. In this work, we aim at finding an optimal SFR system configuration by jointly allocating subcarriers and power in multi-cell Long Term Evolution (LTE) networks. We propose an iterative algorithm that can dynamically allocate subcarriers, power to each cell with inter-cell interference coordination (ICIC) and adjust these parameters to the optimum according to wireless channel status and traffic loads. Simulation results show that the proposed algorithm converges fast to a SFR solution and has better system and cell edge throughput than existing Reuse 1, FFR and static SFR schemes.

Connectivity of Wireless CSMA Multi-hop Networks

Speakers	Tao Yang
Seminar Date	Monday, 22 November, 2010
Location	11am, Room 720, EE Building
Abstract	Connectivity is one of the most fundamental properties of wireless multi-hop networks. In a wireless network with many concurrent transmissions, signals trans- mitted at the same time may mutually interfere with each other. In this paper we consider the impact of interference on the connectivity of CSMA networks. First, it is shown that the aggregate interference experienced by any receiver in a CSMA network with arbitrarily distributed nodes is upper bounded. Then, we derive an equivalent transmission range for CSMA networks where any pair of nodes whose Euclidean distance is smaller than or equal to the transmission range are directly connected. Finally we give a sufficient condition on the transmission power required for a CSMA network with a total of n nodes i.i.d. on a n×n square following a uniform distribution to be asymptotically almost surely connected as n under the SINR model. It is shown that the transmission power only needs to be increased by a constant factor to combat interference and maintain connectivity compared with that considering a unit disk model without interference. This result is also in sharp contrast with previous results considering the connectivity of ALOHA networks under the SINR model.

User Scheduling for the Broadcast Channel Using a Sum-Rate Threshold

Speakers	Malcolm Egan
Seminar Date	Monday, 15 November, 2010
Location	11am, Room 720, EE Building
Abstract	In this presentation, we present a novel user selection scheme for the broadcast channel (BC) using zero-forcing (ZF) precoding with optimal power control. Our scheme is based on a threshold that sets a minimum acceptable sum-rate. In order to design the threshold, we develop a new approximation of the sum-rate and derive simple design rules for the threshold with complexity constraints. We also extend the scheme so that users with differing quality of service (QoS) demands can be accommodated. Simulations show our scheme performs close to the exhaustive search algorithm, but with significantly reduced complexity. Moreover, it significantly reduces the outages compared to non-threshold based low complexity scheduling schemes.

Multi-hop Bi-directional Relay Transmission Schemes Using Amplify-and-Forward and Analog Network Coding

Speakers	Qimin You
Seminar Date	Monday, 8 November, 2010
Location	11am, Room 720, EE Building
Abstract	In this presentation, we investigate two different multi-hop bi-directional relay transmission schemes based on amplify-and-forward (AF) protocol and analogue network coding (ANC). In the first scheme, referred to as the AF-ANC-Central scheme, AF is employed at each of the intermediate nodes, while ANC is only utilized at the central relay. In the second scheme, referred to as the AF-ANC-Even scheme, the even relays perform ANC and the odd relays only perform subtraction and AF. For reference purpose, the one way AF-No-NC scheme is also considered, where the intermediate relays only perform AF and no ANC is employed. Bit error rate (BER) lower bounds of these three schemes are obtained, and are verified by Monte Carlo simulations to be asymptotically tight at high signal-to-noise ratios (SNRs). It is shown that the combination of AF and ANC is able to significantly improve system throughput when compared with the AF-No-ANC scheme. The BER performance of the AF-ANC-Central and AF-No-ANC schemes are of the same at high SNR region, but the AF-ANC-Central scheme is able to double the system throughput. The AF-ANC-Even is able to further increase the system throughput with an SNR loss upper bounded by 1.76 dB when compared to the AF-ANC-Central and AF-No-ANC schemes.

Piecewise-and-Forward Relaying in Wireless Relay Networks

Speakers	Shuang Tian
Seminar Date	Monday, 25 October, 2010
Location	11am, Room 720, EE Building
Abstract	In this paper, we propose a piecewise-and-forward (PF) relaying protocol for wireless relay networks. In PF, the received signal at each relay is compared to an adaptive threshold. If the amplitude of the received signal is above the threshold, the relay will decode the signal, otherwise, the relay will forward the received signal after linear processing. An optimal maximum likelihood (ML) detector is developed at the destination for the proposed PF relaying protocol. Simulation results show that the PF protocol outperforms the existing amplify-and-forward, decode-and-forward and estimate-and-forward relaying protocols, and the gain increases as the number of relays increases.

Exact and Asymptotic SER of Nonregenerative Relaying in MIMO Multi-Relay Networks

Speakers	Phee Lep Yeoh
Seminar Date	Monday, 11 October, 2010
Location	11am, Room 720, EE Building
Abstract	MIMO multi-relay networks with amplify-and-forward (AF) relaying is analyzed under independent but not necessarily identically distributed (i.n.d.) Rayleigh fading. We consider a flexible hierarchical network architecture which encompasses arbitrary L number of parallel relays, each equipped with N transmit and receive antennas, typical of wireless mesh networks. New closed-form expressions are derived for the important statistics of the cumulative distribution function (cdf), the probability density function (pdf), and the moment generating function (mgf) of the end-to-end signal-to-noise ratio (SNR). Based on these, we evaluate the exact and approximate symbol error rate (SER) with M-ary phase-shift keying (MPSK). We further derive a new accurate and concise expression for the SER in the high SNR regime. We explicitly reveal that the diversity order is equal to LN + 1. We highlight that a lower SER is achieved by placing the relays closer to the destination relative to the source.

Design of Distributed Network-Channel Codes for Wireless Sensor Networks

Speakers	Kun Pang
Seminar Date	Monday, 27 September, 2010
Location	11am, Room 720, EE Building
Abstract	In this paper, we use extrinsic information transfer (EXIT) chart to design irregular low density generator matrix (LDGM) codes to form distributed network-channel codes in a wireless sensor network. We formulate the code design and code search as a linear programming (LP) problem. We consider a real-time wireless network with randomly changeable fading channels, resulting in link failures and time varying network topology. In forming such a dynamic network, the connected number of source nodes at each relay needs to satisfy previously obtained degree distributions. At the same time, the channel quality of the data links connecting the source nodes and relay nodes has to be considered. We propose the optimal relaying selection scheme to present such a solution. Simulation results of the proposed irregular codes show that a considerable performance improvement can be achieved in the waterfall region compared with the existing codes.

On The k-hop Partial Connectivity of Finite Wireless Multi-hop Networks

Speakers	Seh Chun Ng
Seminar Date	Monday, 20 September, 2010
Location	11am, Room 720, EE Building
Abstract	In this work, we study the connectivity of wireless multi-hop networks where a finite number of nodes are randomly deployed in a given 2D area. In real applications, it is often desirable to limit the maximum number of hops between the source and destination nodes in order to provide meaningful services. On the other hand, connectivity imposes strong requirements on transmission range/power. Therefore it may be beneficial to allow a small fraction of nodes to be disconnected so that the network is partially connected. Based on the above two considerations, we provide analytical results on the k-hop partial connectivity of ad-hoc networks, which is the fraction of source-destination pairs that are connected by a path in at most k hops. Further, considering wireless multi-hop networks in the presence of infrastructure nodes (gateway), we provide the analytical results on the k hop partial accessibility, which is the fraction of nodes that are connected to at least one infrastructure node by a path in at most k hops. The research provides useful guidelines on the design of ad-hoc networks and infrastructure-based wireless multi-hop networks.

Information Dissemination in Mobile Vehicular Ad Hoc Networks

Speakers	Zijie Zhang
Seminar Date	Monday, 6 September, 2010
Location	11am, Room 720, EE Building
Abstract	Vehicular ad hoc network (VANET) is a mobile ad hoc network formed by vehicles traveling on the road. VANETs have attracted significant interest in recent years due to a large number of potential applications, such as accident avoidance messaging, congestion sensing, traffic metering and other information services (e.g. internet access). Analytical model is proposed to study the process of the information propagation in VANETs. Our model assumes that time is divided into time slots of equal length and that each vehicle changes its speed at the beginning of each time slot. Analytical formulas for the fundamental properties of the information propagation process as well as information propagation speed are derived. Using the formulas, one can straightforwardly study the impact on the information propagation speed of various parameters such as radio range, vehicular traffic density and the time variation of vehicle speed. The accuracy of the results is validated using simulations. The research provides useful guidelines on the design of vehicular ad hoc networks.

Channel Estimation for High Mobility OFDM Systems

Speakers	Neda Aboutorab
Seminar Date	Monday, 30 August, 2010
Location	11am, Room 720, EE Building
Abstract	For orthogonal frequency division multiplexing (OFDM) systems operating in high mobility scenarios, channel estimation becomes a challenging issue, due to the fast channel variation and severe inter-carrier interference (ICI). OFDM techniques have been adopted in 3GPP LTE and WiMAX standards to achieve very high data rates (> 10Mbps). In these standards, high mobility users, moving at speeds higher than 300Km/h, need to be supported. Accurate channel state information (CSI) is required for reliable signal detection at the receiver. In order to accurately estimate the wireless channel, pilot symbols are inserted among sub-carriers before transmission. In a high mobility environment, the wireless channel is time-variant and frequency-selective causing the symbol transmission to be impaired by the Doppler spread. The Doppler spread destroys the orthogonality and creates ICI between OFDM sub-carriers. In addition, the channel changes significantly within one OFDM symbol. As a consequence, the standard channel estimation methods cannot be used in 3GPP LTE and WiMAX to support high mobility users. In this talk we introduce two channel estimation techniques, one based on the pilot symbols and iterative estimate of data symbols and the other one based on the pilot symbols and the knowledge of the Doppler spread . The channel is estimated by time-domain interpolation and least-square (LS) methods. In addition, the second technique is extended further to benefit from the iterative estimate of data symbols in the ICI cancellation process. The simulation results show that the performance of the technique using the Doppler spread outperforms the scheme with iterative data symbol estimation. It is also shown that the performance degradation of the proposed schemes when users move at speed of up to 324Km/h compared to the performance of a system with a zero Doppler shift is very marginal.

Cooperative Spectrum Sharing in Cognitive Radio Networks

Speakers	Raed Manna
Seminar Date	Monday, 23 August, 2010
Location	11am, Room 720, EE Building
Abstract	Cognitive radio (CR) has been recently proposed as a promising technology to improve the utilization efficiency of radio spectrum. It allows the secondary user (SU) networks to coexist with the primary user (PU) networks through spectrum sharing, provided that the secondary spectrum access will not adversely affect the PU�s performance. To allow for this spectrum sharing, three common models have been considered in the literature: the interweave, underlay and overlay model. In this work, we consider an overlay cognitive radio network consisting of a primary transmitter-primary receiver pair, and a secondary base station-secondary receiver pair. We propose a spectrum sharing scheme based on cooperative relay transmission to achieve high performance for both primary user (PU) pair and secondary user (SU) pair. The proposed scheme will involve the design of antenna weights and power allocation, to meet a certain error or sum-rate design criteria for the PU and the SU. To analyze the performance of our proposed scheme, we derive new expressions for the bit error rate and sum-rate. These expressions, along with numerical analysis, reveal that our proposed scheme can achieve significant performance gains, for both the PU and the SU, compared to a conventional underlay scheme with no cooperation.

Cooperative Auction Based Spectrum Sharing

Speakers	Siavash Bayat
Seminar Date	Monday, 16 August, 2010
Location	11am, Room 720, EE Building
Abstract	Cognitive radio (CR) concept is a promising way for combating the scarcity of spectrum and implementing the coexistence of licensed (primary) and unlicensed (secondary) users. Many solutions for spectrum sharing in CR are based on opportunistic access of secondary users (SU) to the primary users (PU) spectrum while primary users (PU) are not aware about the presence of unlicensed users thanks to spectrum sensing techniques. However, all spectrum sensing methods have limitations and are combined with errors and probability of detections less than one that will cause interferences in many cases. The most realistic and applicable method would be a method which both primary and secondary user are aware about the presence of each other and so spectrum sharing in a more efficient manner. In this work we use �property-rights model� in which the secondary users are transparent to primary users and use the spectrum after negotiating with PUs. In fact in property-rights model SUs are granted to use of spectrum by PUs. On the other hand in most of the spectrum sharing models which have been offered so far the dynamic characteristic of the spectrum channels which is modelled by several concepts such as fading and dynamic shadowing has not been considered. In many cases considering the fading makes the spectrum sharing model completely unwieldy and untraceable and because of that most of the models for spectrum sharing in wireless communications does not consider that and so are one-object models. However, in many situations the spectrum buyers and spectrum sellers (leasers) negotiate for short term spectrum buying or spectrum leasing and the difference between the quality of spectrums need to be considered and one-object models cannot help any longer. Moreover, unlike the normal spectrum marketing we do not assume monetary transactions between buyers and sellers and instead the PUs will try to increase their quality of service by doing cooperation with SUs. So we will talk about a mechanism which is distributed and optimized in terms of sum of data rates and also brings in to account the dynamic features of spectrum at any time.

Joint Relay Selection and Network Coding Using Decode-and-Forward Protocol in Two-Way Relay Channels

Speakers	Qimin You
Seminar Date	Monday, 7 June, 2010
Location	11am, Room 720, EE Building
Abstract	In this presentation, we analyze the bit error rate (BER) performance of a single relay selection with network coding (S-RS-NC) scheme in two-way relay channels. In this scheme, two source nodes first broadcast their information to the relays sequentially. A single relay which optimizes the system performance is selected. The selected relay decodes the received signals from two sources, performs network coding on two symbol estimates and then forwards them to two source nodes. Equivalent signal-to-noise-ratio (SNR) of the whole source-relay-destination link is analyzed and its high SNR approximation is derived. Closed form BER expressions are then derived and the results are verified through Monte-Carlo simulations which show that the derived analytical expressions offer a tight bound for the average BER. It is shown that this S-RS-NC scheme can achieve a full diversity order as if all relays were used. Simulation results show that the S-RS-NC scheme can bring considerable gains over the all-participation relaying scheme when the source transmission power is larger than or equal to the relay transmission power. However, when the relay transmission power is greater than the source power, the single relay selection scheme could be inferior to the all-participation relaying scheme. This is quite different from the conventional relay selection scheme in one way relay network, where the relay selection always outperforms the all participation relaying scheme.

A Practical Passive Interference Cancellation Structure for Wireless Networks

Speakers	Bin Guo
Seminar Date	Monday, 31 May, 2010
Location	11am, Room 720, EE Building
Abstract	Co-channel interference exists among co-existing base stations (BSs) transmitting at the same frequency and limits the capacity of wireless networks. In this paper, we propose a novel passive co-channel interference cancellation scheme for wireless networks. In this scheme, co-channel interference cancellation is performed at one inserted common base station (CBS), which is connected to all co-existing BSs through high-speed wired backbones. Co-existing BSs and MSs are not required to cancel co-channel interference. This scheme can be directly applied to wireless networks, such as WLAN or WiMax without altering the configuration of existing BSs and mobile stations (MSs). Linear precoding algorithms based on zero-forcing (ZF) and minimum mean-square-error (MMSE) and lattice precoding are considered at the CBS to eliminate the interference between BSs and to optimize the system average bit error rate (BER) performance. The simulation results show that the proposed scheme can effectively eliminate interference and is superior to the cooperative BSs transmission scheme, where BSs cooperate to transmit to multiple MSs simultaneously.

Distributed Analog Channel Coding for Wireless Relay Networks.

Speakers	Md Shahriar Rahman
Seminar Date	Monday, 24 May, 2010
Location	11am, Room 720, EE Building
Abstract	Distributed coding has been shown to be an efficient scheme to improve the performance of wireless relay networks. In this paper we propose an analog distributed channel coding (ADC) scheme. In the proposed scheme, the intermediate relays directly process the received analog noisy signals without making the hard estimation and performs analog channel encoding of the noisy soft estimates of the transmitted symbols, so that the processed signals forwarded by all relays can form an analog distributed convolutional codeword. Assuming the source packet is encoded by a channel code, the ADC scheme can form serial concatenated codes and iterative decoding can be applied at the destination. We compare our proposed scheme with other existing schemes under Additive White Gaussian Noise (AWGN) and fast Rayleigh fading channel and validate the simulation results with the aid of Extrinsic Information Transfer (EXIT) chart analysis. Simulation results show that the proposed ADC scheme can effectively overcome the error propagation due to the erroneous decoding at the relay in the conventional decode forward (DF) scheme and provide considerable coding gains, thus considerably outperforming the conventional soft information relaying protocols. The coding gains increase as the number of state in the relay encoder increases.

Selection Diversity with Transmit Beamforming in Amplify-and-Forward Relay Networks

Speakers	Phee Lep Yeoh
Seminar Date	Monday, 10 May, 2010
Location	11am, Room 720, EE Building
Abstract	In this presentation, we propose and analyze selection diversity with transmit beamforming as an effective tool to combat channel impairments in relay-assisted cellular networks. We consider the downlink scenario where the base station equipped with N antennas transmits to the mobile station either directly, or indirectly via the relay station, according to the link with the strongest received signal-to-noise ratio (SNR). For this system, we compare two amplify-and-forward relaying protocols: i) fixed gain relaying which requires partial channel state information (CSI), and ii) variable gain relaying which requires full CSI. We derive new exact closed-form expressions for the symbol error rate (SER) which are valid for a wide variety of modulations. Our exact results can be analyzed in the high SNR regime, leading to further design insights. We derive new asymptotic expressions which are decomposed into two key performance parameters: the array gain and the diversity order. Based on these, we demonstrate that the SER of fixed and variable gain relaying are asymptotically the same in the large N limit.

Analysis of k-Hop Connectivity Probability in Wireless Networks with Infrastructure Support

Speakers	Seh Chun Ng
Seminar Date	Monday, 12 April, 2010
Location	11am, Room 720, EE Building
Abstract	Wireless multi-hop networks with infrastructure support have been actively studied to solve the scalability problem in large scale vehicular and sensor networks that the end-to-end throughput and other performance metrics decrease sharply with the increase in the number of nodes in the network. In the infrastructure-based networks, wireless nodes are allowed to access the base stations either directly or via a multi-hop path. In order to provide meaningful services, it is often desirable to limit the maximum number of hops in the wireless multi-hop path to below a certain threshold. In this paper, we study the wireless networks, both in 1-D and 2-D, where users are randomly distributed and base stations are regularly deployed. We obtain analytically the exact and approximate k-hop connectivity probability for k = 2, i.e. the probability that all users can access to at least one base station in at most two hops, under a generic channel model. The results are verified by simulations and can be used in network planning, design and resource management, particularly in the emerging vehicular networks where multi-hop connection between vehicles and infrastructure is envisioned.

On the Information Propagation Speed in Mobile Vehicular Ad Hoc Networks

Speakers	Zijie Zhang
Seminar Date	Monday, 29 March, 2010
Location	11am, Room 720, EE Building
Abstract	Vehicular ad hoc network (VANET) is a 1D mobile ad hoc network formed by vehicles travelling along the road. VANETs have attracted significant interest in recent years due to a large number of potential applications, such as accident avoidance messaging, congestion sensing, traffic metering and other information services (e.g. internet access). We study the expected propagation speed for a piece of information to be broadcasted along the road in a VANET, which is referred to as the information propagation speed. Our model assumes that time is divided into time slots of equal length and that each vehicle changes its speed at the beginning of each time slot, independent of its speed in other time slots and the speeds of other vehicles. Any two vehicles can directly communicate with each other if their Euclidean distance is smaller than a given threshold. Analytical formulas for the information propagation speed in the above network are derived. Using the formulas, we can straightforwardly study the impact on the information propagation speed of various parameters such as vehicle density, speed and radio range. The accuracy of the formula is validated using simulations. The research provides useful guidelines on the design of vehicular ad hoc networks.

Decode-and-Forward and Amplify-and-Forward Protocols for Two-Way Relay Networks

Speakers	Shuang Tian
Seminar Date	Monday, 22 March, 2010
Location	11am, Room 720, EE Building
Abstract	Relayed transmission has been shown to be an effective way to provide spatial diversity and increased coverage in wireless networks. In particular, two-way relayed transmission has recently attracted increasing interest as it achieves higher spectral efficiency by employing network coding (NC), where the transmission in two-way relay channels takes place in two time slots or three time slots. In addition, various NC relay protocols are used at the relay to combine two received signals. The two most common ones are decode-and-forward (DF) and amplify-and-forward (AF) protocols. In this presentation, we�ll focus on the bit error rate (BER) performances of DF and AF protocols in the three time slots transmission. Moreover, to further improve the BER performance, we propose an AF protocol with optimal power allocation by allocating different power coefficients to the two received signals when they are combined at the relay. Simulation results show that optimal power allocation in AF can bring the system considerable gains.

Maximum-Throughput Irregular Distributed Space-Time Code for Near-Capacity Cooperative Communications

Speakers	Dr Michael Soon Xin Ng (Visiting Faculty from University of Southampton UK)
Seminar Date	Monday, 15 March, 2010
Location	11am, Room 720, EE Building
Abstract	An Irregular Distributed Space-Time (Ir-DST) coding scheme is designed for near-capacity cooperative communications where the system�s effective throughput is also maximized with the aid of a joint source-and-relay mode design procedure. At the source node, a serial concatenated scheme comprising an IRregular Convolutional Code (IRCC), a recursive Unity-Rate Code (URC) and a Space-Time Block Code (STBC) was designed for the sake of approaching the corresponding source-to-relay link capacity, where the IRCC was optimized with the aid of EXtrinsic Information Transfer (EXIT) charts. At the relay node, another IRCC is concatenated serially with an identical STBC. Before transmitting the relayed information, the relay�s IRCC is re optimized based on EXIT chart analysis for the sake of approaching the relay channel�s capacity as well as for maximizing the relay�s coding rate, which results in a maximized effective throughput. The topology of the Ir-DST system coincides with that of a Distributed Turbo Code. At the destination node, a novel three-stage iterative decoding scheme is constructed in order to achieve decoding convergence to an infinitesimally low Bit Error Ratio. The numerical results show the proficiency of this joint source-and-relay mode design procedure, demonstrating that the proposed Ir-DST coding scheme is capable of near capacity cooperative communications as well as of maximizing the effective throughput.

Interference Cancellation in Multi-user MIMO Relay Networks Using Beamforming and Precoding

Speakers	Neda Aboutorab
Seminar Date	Monday, 2 November, 2009
Location	11am, Room 720, EE Building
Abstract	In this paper, we investigate transmission methods for a multi-user multiple-input multiple-output (MIMO) network that utilizes base stations (BS) cooperation. To eliminate the interference between users, iterative zero forcing- (ZF) and iterative Tomlinson Harashima precoding-based (THP) schemes are proposed. In the iterative ZF-based scheme, all interference is canceled by using the transmit-receive weights. To reduce the complexity of iterative ZF scheme, in the iterative THP-based scheme, the interference is canceled by using transmit-receive weights and the THP. To achieve symbol error rate (SER) fairness among different users and further improve the performance of multi-user MIMO relay systems, we develop optimal and suboptimal power allocation (PA) methods that ensure signal-to-interference-and-noise-ratio (SINR) across all users are equal, under the power constraints at both BSs and relay station (RS). In the optimal PA scheme, the PA is done at both BSs and RS. In the sub-optimal scheme, to reduce the computational complexity of optimal PA significantly, the PA is done only at BSs, and at the RS a power scaling is performed to satisfy the RS power constraint. The simulation results show that by using optimal and sub-optimal PAs, the iterative ZF-based scheme outperforms the iterative THP-based scheme by an average of 0.4 dB at the cost of a four times higher complexity.

Amplify-and-Forward Relaying with Cooperative Selection Diversity in Nakagami-m Fading Channels

Speakers	Phil Yeoh
Seminar Date	Monday, 26 October, 2009
Location	11am, Room 720, EE Building
Abstract	Cooperative relaying has been the focus of intense research activities as a promising solution for increasing the capacity and achievable rates of wireless communication networks. This presentation investigates the symbol error rate (SER) of amplify-and-forward (AF) relaying under the generalized fading condition of independent non-identically distributed (i.n.d.) Nakagami-m fading channels, which includes Rayleigh fading as a special case. We consider the performance of cooperative selection diversity (CSD), in which both the direct path and a two-hop relay path are processed at the destination. We present new closed-form expressions for two different AF relaying protocols, i.e., fixed gain and channel-state-information (CSI)-based gain relaying. Finally, we highlight the performance gains of CSD over the non-diversity case in unbalanced fading environments.

A Practical Passive Interference Cancelation Structure for Wireless Networks

Speakers	Bin Guo
Seminar Date	Monday, 19 October, 2009
Location	11am, Room 720, EE Building
Abstract	Co-channel interference exists among the co-working base stations (BSs) in the same location and limits the capacity of wireless networks. In this paper, we propose a novel scheme to cancel this interference by using one additional common base station (CBS). CBS is connected with all co-existing BSs through wired back-bone networks. We assumed it knows the channel state information (CSI) of all wireless link of network as well as the transmitted symbols of all BSs. Linear transmit processing such as zero-forcing (ZF) and minimum mean-square-error (MMSE) are considered at CBS to eliminate the interference. BSs and MSs do not perform any signal processing related to interference cancelation. Simulation results show that our proposed scheme can effectively eliminate the interference and outperforms the point-to-point interference free transmission and conventional broadcast transmission, where single CBS broadcasts information to multiple MSs. This scheme does not need to change the existing transceiver structure of BS and MS. This will significantly reduce the system implementation complexity and shorten the deployment period. It can be directly applied to the existing wireless networks, such as WLAN or WiMax.

Multi-hop Delay Performance in Wireless Mesh Networks

Speakers	Dr. Yang Yang (University College London)
Seminar Date	Monday, 28 September, 2009
Location	11am, Room 720, EE Building
Abstract	Wireless Mesh Network (WMN) technology is an attractive solution to meet the demand of broadband network access anywhere and anytime. In order to effectively support delay sensitive applications such as video streaming and interactive gaming in a WMN, it is crucial to develop feasible methodologies and techniques for accurately analyzing, predicting and guaranteeing end-to-end delay performance over multi-hop wireless communication paths. In this talk, we extend the link-layer effective capacity (EC) model and derive a lower bound of delay-bound violation probability, or complementary cumulative distribution function (CCDF), over multi-hop wireless connections. A fluid traffic model with cross traffic and a Rayleigh fading channel with additive Gaussian noise and Doppler spectrum are considered in our study. The average multi-hop delay and jitter performance bounds are also obtained. Analytical results are verified by extensive computer simulations under different traffic load and wireless channel conditions. We find that multi-hop delay performance is much more sensitive to traffic load and maximum Doppler rate than traffic correlation.

On the Hop Count Statistics in Wireless Sensor Networks Subject to Fading

Speakers	Zijie Zhang
Seminar Date	Monday, 21 September, 2009
Location	EIE Seminar; 11am; Monday, 21st September; Rm 720 EE Building
Abstract	In a multi-hop network, the probability distribution of the number of hops between two nodes sheds insight on many research problems, such as the estimation of energy consumption, delay and throughput in end-to-end packet transmissions. We investigate the probability distribution of the number of hops between two arbitrary nodes separated by a known Euclidean distance using the greedy forwarding routing protocol in a wireless sensor network subject to the log-normal shadowing fading. Further, considering the randomness of node deployment and a complex radio environment which may result in a disconnected network, we derive the probability distribution of the number of hops traversed by the packets before being dropped if the transmission is unsuccessful. An end-to-end packet transmission may be unsuccessful if the packet sent from a source to a destination has to be dropped at an intermediate node because the greedy forwarding routing protocol is unable to find a next-hop node. Accurate results on the hop count statistics are obtained, which provides useful guidelines on the design of a randomly deployed network in a realistic radio environment.

An Iterative ZigZag Decoding for Combating Collisions in Wireless Networks

Speakers	Shahriar Rahman
Seminar Date	Monday, 7 September, 2009
Location	EIE Seminar; 11am; Monday, 7th September; Rm 720 EE Building
Abstract	Users increasingly depend on WLAN for business and entertainment, as it provides flexibility of locations and low maintenance efforts. However in presence of hidden terminals sender-receiver pairs experience severe packet loss due to collisions. In case of hidden terminal, IEEE 802.11 retransmission causes successive collision. ZigZag decoding can salvage packets from the collided signal and can improve the system throughput. Experimental results show that ZigZag reduces the average packet loss rate at hidden terminals from 72.6% to about 0.7%. But this may cause error propagation in a packet during decoding process. Here, we propose an iterative ZigZag decoding algorithm to mitigate the error propagation and further improve the system performance in presence of collisions. Simulation results show that the proposed iterative ZigZag decoding can effectively overcome the error propagation and considerably improve the system performance.

On the Connectivity of Infrastructure-based Wireless Multi-hop Networks

Speakers	Seh Chun Ng
Seminar Date	Monday, 24 August, 2009
Location	11am, Room 720, EE Building
Abstract	Many wireless multi-hop networks are deployed with some infrastructure support, where the results on ad-hoc networks cannot be readily extended to understand the properties of those networks. In this seminar, we discuss on the study of those networks in 1-D. Specifically, we consider two types of nodes in the networks: ordinary nodes are i.i.d and Poissonly distributed in a unit interval, powerful nodes are arbitrarily distributed within the same unit interval. These powerful nodes connect to each other via some backbone infrastructure. The network is said to be connected, i.e. any two nodes can communicate with each other, if each ordinary node is connected to at least one of the powerful nodes. We call this type of connectivity problem the type-II connectivity problem. Exact and simplified asymptotic formulas for network type-II connectivity probability and the average hop count between two arbitrary nodes are obtained. Further we show that the optimum powerful node distribution that maximizes the asymptotic connectivity probability is to deploy these powerful nodes in an equidistant fashion. These results are important for the design and deployment of 1-D infrastructure-based networks and provide useful insights into the analysis of higher dimensional networks.

Codes Design on Graph - From a centralized to a distributed way

Speakers	Kun Pang
Seminar Date	Monday, 17 August, 2009
Location	11am, Room 720, EE Building
Abstract	Codes on graphs attracted considerable attention owning to their capacity-approaching performance and low complexity iterative decoding. Prime examples are the low density parity check (LDPC) and LDPC-like codes. As a graphic representation of LDPC codes, Tanner graph can be used to represent the predefined network topology, the source nodes correspond to the variable nodes and the relay nodes represent the check nodes, and the edges to connect them can be assumed to be the transmission link in wireless network. As the real networks consist of randomly faded channels, the link failure and topology change are unavoidable. To match the instantaneous network topologies and to achieve excellent performance, the proper codes design on the graph of the network has to be considered. In this talk, we will first introduce the properties of LDPC codes and the ways to construct those. Some simulation results will be shown to discuss the trade-off between the performance and the complexity of codes design and construction. The second part of this talk will introduce the scheme of adaptive network coded cooperation (ANCC). Based on the idea of ANCC, we present a real network scenario; the graph will be constructed according to the log-likelihood ratios (LLR) selection at the relays. To solve the unselected variable nodes problem, we propose a novel method by using brute force cluster selection (BFCS). This makes sure that all the information from the sources will be used for decoding.

Dynamic Spectrum Sharing: Bargaining and Optimization Approach

Speakers	Siavash Bayat
Seminar Date	Monday, 3 August, 2009
Location	11am, Room 720, EE Building
Abstract	All wireless systems require spectrum to propagate radio waves, but interference will happen if radios in the same location simultaneously operate on the same band. Therefore, spectrum is a potentially scarce resource because spectrum is regulated so that most bands are allocated exclusively to a single system licensed to use that band in any given location. However, such static spectrum allocation policies lead to significant under use of spectrum. Thus new concepts such as sharing the spectrum have been important. And new researches in this area are extending. In this seminar, a fair scheme to allocate frequency, rate, and power for multi-user orthogonal multiple-access systems will be discussed. A group of Cognitive Radios access the spectrum of a primary system and are allowed to use the spectrum while there is not any problem for primary users QOS. It is assumed that the primary system is a cellular OFDM based network operating in uplink. The problem is to maximize the total system rate, under each user�s maximal available power and minimal rate constraints, while considering the fairness among users. The aim is to develop an optimum resource allocation method, using cooperative Game Theory, which does not interfere with the primary users and does not destroy the QOS of these users. The proposed Cognitive Radio Game is a network-assisted resource management and there is a base station which collects the information such as channel state information of both primary and the secondary users and manages the system. Some optimization methods which will be used in the future works also will be reviewed.

Interference Cancellation in Two Hop Cognitive Radio Network

Speakers	Raed Manna
Seminar Date	Monday, 27 July, 2009
Location	11am, Room 720, EE Building
Abstract	Cognitive radio has been recently proposed as a promising technology to improve the spectrum utilization by allowing unlicensed users to coexist in licensed bands. However, the interference caused by sharing the same radio channel becomes an obstacle that limits the system performance, such as the system throughput. Hence, interference suppression and cancellation schemes are seen as key technologies for enabling coexisting systems, and the application of multiple antennas might be one solution to tackle interference. In this research, we have considered a two hop cognitive radio (CR) network scenario operating in a frequency band allocated to N primary users (PUs). In this CR network a secondary users� base station is transmitting in the same frequency band and time slot to K secondary users (SUs) through a fixed relay. We assume that the secondary users� base station and the relay each have multiple antennas but that the secondary users have only a single receive antenna. We have applied multi-user MIMO interference cancellation schemes based on zero-forcing (ZF) schemes in order to cancel the interference among the SUs based on fairness criteria as well as to cancel the interference received by each PU. In addition, we derive the analytical expression for the SINR at each SU which has been verified through simulations

Interference cancellation in two-hop MIMO multi-user networks using beamforming and precoding

Speakers	Neda Aboutorab
Seminar Date	Monday, 13 July, 2009
Location	11am, Room 720, EE Building
Abstract	The very high data rates envisioned for next generation wireless systems in reasonably large or dense areas do not appear to be feasible with conventional cellular architectures. Even with the advanced techniques employed in such networks (e.g., MIMO and interference cancellation algorithms), the target data rates will require a SINR ratio at the receiver that may be difficult to obtain at the cell edge and even using these techniques by themselves cannot provide enough leverage to achieve the desired level of performance. In this research, we have considered a multi-user two-hop MIMO wireless network. In our scenario several base stations will cooperate to send the data in the same frequency band and time slot to several users in a wireless network using a relay station. All nodes are considered to be equipped with multiple antennas and we have applied a cooperative downlink transmission scheme employing precoding and beamforming that eliminates the interference and improves the performance of the system. We have applied multi-user MIMO interference cancellation schemes based on zero-forcing (ZF) and Tomlinson Harashima Precoding (THP) methods as the combination of these two approaches seems to offer the most practical solution for the present practical systems.

Relaying Protocols for Cooperative Communication Networks

Speakers	Qingfeng Zhou
Seminar Date	Monday, 29 June, 2009
Location	11am, Room 720, EE Building
Abstract	I will begin this talk with a short summary of the concepts of channel, achievable rate and channel capacity. Then I will introduce the classic information-theoretical model of a single-relay cooperative network, and recall the three prevailing relaying protocols, namely amplify-and-forward (AF), decode-and-forward (DF) and incremental amplify-and-forward (IAF) protocols. Following will be the highlight of this talk, the presentation of two new-proposed incremental protocols, which are incremental selection amplify-and-forward (ISAF) and joint incremental selection relaying (JISR) protocols. At the second part of this talk, I will discuss the cooperative networks with multiple relays, mainly elaborating the networks employing opportunistic relaying technique. In the opportunistic relaying mode, an opportunistic relay selection algorithm runs once every channel coherence time period to select the "best" relay among the available relays. The selected "best" relay then forwards the received information to the destination while the other relays remain idle. I will analyze the asymptotic outage performances of IAF, ISAF and JISR when they are employed to the opportunistic relaying scenario. Comparison of the analytical asymptotic outage probabilities and the simulated ones will also be presented and discussed.

Active interference cancellation technology for wireless communication system

Speakers	Bin Guo
Seminar Date	Monday, 1 June, 2009
Location	11am, Room 720, EE Building
Abstract	Co-channel interference exists among the co-working base stations (BS) in the same location and reduces the throughput of the co-working wireless local area networks (WLANs). We propose a novel system to cancel this interference by using the concepts of common relay base station (RBS). RBS is connected with all the base stations (BS) and we assumed that it knows all channel state information (CSI) as well as the transmitted symbols of both base stations. Linear transmit processing are considered at the RBS, which are known as zero-forcing (ZF) and minimum mean-square-error (MMSE). The proposed structure can be directly applied to existing cellular networks, WLAN or WiMax, without changing the transmitter and receiver structure of existing base stations (BS) and mobile stations (MS). The interference and precoding will be done only at the relay BS. Therefore, the main advantage of our scheme is that it can remove the interference without redesign the transceivers of existing BS and MS. This will significantly reduce the system implementation complexity and shorten the deployment period. Simulation results show that our scheme can achieve the good performance and is less than 5 dB away from the interference-free channel under the same condition.

Design of Smart Metering network and Home Automation using ZigBee technology

Speakers	David Qi
Seminar Date	Monday, 25 May, 2009
Location	11am, Room 720, EE Building
Abstract	Wireless sensor networks are an emerging technology for low-cost, unattended monitoring of a wide range of environments. Taking advantage of the IEEE 802.15.4 standard for the physical and MAC layers, wireless sensor network has been enforced by the recent delivery of the ZigBee standard which contains the advantages of cheapness, small size, low-power consumption, licenses free radio bands operation, flexible and extendable network and integrated intelligence for network configuration and message routing. Thus, a ZigBee smart metering and home automation system is considered the competent solution to innovate the problems existed in the manual traditional expensive meter reading. In order to extend the coverage rang from every home side to city central server, the system structure employs distributed structure based on the ZigBee network, WiMax network and the central server. The innovations are not only introducing WiMax network coexisted with ZigBee network to link the separated households� local area network to the whole system, but also providing a compatible wireless platform to implement both automatic meter reading (AMR) and smart home purposes. The particular implementation is the subsystem of ZigBee network, which consists of ZigBee metering nodes, ZigBee coordinator, ZigBee remote hand held device, ZigBee smart appliance. Furthermore, the start network and multi cluster tree network topology structures were adopted in designing system.

Cooperative Selection Diversity with Amplify-and-Forward Relays

Speakers	Phil Yeoh
Seminar Date	Monday, 18 May, 2009
Location	11am, Room 720, EE Building
Abstract	In wireless network deployments which make use of relay nodes, it is often desirable to provide multiple relays with overlapping coverage. For such networks, we consider a cooperative selection diversity (CSD) scheme for selecting the strongest diversity path amongst N relay paths and the direct path. We derive simple closed-form expressions for a lower bound on the outage probability and symbol error rate (SER). Our analytic results are shown to closely approximate the exact system performance at medium to high signal-to-noise ratios (SNRs). In addition, we derive asymptotic expressions which indicate that a full diversity order of N+1 is achieved. Numerical examples are provided to substantiate our analysis.

Physical Layer Network Coding in Two-Way Relay Channels

Speakers	Raymond Louie
Seminar Date	Monday, 11 May, 2009
Location	11am, Room 720, EE Building
Abstract	This presentation investigates the performance of practical physical layer network coding (PLNC) schemes for two-way relay channels. We consider a network consisting of two source nodes and a single relay node, which is used to aid communication between the two source nodes. For this scenario, we investigate various transmission schemes, where transmission takes place over two, three or four time slots. We show that the two time slot PLNC scheme offers a higher maximum sum-rate, but a lower sum-bit error rate (BER) than the four time slot transmission scheme for a number of practical scenarios. We also investigate a three time slot PLNC scheme, which we show offers a good compromise between the two time slot PLNC and four time slot transmission schemes, and also achieves the best maximum sumrate and/or sum-BER in certain practical scenarios. To facilitate comparison, we derive new closed-form expressions for the outage probability, maximum sum-rate and sum-BER.

The Energy Optimizing Transmission Range in Wireless Sensor Networks

Speakers	Zijie Zhang (Jeffrey)
Seminar Date	Monday, 27 April, 2009
Location	11am, Room 720, EE Building
Abstract	Minimizing energy consumption is one of the major considerations in the design of battery powered Wireless Sensor Networks (WSNs). In many scenarios it is difficult to change or recharge a battery for the sensors. The problem of minimizing energy consumption in WSNs has attracted many researchers, and various approaches for reducing energy consumption have been proposed. We consider the optimum transmission range (power) that minimizes the energy consumed in end-to-end packet transmissions. Most of the research on this topic assume that the network is connected, i.e. there is at least one path between any pair of nodes. Different from previous work, we consider the possibility that a transmission is unsuccessful if the packet sent from a source to a destination has to be dropped at an intermediate node because the greedy forwarding protocol is unable to find a next-hop node. We study the average number of hops between any two nodes and show that there exists an optimum transmission range which minimizes the effective energy consumption. Finally, some ideas about the future work will be introduced.

A Low Complexity Limited Feedback Scheme in MIMO Broadcast Channels

Speakers	Lei Lei Wu
Seminar Date	Monday, 20 April, 2009
Location	11am, Room 720, EE Building
Abstract	Multiple-input multiple-output (MIMO) technologies have the advantage of improving system throughput and error performance by exploiting the spatial diversity, owing to the multiple independent transmit-receive paths. A lot of efforts have been put into MIMO broadcast channels research and it is expected to be a core technology for 3G and 4G wireless systems. Generally, channel state information (CSI) is required at the transmitter, to fully exploit the diversity. In the case that channel reciprocity does not apply, limited feedback is one effective solution to get CSI at the transmitter but it increases system complexity and degrades the performance. We propose a low complexity limited feedback scheme for MIMO broadcast channels. Non-cooperative receivers without full CSI and a feedback link with a limited rate are considered. The proposed scheme is based on a novel Grassmannian-based two-layer codebook, each layer having a small dimension, and a vector selection algorithm, to maximize the effective channel gain. A simulation is presented for the proposed scheme, showing that the performance is very close to the best known conventional high dimensional codebook, while the search complexity and memory requirement are reduced by several orders of magnitude.

A performance study of uplink scheduling algorithms in WiMax networks with high accuracy physical layer

Speakers	Xinyi Li
Seminar Date	Monday, 30 March, 2009
Location	11am, Room 720, EE Building
Abstract	As the demand for high speed wireless broadband services increases rapidly, the IEEE 802.16 standard defines a wireless broadband access technology called WiMax. The specifications for medium access control (MAC) and physical (PHY) layers are defined in the standard. A critical part of the MAC layer specification is packet scheduling, which resolves the bandwidth allocation and service order of the subscriber stations. Although the 802.16 standard suggests the main principles in designing QoS architecture to support different types of broadband services, scheduling algorithms for uplink and downlink traffics are left to researchers and manufacturers to design and implement. Evaluating the performance of scheduling algorithms under a realistic channel condition thus becomes one of the most important tasks in realization of large scale WiMax networks. This project aims to conduct a comprehensive performance study in point to multipoint, OFDM-based WiMax networks, under a high accuracy channel propagation model based on ray tracing algorithm. A number of well known scheduling algorithms are modified and implemented in NS2 for this study, and performances are evaluated in terms of system throughput, ability to meet QoS requirements for each type of service and fairness to different service classes.

Type-II Connectivity of 1-D Wireless Multi-hop Networks

Speakers	Seh Chun Ng
Seminar Date	Monday, 16 March, 2009
Location	11am, Room 720, EE Building
Abstract	The connectivity of wireless ad hoc networks has been actively studied in recent years. A well known result of asymptotic connectivity for two-dimensional networks were reported by Gupta and Kumar in their paper in 1998. There are many other papers presented the analysis on the connectivity of one or higher dimensional networks, either analytically or empirically. The network models used in these work have one configuration in common, that is the nodes in the network are assumed to be identical in transmission capabilities. The critical requirement for these networks is all nodes should be connected into a single component. What if we only require every node connects to any one of some "powerful" nodes? The real applications for this type of networks include the vehicular networks formed by the vehicles and infrastructure along the road, the cellular networks formed by the mobile units with multi-hop functions and base-stations. In this presentation, we present the exact formula on the connectivity for this type of hybrid networks, namely the "Type-II" connectivity, as well as some other interesting properties such as average number of clusters and average hop count. The current work only focus on one-dimensional hybrid wireless networks, and shall be extended to two-dimensional networks in the near future.

Connectivity of Wireless Multi-hop Networks with Arbitrary Wireless Channel Models

Speakers	Jason Ta
Seminar Date	Monday, 2 March, 2009
Location	11am, Room 720, EE Building
Abstract	Connectivity is one of the most fundamental properties of wireless multi-hop networks, e.g. wireless ad hoc/sensor networks. There has been extensive research and outcome on network connectivity in the last several years. So far, almost all the results in the literature were derived under two widely used wireless channel models, viz. the unit disk communication model and the log-normal shadowing model. It can be shown that there are many similarities between the results derived under these two models, which raises an interesting question: whether some well-known results derived under these two models hold under more generic channel models. In reality, the results obtained under a specific channel model may not necessarily apply for other channel models or scenarios. Hence, it is important to investigate whether these results are applicable for more generic channel models satisfying certain reasonable conditions. This talk will focus on one widely studied problem, i.e. the probability of connectedness, under an arbitrary channel model.

Multiuser MIMO Systems with Quantized Feedback

Speakers	Prof. Nevio Benvenuto
Seminar Date	Thursday, 27 November, 2008
Location	11am, Rm. 952, EE Building
Abstract	This presentation deals with multiuser wireless communication systems, namely, the multiple-input multiple-output (MIMO) broadcast channel (BC) where a multi-antenna transmitter serves multiple users along spatially multiplexed channels. In MIMO BC, channel state information at transmitter (CSIT) is essential to achieve spatial multiplexing across users. Special interest is on frequency division duplexing systems where CSIT is provided through limited uplink feedback (FB) from the receivers. Either in case of single antenna or multi-antenna receivers the main contributions are: i) the design of novel linear transceiver strategies that account for limited CSIT, ii) the proposal of channel quantization techniques and FB strategies that exploit spatial and time correlation of the MIMO channel and iii) the derivation of efficient and robust user selection schemes for the maximization of the achievable throughput. In case of multiuser MIMO orthogonal frequency division multiplexing (OFDM) downlink systems we provide non trivial generalizations of channel quantization strategies proposed for single carrier flat fading systems. Interestingly, concentrating FB bits to characterize only a portion of the available bandwidth at receivers and the possibility of exploiting multiuser diversity can increase significantly the achievable throughput.

Performance Analysis of Zero Forcing in General Two Hop Relay Networks with Multiple Source and Destination Nodes

Speakers	Raymond Louie
Seminar Date	Monday, 17 November, 2008
Location	11 am, Room 720, EE Building
Abstract	In this presentation, we consider a two hop relay network with multiple source and destination nodes. In particular, we investigate systems with various processing capabilities at the source, relay and destination nodes. We derive new exact outage probability expressions when zero forcing processing is used at the i) source and relay, ii) relay, iii) relay and destination and iv) destination nodes only. For the schemes considered, we show significant decreases in outage probability for certain node configurations.

Design and implementation of WiMax module in NS2 with high-accuracy channel propagation model

Speakers	Xinyi Li
Seminar Date	Monday, 20 October, 2008
Location	11am Room 720, EE Building
Abstract	As the demand for high speed broadband wireless systems increases rapidly, the IEEE 802.16 standard defines a wireless Broadband access technology called WiMax. As for every new technology, studies based on simulations are required in order to estimate the performance of the system by different parameters. Network Simulator 2 (NS2) is one of the most popular simulators for wireless networks, which is widely used in both industry and academy. However, there is not much contribution for WiMax module in NS2. Furthermore, NS2 only contains three simple channel propagation models, which is based on the distance threshold and the coverage is determined by a constant radius, it does not contain link parameters such as SNR and PER. In this project, we propose a QoS included WiMax module in NS2 which is able to simulate five different scheduling classes with a number of scheduling algorithms, together with a high-accuracy channel propagation model based on ray tracing is included.

Research Methodology - A Personal View

Speakers	Professor Branka Vucetic
Seminar Date	Monday, 13 October, 2008
Location	Level 9, Sky Room, EE Building
Abstract	The lecture will give an introduction to standard research methodologies in electrical engineering. The presenter will give a personal view on required attributes for research, steps in a research process and nature of creativity. A special focus will be on recent education and psychology views on nurturing and stimulating creativity. Communication and commercialization of research results will also be addressed.

Self-Similarity in Network Traffic and Traffic Prediction Methods

Speakers	Yanqiang Luan
Seminar Date	Monday, 25 October, 2004
Location	11.00am, Room 720, Electrical Engineering Building (J03), Maze Crescent, Darlington Campus
Abstract	Recent measurements and simulation studies have revealed that wide area network traffic has complex multifractal characteristics on small timescales, and is self-similar on large timescales. Conventional prediction methods such as LMS, RLS and Kalman Filter are incapable to capture the multiscale behavior of network traffic. Some new methods include: FARIMA (fractional autoregressive integrated moving average) model , neural network approach and multiresolution analysis method. In out research, wavelet transform is used as a mathemetical tools for multiresolution analysis. More accurate traffic prediction can be achieved by combined traditional prediction algorithm with wavelet transform.

Enhancement to IEEE 802.11e EDCA with QoS Differentiation Based on Station Priority

Speakers	Lixiang Xiong
Seminar Date	Monday, 18 October, 2004
Location	11.00am, Room 720, Electrical Engineering Building (J03), Maze Crescent, Darlington Campus
Abstract	The seamless integration of WLAN and 3G cellular network is one of the widely accepted visions for future 4G network, therefore the QoS consistency between WLAN and 3G cellular network should be considered. QoS differentiation based on station priority has been considered in cellular network. However, it is not defined in the current draft of IEEE 802.11e MAC protocol yet. In this presentation, I will propose several approaches to enhance IEEE 802.11e Enhanced Distribution Channel Access (EDCA) with QoS differentiation based on station priority.

A network Selection Mechanism for the integrated 3G/Wireless LAN systems

Speakers	Lily Song
Seminar Date	Monday, 20 September, 2004
Location	11.00am, Room 720, Electrical Engineering Building (J03), Maze Crescent, Darlington Campus
Abstract	Abstract: 3G cellular systems provide users high mobility with low bandwidth. Wireless LAN is adopted as a complementary service provider to support high bandwidth in some hot spots. In this seminar, a network selection mechanism for the integrated 3G/wireless LAN networks will be presented. The proposed network selection combines Analytic Hierarchy Process and Grey Relational Analysis. The selection procedures are presented and evaluated in three simulation cases. The results show that the proposed scheme can effectively choose the optimum network for users through balancing between objective network conditions and subjective user requirements.

Adaptive LS Receivers for CDMA and MIMO

Speakers	Matthew Peacock
Seminar Date	Monday, 13 September, 2004
Location	11.00am, Room 720, Electrical Engineering Building (J03), Maze Crescent, Darlington Campus
Abstract	Recent results on the large-system performance of adaptive least-squares (ALS) receivers for CDMA and MIMO will be presented. This extends previous analysis of MMSE receivers, which have full knowledge of the channel and spreading signatures. In contrast, an ALS receiver uses training symbols to approximate the autocorrelation matrix of the received signal with a sample autocorrelation matrix. This analysis provides insight into the effect of the number of training symbols used to train the receiver, the effect of the diagonal loading constant which is used to prevent ill-conditioning of the sample autocorrelation matrix, and exponential weighting schemes (such as recursive LS).

An Adaptive Iterative Receiver for Coded CDMA System in MIMO Fast Fading Channel

Speakers	Chakree Teekapakvisit
Seminar Date	Monday, 6 September, 2004
Location	11.00am, Room 720, Electrical Engineering Building (J03), Maze Crescent, Darlington Campus
Abstract	In this seminar, an adaptive iterative receiver structure for decoding multiuser information data in a convolutionally coded synchronous DS-CDMA multi-input multi-output (MIMO) system is presented. The receiver performs two successive soft-output decision, achieved by an adaptive soft-input soft-output multiuser detector and a bank of of single-user soft-input soft-output channel decoders through an iterative process. The proposed receiver, based on a combining of a normalized LMS algorithm and a Maximum A Posteriori (MAP) algorithm, adaptively suppresses and cancels co-channel interference. Simulation results in fast-fading channel show that the performance of the proposed receiver approaches the single-user performance. The computation complexity of the proposed receiver is also compared to an iterative MMSE receiver.

Source Coding for Broadcast Applications

Speakers	J. K. Wolf
Seminar Date	Friday, 3 September, 2004
Location	4pm, University of Sydney, Electrical Engineering Building, Level 9, Sky Room. (Main Campus, Darlington)
Abstract	RSYNC is a method of efficiently transmitting the contents of one computer to a second computer which contains an older but strongly correlated version of this information. The efficiency in transmission is achieved by the first computer sending only the difference between its information and the information stored in the second computer. Upon receiving the difference, the second computer can update its information. To make RSYNC work, however, the second computer must first send enough information to the first computer so that it can compute the required difference. In this paper we generalize this problem to the case where we have one computer broadcasting information to several different computers each of which contains different versions of the information stored in the first computer. We will further assume a knowledge of the joint statistics of the information in all of the computers. Based upon this knowledge we will describe a very efficient broadcast strategy for updating all of the other computers. Biography Jack Keil Wolf received his undergraduate degree from the University of Pennsylvania and his graduate degrees from Princeton. He has taught at several universities. Presently he is on the faculty of the University of California, San Diego and a member of the Center for Magnetic Recording Research. He also works part-time at Qualcomm.

Adaptive Modulation Design Based on Imperfect CSI in MIMO Systems

Speakers	Zhendong (Kyle) Zhou
Seminar Date	Monday, 30 August, 2004
Location	11.00am, Room 720, Electrical Engineering Building (J03), Maze Crescent, Darlington Campus
Abstract	The design of adaptive coded/uncoded modulation for MIMO systems under imperfect channel state information (CSI) is addressed. Firstly the impact of imperfect CSI on an adaptive modulation MIMO system designed under perfect CSI assumption is investigated, which gives the insight to the design under imperfect CSI condition. The system is simulated to verify the effectiveness of the design to trade off between the spectral efficiency and bit error rate performance.

An Iterative Channel Estimation Scheme for Beamforming Transmission and Detection in MIMO Systems

Speakers	Yang Tang
Seminar Date	Monday, 23 August, 2004
Location	11.00am, Room 720, Electrical Engineering Building (J03), Maze Crescent, Darlington Campus
Abstract	A novel iterative singular vector estimation scheme has been proposed for a beamforming transmission and detection in wireless multiple input multiple output (MIMO) systems. Two singular channel matrix vectors, which correspond to the largest singular value, are iteratively obtained at the transmitter and the receiver without estimating the channel coefficients. The proposed singular vectors estimation strategy has advantages over the conventional MIMO channel estimation schemes in terms of both bit-error-rate performance and bandwidth efficiency.

ECDH Exchange for Wireless LAN Security

Speakers	Mohammad Abdul Azim
Seminar Date	Monday, 9 August, 2004
Location	11.00am, Room 720, Electrical Engineering Building (J03), Maze Crescent, Darlington Campus
Abstract	WEP (Wired Equivalent Privacy), the security technique for WLAN (Wireless LAN), largely fails to provide security requirements because it lacks a robust mutual authentication mechanism for STAs and APs, an automated key exchange mechanism for RC4 algorithm and a keyed hash function for message integrity check. To resolve the wireless LAN security problems we propose a model using public key cryptography for the authentication and key exchange and privet key cryptography for the data encryption and integrity check. In our proposed representation an AKA (Authenticated Key Agreement) based on ECDH (Elliptic Curve Diffie-Hellman) is used for the automated key exchange and mutual authentication function. Two different versions of the exchange will be required for the BSS (Basic Service Set) network and IBSS (Independent Basic Service Set) network. We also propose a combination of authentication protocol SOLA (Statistical One bit Lightweight Authentication) and AES-CCM mode for the synchronization, encryption and integrity check. The combination will give us better security than SOLA and will free 47 bits from the CCMP header.

Call admission control for self similar traffic using shared channels

Speakers	Robert Fry
Seminar Date	Monday, 22 March, 2004
Location	Room 720, Electrical Engineering Building (J03), Maze Crescent, Darlington Campus at 11am
Abstract	Time: 11:00am The increasing presence of self-similar and real-time traffic types in wireless cellular networks places a large demand on the limited bandwidth available over the wireless spectrum. This is due in part to the fact that in order to meet the stringent QoS requirements of real time traffic, bandwidth is assigned to the user for the duration of the call. This means that for a significant proportion of the duration of the call bandwidth is available that is not being used. We look at call admission control and traffic scheduling for real time and self similar traffic transmitted over shared channels, in order to improve the bandwidth utilization whilst maintaining QoS requirements.

An Overview of Asymptotic Analysis of Random Matrices in Communication Systems

Speakers	Mat Peacock
Seminar Date	Monday, 31 March, 2003
Location	Telecomms Lab Seminar Room, Level 7, J03
Abstract	The application of random matrix theory to is a powerful technique in analysing communication systems. First used in 1999 to analyse performance of DS-CDMA with random spreading by Tse & Hanly and Verdu & Shamai, random matrix techniques have recently begun to be used in characterising MIMO channels, convergence of adaptive tehniques, multi-signature MC-CDMA systems, and space time codes. This talk will be an introduction to the theory and application of random matrix methods, and provide an overview of the insight provided by such techniques, including recent results in our current research on multi-signature MC-CDMA.


	© 2002-2011 The University of Sydney, NSW 2006 Australia. Phone: +61 2 9351 2222. ABN: 15 211 513 464. CRICOS Number: 00026A. Authorised by: Head of School. Our People \| School Contacts \| Sitemap \| Jobs \| Library \| Disclaimer \| Privacy Statement