By Topic

Communications, IEE Proceedings-

Issue 4 • Date Aug 2002

Filter Results

Displaying Results 1 - 9 of 9
  • Novel TCP congestion control scheme and its performance evaluation

    Page(s): 217 - 222
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (414 KB)  

    A novel self-tuning proportional and derivative (ST-PD) control based TCP congestion control scheme is proposed. The new scheme approaches the congestion control problem from a control-theoretical perspective and overcomes several Important limitations associated with existing TCP congestion control schemes, which are heuristic based. In the proposed scheme, a PD controller is employed to keep the buffer occupancy of the bottleneck node on the connection path at an ideal operating level, and it adjusts the TCP window accordingly. The control gains of the PD controller are tuned online by a fuzzy logic controller based on the perceived bandwidth-delay product of the TCP connection. This scheme gives ST-PD TCP several advantages over current TCP implementations. These include rapid response to bandwidth variations, insensitivity to buffer sizes, and significant improvement of TCP throughput over lossy links by decoupling congestion control and error control functions of TCP View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Least load dispatching algorithm for parallel Web server nodes

    Page(s): 223 - 226
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (364 KB)  

    A least load dispatching algorithm for distributing requests to parallel Web server nodes is described. In this algorithm, the load offered to a node by a request is estimated based on the expected transfer time of the corresponding reply through the Internet. This loading information is then used by the algorithm to identify the least load node of the Web site. By using this algorithm, each request will always be sent for service at the earliest possible time. Performance comparison using NASA and ClarkNet access logs between the proposed algorithm and commonly used dispatching algorithms is performed. The results show that the proposed algorithm gives 10% higher throughput than that of the commonly used random and round-robin dispatching algorithms View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Modified branch metric for turbo codes using PSA technique in Rician fading channels

    Page(s): 227 - 231
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (403 KB)  

    The authors study a log-maximum a posteriori (MAP) algorithm for use with turbo-coded systems incorporated with pilot-symbol-aided (PSA) transmission in the satellite mobile channels. The channel probability used in a conventional log-MAP algorithm assumes the channel is memoryless white Gaussian noise. If this same channel probability is used in the log-MAP algorithm for decoding turbo-coded signals with fading compensation in the satellite mobile channels, which are Rician-faded, performance degradation occurs. A new channel probability for use in the log-MAP algorithm to decode turbo-coded signals incorporated with PSA transmission on Rician fading channels is derived. Monte Carlo simulation results have shown that the proposed log-MAP algorithm can give substantial improvements in the bit-error-rate performances of a turbo-coded quarternary-phase-shift-keyed (QPSK) signal over the conventional log-MAP and other log-MAP algorithms View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Linear complexity of polyphase power residue sequences

    Page(s): 195 - 201
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (420 KB)  

    The well known family of binary Legendre or quadratic residue sequences can be generalised to the multiple-valued case by employing a polyphase representation. These p-phase sequences, with p prime, also have prime length L, and can be constructed from the index sequence of length L or, equivalently, from the cosets of pth power residues and non-residues modulo-L. The linear complexity of these polyphase sequences is derived and shown to fall into four classes depending on the value assigned to b0, the initial digit of the sequence, and on whether p belongs to the set of pth power residues or not. The characteristic polynomials of the linear feedback shift registers that generate these sequences are also derived View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Concatenated wavelet-based ring-TCM

    Page(s): 189 - 194
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (457 KB)  

    A new coded modulation technique for signal sets designed combining wavelet functions and M-quadrature amplitude modulation (MQAM) is presented. Trellis-coded modulation (TCM) is implemented using the ring of integers modulo-Q as the algebraic structure of the multilevel convolutional encoder. These schemes are suitable for adaptive time-frequency unequal error protection, multiple-access and multiplexing, as well as being effective replacements for existing schemes. Wavelet-based ring-TCM schemes and MQAM ring-TCM schemes are first presented, and then combined to constitute a concatenated ring-TCM scheme. The concatenated scheme shows a coding gain of around 8 dB over uncoded BPSK, and does not require iterative decoding View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fixed channel allocation scheme performance enhancement for cellular mobile systems

    Page(s): 232 - 236
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (433 KB)  

    An optimisation technique for fixed channel allocation (FCA) schemes proposed. This mechanism can respond to the variations in traffic to reassign channels to cells. Traffic variability can be automatically adapted after applying this mechanism to FCA schemes. It is unnecessary to handoff a call when its allocated channel is reassigned to other cells. Moreover, freedom from deadlock and starvation is ensured. Simulation results demonstrate that this mechanism enhances the traffic-carrying capacity of FCA schemes with the flexibility and adaptability to handle variations in traffic View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fault recovery for zoned VP-based ATM networks

    Page(s): 211 - 215
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (402 KB)  

    The virtual path simplifies cell processing in ATM networks, while a good resource management strategy enables virtual channel connection to be established quickly and efficiently. Methods of constructing VP and the strategies of managing and allocating resources greatly affect the system performance. The paper proposes an architecture called 'zoned architecture', and the corresponding method of constructing VP. Related work about the multicast routing of the zoned architecture has been studied. The paper mainly focuses on routing and fault recovery; the various methods and strategies are studied for bandwidth control, re-routing, resource management and fault recovery. This model is evaluated by simulation experiment; and the performance measurements, such as bandwidth utilisation, blocking probability, re-routing ratio and recovery ratio, are obtained View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Generalised Berlekamp-Massey algorithm

    Page(s): 207 - 210
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (299 KB)  

    The Berlekamp-Massey algorithm is revisited and proven again by using the matrix representation. This approach makes the derivation and proof of the algorithm straightforward, simple and easily understood. It further enables the presentation of a generalised Berlekamp-Massey algorithm, including the conventional algorithm and the inversion-free algorithm as two special cases View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Analysis of the decoding process of T-codes

    Page(s): 202 - 206
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (406 KB)  

    T-codes are variable-length codes (VLC) generated by an iterative construction algorithm known as T-augmentation. All T-codes possess strong self-synchronising properties by virtue of T-augmentation, which spreads synchronisation information throughout the code construction. A number of applications have been reported that reference the T-codes for their desirable properties. However, further theoretical development is needed to realise their full benefits. The authors highlight important applications of self-synchronising codes and survey reported research on the decoding process of T-codes. Building on previous work, the authors present a thorough treatment of this process. Understanding the decoding process provides insight into the self-synchronisation process of T-codes. It also leads to an efficient method of determining the average synchronisation delay associated with each T-code, which is important for practical applications View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.