Journal of Communications and Networks

We present a prototype for a new architecture, MCN (multihop cellular network), implemented over a wireless LAN platform. MCN preserves the virtue of traditional single-hop cellular networks where the service infrastructure is constructed by many bases, but it also adds the flexibility of ad-hoc networks where wireless transfer through mobile stations in multiple hops is allowed. The MCN can reduce the number of required bases or improve the throughput performance. On IEEE 802.11 compliant wireless LAN products, a bridging protocol, our BMBP (base-driven multihop bridging protocol), runs between mobile stations and access points to build bridging tables. The demonstration shows that MCN is a feasible architecture for wireless LANs

Reliable data transmission over a wireless multi-hop network, called the ad hoc network, has proven to be non-trivial. The TCP (transmission control protocol), a widely used end-to-end reliable transport protocol in a wired network, is not entirely suitable when applied to a wireless ad hoc network due to TCP's congestion control schemes. In particular, the TCP at the source considers the network as congested when detecting packet losses or timeouts. However, in a wireless ad hoc network when a route disconnection occurs because of node movement, the network mistakes this as a congestion. Therefore, the conventional TCP congestion control mechanism cannot be applied, because a route disconnection must be handled differently from a network congestion. We propose a new mechanism that improves the TCP performance in a wireless ad hoc network where each node can buffer packets during a route disconnection and reestablishment. Additionally, we incorporate new measures to deal with the reliable transmission of important control messages. Our simulation results further confirmed these advantages

In an IP-based network, automated dynamic assignment of IP addresses is preferable. In most wired networks, a node relies on a centralized server by using dynamic host configuration protocol (DHCP) to obtain a dynamic IP address. However, the DHCP-based approach cannot be employed in a mobile ad hoc network (MANET) due to the uncertainty of any centralized DHCP server. That is, a MANET may become partitioned due to host mobility. Therefore, there is no guarantee to access a DHCP server. A general approach to address this issue is to allow a mobile host to pick a tentative address randomly, and then use duplicate address resolution (DAR) protocol to resolve any duplicate addresses. In this paper, an innovative distributed dynamic host configuration protocol designed to configure nodes in MANET is presented. The proposed protocol not only can detect the duplicate address, but also can resolve the problem caused by duplicate address. It shows that the proposed protocol works correctly and is more universal than earlier approaches. An enhanced version of DAR scheme is also proposed in this paper to solve the situation of duplicate MAC address. The new and innovative approach proposed in this paper can make the nodes in MANET provide services to other networks and avoid packets from being delivered to incorrect destinations.

In the near future packet networks should support applications which can not predict their traffic requirements in advance, but still have tight quality of service requirements, e.g., guaranteed bandwidth, jitter, and packet loss. These dynamic characteristics mean that the sources can be made to modify their data transfer rates according to network conditions. Depending on the customer's needs, network operator can differentiate incoming connections and handle those in the buffers and the interfaces in different ways. In this paper, dynamic QoS-aware scheduling algorithm is presented and investigated in the single node case. The purpose of the algorithm is — in addition to fair resource sharing to different types of traffic classes with different priorities — to maximize revenue of the service provider. It is derived from the linear type of revenue target function, and closed form globally optimal formula is presented. The method is computationally inexpensive, while still producing maximal revenue. Due to the simplicity of the algorithm, it can operate in the highly nonstationary environments. In addition, it is nonparametric and deterministic in the sense that it uses only the information about the number of users and their traffic classes, not about call density functions or duration distributions. Also, Call Admission Control (CAC) mechanism is used by hypothesis testing.

The BcN architecture of VPWS is introduced as a service model. The network elements forming the BcN data plane and a centralized controller forming the BcN control plane functionality are briefly described. A mapping architecture which transports the VPWS signal is presented. We examine the propagation of the maintenance signal under the network failure condition. The main purpose of this architecture is to find out the relationship between the maintenance signals in each layer. Finally, we briefly review the MPLS OAM ITU-T Rec. Y.1711 and Ethernet OAM being recommended in ITU-T Rec. Y.1731

The performance implications of retransmission diversity packet combining on the RLC (radio link control)/MAC (medium access control) layer and transport layer protocol performance are investigated for three different heuristic-based RLC/MAC layer access control schemes in a CDMA S-ALOHA network under frequency selective Rayleigh fading. The transport layer protocol implements a two-level error recovery mechanism for reliable data transmission. Two different transport layer timer control mechanisms are considered. Implications of some physical layer parameters on system performance are discussed. It is observed that, for two-level error recovery through a reliable transport protocol, the achieved throughput depends on the transport protocol timer control mechanism and a suitable mechanism can be identified for an underlying RLC/MAC layer access control scheme and a particular physical layer design

In this paper, we report our investigation into the effect of using fixed multibeam antennas on the performance of the slotted ALOHA protocol with capture in a mobile communications environment with Rayleigh and Log-normal fading. We consider the configuration where multiple receivers are present at each base station and calculate the capture probability, its asymptotic value as the number of colliding packets tends to infinity, and throughput when multiple receivers are used. The results demonstrate that by using fixed multibeam antennas, we can achieve higher performance in terms of capture probability and throughput when compared to a conventional antenna system using the slotted ALOHA protocol.

The regrowth of OQPSK power spectral sidelobes from AM/AM and AM/PM amplifier nonlinearity is analyzed. The time-domain expression for amplifier output shows how spectral regrowth will depend on the cubic coefficient of the Taylor's series of the amplifier nonlinearity as well as input amplitude ripple. Closed form spectrum calculations show that the spectral sidelobes produced by AM/PM take the same form as those produced by AM/AM. The rate of growth of AM/PM sidelobes is, however, not as great as for AM/AM.

We propose to use turbo codes for wireless communication systems with multiple transmit and receive antennas over Rayleigh fading channels. We show that a simple, arbitrarily picked, turbo coded modulation scheme with a sub-optimal decoding algorithm outperforms the space-time codes significantly. We present examples for both block and fast fading channels, and observe gains as high as 8 dB at a bit error rate of 10^-5 for large interleaver lengths, suitable for data communications. Furthermore, we show that, depending on the channel model, the turbo code block size can be chosen small enough to be suitable for speech applications, and still offer a significant performance improvement in terms of bit and frame error rates

We propose a parallel and distributed routing algorithm with a hierarchical connection management architecture for the ATM/B-ISDN transport networks. In the proposed routing algorithm, a hierarchical connection management architecture is assumed where each subnetwork has its own routing functions to find the shortest route for the request subnetwork connections, and the routing information is merged by the upper-level domain subnetwork to find shortest path in the domain. This subnetwork routing is performed in each subnetwork in hierarchy, providing maximized parallel and distributed processing capability. The proposed parallel and distributed routing algorithm can reduce the routing time in a large network, such as a public B-ISDN

We provide the cross-layer analysis of wireless TCP systems. The effect of error correlation on the behavior of link retransmission strategy and the end-to-end throughput of TCP layer are investigated. Based on the cross-layer analysis, a refinement of link layer protocol is proposed by consciously utilizing the information of channel correlations, which leads to the performance improvement of wireless TCP systems.

We propose a framework to study how to route packets efficiently in multipath communication networks. Two traffic congestion control techniques, namely, flow assignment and packet scheduling, have been investigated. The flow assignment mechanism defines an optimal splitting of data traffic on multiple disjoint paths. The resequencing delay and the usage of the resequencing buffer can be reduced significantly by properly scheduling the sending order of all packets, say, according to their expected arrival times at the destination. We consider a multiple-node M/M/1 tandem network with a delay line as the path model. When end-to-end path delays are all Gaussian distributed, our analytical results show that the techniques are very effective in reducing the average end-to-end path delay, the average packet resequencing delay, and the average resequencing buffer occupancy for various path configurations. These promising results can form a basis for designing future adaptive multipath protocols

In cellular mobile communications, how to achieve optimum system capacity with limited frequency spectrum is one of the main research issues. Many dynamic channel assignment (DCA) schemes have been proposed and studied to increase the capacity of cellular systems. Reuse partitioning (RP) is another technique to achieve higher capacity by reducing the overall reuse distance. In this paper, a new network-based DCA scheme with the use of RP technique is proposed, namely dynamic reuse partitioning with interference information (DRP-WI). The scheme aims to minimize the effect of assigned channels on the availability of channels for use in the interfering cells and to reduce their overall reuse distances. Simulation results have confirmed the effectiveness of DRP-WI scheme. Under both uniform and nonuniform traffic distributions, DRP-WI exhibits outstanding performance in improving the system capacity. It can provide about 100% capacity improvement as compared to conventional fixed channel assignment scheme.

The capacity and the interference statistics of the sectors of the cigar-shaped W-CDMA microcell are studied. A model of multimicrocells is used to analyze the uplink when the users are within an under-ground train. The microcells are assumed to exist in a long under-ground tunnel. The capacity and the interference statistics of the microcells are studied for different propagation exponent, different antenna side lobe levels and different bends loss. The capacity for the best case and worst case are given.

It is well known that MIMO systems offer the promise of achieving very high spectrum efficiencies (many tens of bits/Hz) in a mobile environment. The gains in MIMO capacity are sensitive to the presence of spatial and temporal correlation introduced by the radio environment. In this paper we examine how MIMO capacity is influenced by a number of factors, e.g.: a) temporal correlation, b) various combinations of low/high spatial correlations at either end, c) combined spatial and temporal correlations, In all cases we compare the channel capacity that would be achievable under independent fading. We investigate the behaviour of "capacity fades", examine how often the capacity experiences the fades, develop a method to determine level crossing rates and average data durations and relate these to antenna numbers.

We consider a power controlled CDMA system with N nodes and F flow types, and with the constraint that each node uses the same power level for all flows. For the uplink case with F=1, the optimum sequences to minimize the total power are found and proved. For the uplink problem with N=2, the necessary and sufficient condition to have a solution is found and proved, and the iterative algorithm to find the optimal solution is provided. For the uplink problem with arbitrary N the iterative algorithm in finding the optimal solution is provided and its convergence is proved. For the downlink case, the power assignment problem is solved and some properties of the optimum sequences are proved.

We propose a new channel reservation protocol using a counter for detection of a source conflict in a WDM single-hop network with non-equivalent propagation delay. A source conflict occurs when a source node has the right to transmit more than two messages to their destination nodes using different wavelengths in the same time slot. In the proposed protocol, a source node can detect a source conflict before the assignment of wavelengths by investigating information about the final message which has succeeded in reservation. We approximately analyze the throughput considering the effect of a source conflict. Also, we show by computer simulation that our proposed protocol can reduce mean message delay dramatically without degrading the throughput performance as the offered load becomes large

This paper deals with the problem of channel identification for single input multiple output (SIMO) slow fading channels using clustering algorithms. The received data vectors of the SIMO model are spread in clusters because of the AWGN. Each cluster is centered around the ideal channel output labels without noise. Starting from the Markov SIMO channel model, simultaneous maximum-likelihood estimation of the input vector and the channel coefficients reduces to one of obtaining the values of this pair that minimizes the sum of the Euclidean norms between the received and the estimated output vectors. The Viterbi algorithm can be used for this purpose provided the trellis diagram of the Markov model can be labeled with the noiseless channel outputs. The problem of identification of the ideal channel outputs, which is the focus of this paper, is then equivalent to designing a vector quantizer (VQ) from a training set corresponding to the observed noisy channel outputs. The Linde-Buzo-Gray (1980) type clustering algorithms could be used to obtain the noiseless channel output labels from the noisy received vectors. This paper looks at two critical issues with regards to the use of VQ for channel identification. The first has to deal with the applicability of this technique in general. We present theoretical results showing the conditions under which the technique may be applicable. The second aims at overcoming the codebook initialization problem by proposing a novel approach which attempts to make the first phase of the channel estimation faster than the classical codebook initialization methods

Space-time coding is a bandwidth and power efficient method of communication over fading channels that realizes the benefits of multiple transmit and receive antennas. This novel technique has attracted much attention. However, currently the only analytical guide to the performance of space-time codes is an upper bound which could be quite loose in many cases. In this paper, an exact pairwise error probability is derived for space-time codes operating over Rayleigh fading channels. Based on this expression, an analytical estimate for the bit error probability is computed, taking into account dominant error events. Simulation results indicate that the estimates are of high accuracy in a broad range of signal-to-noise ratios

Cooperative transmission protocols are always designed to reach the largest diversity gain and the largest network capacity simultaneously. The concept of diversity-multiplexing tradeoff (DMT) in MIMO systems put forward by Zheng and Tse has been extended to this field. However, the concept of multiplexing gain in DMT constrains a better understanding of the asymptotic interplay between transmission rate, frame error probability (FEP) and signal-to-noise ratio (SNR), and also fails to predict FEP curves accurately. Two improved methods are then put forward. One is by Narasimhan who proposes finite-SNR diversity-multiplexing gain tradeoff which gives a tighter lower bound of the FEP curves by applying nonlinear programming in MIMO systems, and the other is by Azarian and Gamal who propose a new rule called the throughput-reliability tradeoff (TRT) which avoids the limitation of the conception of multiplexing and elucidates the linearly asymptotic trends exhibited by the FEP curves in block-fading MIMO channels. The finite-SNR diversity-multiplexing gain tradeoff has already been applied to cooperative relay channels. However, this method is time-consuming in computation since nonlinear programming is used, especially in large networks. In this paper, we will use TRT rule to give the relationship between transmission rate, FEP and SNR in decode-and-forward (DF) cooperative protocols. We also exhibit the FEP curves predicted by TRT. To do this, We first propose a symbol based slotted decode-and-forward (SSDF) protocol as the infrastructure. Network information theory is also used to bound the capacity of the protocol.

In cognitive radio systems, unlicensed users can use the frequency bands when the licensed users are not present. Hence reliable detection of available spectrum is the foundation of cognitive radio technology. To ensure that the unimpaired operation of licensed users and to improve the spectrum sensing performance, a novel cooperative spectrum sensing algorithm based on credibility is proposed. In particular, the close-form expressions for the probability of the detection and the false-alarm is derived for the novel algorithm, and the expression for the average overhead used for cooperation is given in the performance analysis. The conclusion is proved by computer simulations.

Ultra-wideband (UWB) radios have attracted great interest for their potential application in short-range high-data- rate wireless communications. High received signal noise ratio and compliance with the FCC spectral mask call for judicious design of UWB pulse shapers. In this paper, even and odd order derivatives of Gaussian pulse are used respectively as base waveforms to produce two synthesized pulses. Our method can realize high efficiency of spectral utilization in terms of normalized effective signal power (NESP). The waveform design problem can be converted into linear programming problem, which can be efficiently solved. The waveform based on even order derivatives is orthogonal with the one based on odd order derivatives.

We propose a delayed multiple copy retransmission (DMCR) scheme for data communication in wireless networks, by which multiple copies of a lost link layer frame are retransmitted at the link layer one-by-one with a delay in between. The number of the copies gradually increases as the number of retransmissions increases. For implementing DMCR scheme in a typical mobile communication system, an interleaving scheme is also proposed. Moreover, a simplified method called polling is suggested when retransmission times are very limited. We compare our scheme with the previous non-delayed retransmission scheme on the performance of both channel capacity and total transmission time. Numerical results show that the DMCR can achieve higher performance. The effect of the delay time on end-to-end TCP throughput is investigated as well.

The layer 3 switch enables us to transmit IP datagrams using the cut-through technique. There are mainly two schemes of connection setup; one is the flow-driven case and another is the topology-driven one. In this paper, we analyze the cut-through rate, the datagram waiting time and the mis-ordered rate as performance measures of both cases and compare these performances. In the analysis, by using the interrupted Bernoulli process (IBP), we model the arrival process of the IP flow and the IP datagram from each source. Furthermore, we investigate the impacts of the arrival rate and the average IP flow length on performance

To provide network services consistently under various network failures, enterprise networks increasingly utilize path diversity through multi-homing. As a result, multi-homed non-transit autonomous systems (ASes) has surpassed the single-homed networks in number. In this paper, we address an inevitable problem that occurs when networks with multiple entry points deploy stateful inspection firewalls in their borders. In this paper, we formulate this phenomenon into a state-sharing problem among multiple firewalls under the asymmetric routing condition. To solve this problem, we propose a stateful inspection protocol that requires a very low processing and messaging overhead. Our protocol consists of the following two phases: 1) generation of a TCP SYN cookie marked with the firewall identification number upon a SYN packet arrival, and 2) state sharing triggered by a SYN/ACK packet arrival in the absence of the trail of its initial SYN packet. We demonstrate that our protocol is scalable, robust, and simple enough to be deployed for high speed networks. It also transparently works under any client-server configurations. Last but not the least, we present the experimental results through a prototype implementation.