In this paper, we study the feature of Multi-Flow HSDPA (MF-HSDPA), which allows simultaneous transmission of separate data packets from two or more serving cells (across different sectors or NodeBs) to a UE. This new feature brings challenges for system design, especially for the Inter-NodeB case, in which the serving cells reside in different NodeBs. To address these challenges, we propose enhancements to the Radio Link Control (RLC) layer and Media Access Control (MAC) layer. Simulation results show significant performance gain can be achieved by the MF-HSDPA feature.

In this paper, we study the feature of Multi-Flow HSDPA (MF-HSDPA), which allows simultaneous transmission of separate data packets from two or more serving cells (across different sectors or NodeBs) to a User Equipment (UE). This new feature brings challenges for system design, especially for the Inter-NodeB case, in which the serving cells reside in different NodeBs. To address these challenges, we propose enhancements to the Radio Link Control (RLC) layer and backhaul flow control/congestion control. Simulation results show significant performance gain can be achieved by the MF-HSDPA feature. The gain is more promising when the backhaul is congested, because Inter-NodeB MF-HSDPA offers "backhaul diversity" by utilizing two backhaul links to serve a MF UE.

In this paper, we will study the scheme of Single-Frequency Dual-Cell HSDPA (SF-DC HSDPA) which is introduced to simultaneously transmit two separate packet streams from two cells to a mobile. The inter-stream interference is suppressed by the linear-MMSE equalizer at the mobile. In the simple implementation studied in this paper, the schedulers at the two cells are independent without information exchange. Our study is based on both analysis and simulations. SF-DC HSDPA offers significantly higher data rates to users in the handover region. It also achieves dynamic load balancing in terms of resource utilizations across cells. Furthermore, all these gains can be obtained without degrading the performance of users not in handover and the legacy users. The gain from SF-DC is particularly substantial for users in heavily loaded cells surrounded by lightly loaded neighboring cells. In addition, such gain can increase with loading at a given loading ratio between the heavily loaded and lightly loaded cells. SF-DC is easy to implement with only incremental complexity to the current mobile and network equipment.

Multiple Input Multiple Output (MIMO) transmission improves downlink spectral efficiency and is a key feature of Release 7 High Speed Downlink Packet Access (HSDPA). When the Node-B is configured in the MIMO mode with secondary common control pilot channel (S-CPICH), the power amplifiers (PAs) for the two transmit antennas are not balanced. A common pre-coder is required to balance the PAs for legacy and dual-stream MIMO transmissions. In addition, codebook restriction is needed to balance the PAs for single-stream MIMO transmissions. Despite codebook restriction, MIMO performance degradation is small. Legacy user performance in a MIMO cell suffers due to power and interference penalties associated with the Secondary Common Pilot Channel (S-CPICH). By reducing the S-CPICH transmit power, this degradation can be reduced, without significantly compromising the channel estimation quality for MIMO users in the cell.

In this paper. we consider an admission control algorithm for cellular networks. It is based on direct/dynamic monitoring of the QoS performances - the delay tail of the admitted users. The main purpose of the direct monitoring of QoS performances is to guarantee the required QoS provision more precisely. The system delay tail is effectively computed by the proposed delay tail estimation algorithm. With the estimated delay tails, each base station determines the admission or rejection for a new user to the system. By doing so, the admission control algorithm becomes more robust against time-varying fading channel environment. Simulation results in OFDMA cellular system shows that the proposed algorithm can provide substantial capacity gains in offering QoS. We also use a general Neyman-Pearson-like framework in evaluating various admission control mechanisms.

The newly standardized Dual-Carrier HSDPA offers two-carrier aggregation on the downlink. In this paper, we investigate the performance gains from the dual-carrier operations with both full buffer and bursty traffic models. With full buffer traffic, dual-carrier offers dynamic load balancing and increase in user data rate, system capacity and spectral efficiency. All the users throughout the system enjoy the data rate increase. With bursty traffic, dual-carrier provides substantial gain in user experience measured by burst rates or latency under all loading conditions. Moreover, the gain with bursty traffic is universal not only for all the users but also for a very wide range of scheduling algorithms. Our results are based on analysis and extensive simulations. Design issues are also addressed in this paper.

This paper discusses several important aspects in the HSUPA scheduling algorithms. First, it clearly demonstrates the benefit of explicitly utilizing the directly measured Rise-over- Thermal(RoT) values in term of capacity and robustness improvement. This paper also briefly presents the RoT measurement schemes designed specifically for the asynchronous systems such as WCDMA/HSUPA. If pilot and traffic interference cancellation (IC) schemes are implemented at the Node B receiver, this paper presents a scheduling algorithm that attains IC benefit in such a way that the capacity enhancement is maximized while fairness and link budget are maintained. The presented algorithm defines the effective RoT by monitoring the IC performance in real time and adjusts its grants accordingly. In addition, the user priority function used here optimizes the sum utility functions and provides multi-user diversity.

The optimal resource allocation policy is studied for non-real-time users in CDMA reverse link. The resource allocation policy of interest includes channel coding, spreading gain control and power allocation under the conventional receiver operation. The constraints in the optimization include peak transmit power of the mobile station, total received power at the base station and QoS in the form of minimum SINR for each user. The coding and spreading gain control can be separated from the power allocation strategy. Our results show that the optimal power allocation policy depends on the objective function: a greedy policy is optimal to maximize the sum of throughput from each user, whereas a fair policy is optimal to maximize the product of throughput from each user. A unified approach is taken to derive the optimal policies, and it can also be applied to other power allocation problems in CDMA reverse link. Numerical results of the channel capacity are presented for both objectives along with the effect of QoS constraints.

The optimal resource allocation policy is studied in this paper, for nonreal-time users in CDMA reverse link. The resource allocation policy of interest includes channel coding, spreading gain control and power allocation under the conventional receiver operation-multiuser detection technique is not considered. The constraints in the optimization include peak transmit power of the mobile station, total received power at the base station and QoS in the form of minimum SINR for each user. The coding and spreading gain control can be separated from the power allocation strategy. Our results show that the optimal power allocation policy depends on the objective function: to maximize the sum of throughput from each user, a greedy policy is optimal where the good users (mobile stations with good radio conditions) transmit with full power, bad users (mobile stations with poor radio conditions) barely achieve their QoS and there is at most one user in-between; on the other hand, to maximize the product of throughput from each user, a fair policy is optimal where the bad users transmit with full power and the good users transmit with powers to reach a common received power level. The fairness of the optimal policy in the latter case makes QoS constraint unnecessary. Furthermore, we have found a unified approach in deriving these optimal policies. This approach can also be applied to other power allocation problems in CDMA reverse link. We also present numerical results on the channel capacity under both objectives and the effect of QoS constraint.

This work presents a simple and robust mechanism, called change-point monitoring (CPM), to detect denial of service (DoS) attacks. The core of CPM is based on the inherent network protocol behavior and is an instance of the sequential change point detection. To make the detection mechanism insensitive to sites and traffic patterns, a nonparametric cumulative sum (CUSUM) method is applied, thus making the detection mechanism robust, more generally applicable, and its deployment much easier. CPM does not require per-flow state information and only introduces a few variables to record the protocol behaviors. The statelessness and low computation overhead of CPM make itself immune to any flooding attacks. As a case study, the efficacy of CPM is evaluated by detecting a SYN flooding attack - the most common DoS attack. The evaluation results show that CPM has short detection latency and high detection accuracy.

An algorithm for generating sample paths of linear fractional stable motion (LFSM) is introduced. It is based on the approximation of LFSM by a linear process and exhibits low computational complexity. A detailed analysis of the error term involved in the approximation is provided, which in turn guides the user on selecting the size of the generated sequence.

This paper addresses the problem of dynamic resource allocation in a multiservice direct-sequence code-division multiple-access (DS-CDMA) wireless network supporting real-time (RT) and nonreal-time (NRT) communication services. For RT users, a simple transmission power allocation strategy is assumed that maximizes the amount of capacity available to NRT users without violating quality of service requirements of RT users. For NRT users, a joint transmission power and spreading gain (transmission rate) allocation strategy, obtained via the solution of a constrained optimization problem, is provided. The solution maximizes the aggregate NRT throughput, subject to peak transmission power constraints and the capacity constraint imposed by RT users. The optimization problem is solved in a closed form, and the resulting resource allocation strategy is simple to implement as a hybrid CDMA/time-division multiple-access strategy. Numerical results are presented showing that the optimal resource allocation strategy can offer substantial performance gains over other conventional resource allocation strategies for DS-CDMA networks.

Presents a simple and robust mechanism called SYN-dog to sniff SYN flooding sources. We install SYN-dog as a software agent at leaf routers that connect stub networks to the Internet. The statelessness and low computation overhead of SYN-dog make itself immune to any flooding attacks. The core mechanism of SYN-dog is based on the protocol behavior of TCP SYN-SYN/ACK pairs, and is an instance of the sequential change detection. To make SYN-dog insensitive to site and access pattern, a non-parametric cumulative sum (CUSUM) method is applied, thus making SYN-dog much more generally applicable and its deployment much easier. Due to its proximity to the flooding sources, SYN-dog can trace the flooding sources without resorting to expensive IP traceback.

We propose a simple and robust mechanism for detecting SYN flooding attacks. Instead of monitoring the ongoing traffic at the front end (like firewall or proxy) or a victim server itself, we detect the SYN flooding attacks at leaf routers that connect end hosts to the Internet. The simplicity of our detection mechanism lies in its statelessness and low computation overhead, which make the detection mechanism itself immune to flooding attacks. Our detection mechanism is based on the protocol behavior of TCP SYN-FIN (RST) pairs, and is an instance of the Seqnential Change Point Detection [l]. To make the detection mecbanism insensitive to site and access pattern, a non-parametric Cnmnlative Sum (CUSUM) method [4] is applied, thus making the detection mechanism much more generally applicable and its deployment much easier. The efficacy of this detection mechanism is validated by trace-driven simulations. The evaluation results show that the detection mechanism has short detection latency and high detection accuracy. Moreover, due to its proximity to the flooding sources, our mechanism not only sets alarms upon detection of ongoing SYN flooding attacks, but also reveals the location of the flooding sources without resorting to expensive IP traceback.

In this paper, we are interested in developing a control algorithm or transmission scheme for wireless data communication over time-varying channels with memory that determines when to attempt a transmission and at what power level, so as to achieve a suitable balance between throughput and energy consumption; and studying the effect of channel memory on the performance, design, and structure of this scheme. The channel state is not directly observable, and thus the transmission decisions must be based on partial or incomplete channel state information, provided to the user by the base station over a feedback channel. More specifically, we assume that the delayed channel state information is provided to the sender at the end of those time slots during which packet transmissions are made. We cast the problem as a partially observable Markov decision process. Despite the difficulty associated with the incomplete channel information, we obtain the optimal transmission scheme, which can be interpreted as a back-off rule: at the end of the transmission, depending on the channel quality during the past time slot, the transmission may be suspended for some time slots, which can be computed in advance. Numerical results are also provided.

We set up a general framework for evaluating the accuracy of the Markov model for the packet transmission over a wireless channel. Unlike most previous studies, a broad range of fading/shadowing models and coding/modulation schemes are included in our framework. We study the effect of both the packet length and mobile speed. Our analysis is based on the approximate normal distribution and the second order statistics in both the time and frequency domain. We make the generalized conclusion that the Markov model is most accurate when the mobile movement is either very fast (packets i.i.d.) or very slow (packets highly correlated). For the first time, we explicitly show that at intermediate speed when the correlation length in the fading is comparable with the packet length, the accuracy of the Markov model is highly dependent on the specific bit level model. This intermediate range changes with the packet length. A long packet will make this range narrow and occur at low speed.

We develop a control algorithm or transmission scheme for wireless data communication over time-varying channels with memory which determines when to attempt a transmission so as to achieve a suitable balance between throughput and energy consumption; and we study the effect of channel memory on the performance, design, and structure of this scheme. The channel state is not directly observable, and thus the transmission decisions must be based on ACK/NAK information provided over a feedback channel. We cast the problem as a partially observable Markov decision process, and derive the optimal transmission scheme. Numerical results are also provided comparing performance of the optimal scheme with some other existing mechanisms.