摘要:

A new approach is proposed for the prediction of bit error rate (BER) in binary FSK receivers subjected to impulsive noise. This is accomplished by calculating the conditional error probability of the receiver and then integrating over the probability distribution of the incoming noise. The analysis shows that the BER prediction is crucially dependent on accurately modelling the envelope of the receiver filter impulse response. The closer the model is to the actual receiver impulse response, the better the agreement between prediction and measurement. The new approach is verified experimentally, first for a uniform train of constant-strength impulses and secondly for different noise strength distributions (such as those determined from the noise amplitude distribution) for both steady and fading signal cases.

DOI:10.1049/ip-f-1.1987.0105

出版商:IEE    

年代:1987

数据来源: IET

摘要:

In the paper we analyse an improved soft-decision demodulation technique applied to μ-law pulse code modulation (PCM). The PCM-encoded waveforms are transmitted over Gaussian and Rayleigh fading channels by means of coherent phase shift keying modulation. Received bits which are unreliable are identified by means of soft-decision demodulation and erased. Each bit in the PCM word is assigned its own erasure threshold. The thresholds are optimised and are thus theoretically determined as a function of input power level to the quantiser, channel type and signal/noise ratio, and the relative mean-square error power that occurs when a PCM word containing an erased bit is estimated by means of prediction or interpolation. The predicted or interpolated sample amplitude is used as an estimate of the missing speech amplitude. This procedure works well because of the redundancy in the decoded speech waveform. In previous work the estimate was used to replace the PCM word with an erased bit. In the present paper the estimate is used to decide the most likely value of the erased bit, thus correcting most of the erasures without introducing errors in the replaced word. The new soft demodulation algorithm performs significantly better than the previous one. A gain of several decibels in overall speech signal/noise ratio is typical both for Gaussian and Rayleigh fading channels. Compared to no soft demodulation, the improvement in required channel signal/noise ratio is up to about 4 dB for the Gaussian channel. The combination of bit-position weighting and soft demodulation is also analysed, as well as the combination of a PCM bit assignment more robust against transmission errors and soft demodulation. For both of these cases, the combined schemes yield improvements, but not as large as for unweighted transmission of the standard bit assignment plus the improvement of the weighting or the robust bit assignment.

DOI:10.1049/ip-f-1.1987.0107

出版商:IEE    

年代:1987

数据来源: IET

摘要:

The paper attempts to clarify some issues concerning the use, and usefulness, of the packet-speech multiplexer. An accurate mathematical analysis is used to find the trade-off between delay and packet loss. The subjective effect of delay and packet loss is determined using data from our own, and published, conversation tests. It is found that the multiplexer will operate most efficiently if queueing delay is limited, with a large delay limit for heavy traffic conditions and a small delay limit for light traffic conditions. The improvement in TASI (time assignment speech interpolation) advantage gained through allowing delay is small, and it is concluded that the packet-speech multiplexer is unlikely to be used purely for its buffering facility. However, its ability to multiplex speech channels where the information rate is continually varying may be of significance for the next generation of speech coders.

DOI:10.1049/ip-f-1.1987.0108

出版商:IEE    

年代:1987

数据来源: IET

摘要:

After a brief review of discrete Fourier transformation (DFT) — and fast Fourier transformation (FFT) — properties, two representatives of partial (narrowband) spectrum computation methods, the zoom FFT and a specific type of zoom transform, are introduced. With respect to the FFT, only the zoom transform sufficiently fits demands concerning memory space and computional speed. Moreover, it offers improved accuracy and applicability relative to the existing type of zoom transform. The described transform method, a link between an FFT and a single DFT sum, is compared with the use of separate DFT sums.

DOI:10.1049/ip-f-1.1987.0109

出版商:IEE    

年代:1987

数据来源: IET

摘要:

The paper deals with the synthesis of the coherent asymptotically optimum detector (AOD) for fluctuating targets embedded in non-Rayleigh clutter. For the proposed detection scheme the asymptotic (large sample size) performance is evaluated. Comparison with the linear detector shows that, in principle, significant improvements are achievable by resorting to the AOD if the sample size is arbitrarily large. Since in radar system design the sample size is dictated by technological limits, the paper also investigates by computer simulation the performance of the AOD under such constraints. It turns out that the promises of the asymptotic theory cannot be achieved using moderately large sample sizes. Finally, a rather simple hybrid detection scheme is proposed, that performs better than both the asymptotically optimum and the linear detectors in actual operational situations.

DOI:10.1049/ip-f-1.1987.0110

出版商:IEE    

年代:1987

数据来源: IET

摘要:

computer simulation comparison of the tolerances to additive white Gaussian noise of two maximum-likelihood trellis decoding systems for use on discrete memoryless channels is presented. The first decoding system applies the Viterbi algorithm to the encoder trellis of a convolutional code; this is the well known standard Viterbi decoding system. The second decoding system, proposed by Yamada, uses the same algorithm but applies it to the syndrome-former trellis of the code. High-rate (n,n−1) systematic and nonsystematic convolutional codes, with rates ½, ¾, ⅚ and ⅞ are used throughout the tests. Simulation results are presented for hard- and soft-decision decoding with BPSK modulation and coherent detection. Results show that the Viterbi and Yamada decoding systems give identical error performance for the same code. The implementation complexity of the systems is also examined; a useful reduction in the number of binary comparisons required by the Yamada system can be achieved.

DOI:10.1049/ip-f-1.1987.0111

出版商:IEE    

年代:1987

数据来源: IET

摘要:

The paper presents a method of traffic analysis for a network which has extensive alternative routing and mutual overflow. The model is based on the mean and variance of the traffic and uses the Bernouilli-Pascal-Poisson (BPP) procedure to describe individual traffic streams on each link of the network. The method is applied to product graph networks where the load on each link includes a large number of traffic streams resulting from one or more stages of overflow. The node-to-node blocking probabilities and other relevant parameters are computed and compared with values obtained from a simulation study, and with values resulting from a simplification based on the equivalent random traffic (ERT) method. The model is also used to calculate the sum of the covariances between the marginal streams of traffic carried on individual links.

DOI:10.1049/ip-f-1.1987.0113

出版商:IEE    

年代:1987

数据来源: IET

摘要:

The problem of finding theoptimum linear detectorfor a general binary composite hypothesis testing problem in additive white Gaussian noise is addressed in the paper. The signal set consists of a limited number of known signals with knowna prioriprobabilities on each binary hypothesis. Thea prioriprobability for each hypothesis is also assumed known. The linear detector to this binary decision problem consists of a linear filter and a comparison with a threshold. In the paper we show how to find the optimum filter and threshold for this linear detector, for the limiting cases of infinitely large and vanishingly small noise power, respectively. An analytical solution is given for the optimum solution in the case of infinitely large noise power and a recursive algorithm, giving the optimum solution in the case of vanishingly small noise power, is presented. These solutions are valid without any restrictions on signals anda prioriprobabilities.

DOI:10.1049/ip-f-1.1987.0115

出版商:IEE    

年代:1987

数据来源: IET

摘要:

Binary continuous phase modulations with constant amplitude (CPM) with modulation index 1/2 have been shown to provide both good spectral and error probability properties. These modulations can also be detected with a very simple linear detector, which can be implemented simply at high speeds. This can be done with only a minor degradation in error performance. The linear detector can be implemented both as a serial and as a parallel detector, and these have slightly different sensitivity to errors in bit timing and phase synchronisation. In the paper we study the effects of multiple interfering signals on both these linear detectors for CPM. Coherent detection is assumed, and both Gaussian and slow Rayleigh fading channels are considered. Analytical formulas are derived on both channels, where only a single integral has to be calculated numerically. This formula can also easily be used for detector filters with impulse response of long time duration. It is shown that the difference in performance due to the number of interferers is small, when the total interference power is kept constant. Further, it is shown that the performance is improved by several decibels for the longer asymptotically optimum filters in adjacent channel interference. As one would expect, the smoothed phase modulations have significantly larger tolerances against adjacent channel interference, while the difference is small in cochannel interference. On the Rayleigh fading channel, the longer asymptotically optimum filters on the Gaussian channel lead to much lower irreducible error probabilities than does the MSK filter; a decrease by a factor of as much as 10 has been found.

DOI:10.1049/ip-f-1.1987.0117

出版商:IEE    

年代:1987

数据来源: IET

摘要:

The paper provides a new algorithm to recursively update the parameter vector of a linear predictor. The proposed method is based on modified Cholesky (LD) factorisation of an augmented correlation matrix. In essence the algorithm suggested is a modified version of the method presented by Ljung and Soderstrom, with distinct advantages. Although computationally the burden of both methods is the same, the proposed algorithm has been mathematically restructured in such a way that a computational step is eliminated. This results in considerable simplifications when implementing the suggested algorithm in a systolic array configuration, since the problem reduces to the processing of a single matrix structure.

DOI:10.1049/ip-f-1.1987.0119

出版商:IEE    

年代:1987

数据来源: IET