William Webb

Subband-multipulse digital audio broadcasting for mobile receivers

The audio quality, robustness and implementational complexity of a novel mobile digital audio broadcast scheme are addressed. The audio codec proposed is based on an efficient combination of subband coding (SBC) and multipulse excited linear prediction coding (MPLPC). The bit allocation is dynamically adapted according to both the signal power in different subbands and a perceptual hearing model. Typically a segmental signal to noise ratio (SEGSNR) in excess of 30 dB associated with high fidelity subjective quality was achieved for 2.67-b/sample transmissions at a bit rate of 86 kb/s. Perceptually unimpaired audio quality was achieved for a bit error rate (BER) of about 10/sup -4/, when injecting random errors, which was degraded for increased BERs. In order to provide robust error protection, the audio codec was also subjected to a rigorous bit sensitivity analysis. Four different forward error correction schemes were investigated in order to explore the complexity, bit rate, and robustness tradeoffs. >

A packet reservation multiple access assisted cordless telecommunication scheme

A packet reservation multiple access (PRMA) assisted microcellular cordless telecommunications (CT) system is investigated for office type environments in absence of cochannel interference. The objective speech quality is found as a function of channel signal-to-noise ratio (SNR) for transmissions over narrowband Rayleigh fading channels, parameterized with the number of PRA users. A moderate complexity 16-ary CT scheme constituted by a 4-bit/symbol 32-kbps adaptive differential pulse code modulation (ADPCM) speech codec, Reed-Solomon forward error correction (FEC) codec, and diversity-assisted 16-level star quadrature amplitude modulation (16-StQAM) modem is proposed. The 264-kBd 20-slot PRMA scheme supports 36 users while maintaining virtually imperceptible speech degradation for channel SNRs in excess of about 25 dB, and for a mobile speed of 2 mi/h if the signal-to-interference ratio (SIR) is above 30 dB. >

Does 16-QAM provide an alternative to a half-rate GSM speech codec?

The computational complexity, speech quality, spectral efficiency, and robustness tradeoffs of speech transmission schemes for personal communications networks (PCNs) are addressed. Differentially encoded star 16-QAM arrangements with and without oversampling and diversity are compared, in terms of robustness, to the parallel frequency-division multiplex (FDM) 16-quadrature amplitude modulation (QAM) system to find the most appropriate modem scheme. The 13.4-kb/s regular-pulse-excited-long-term predictor (RPE-LTP) speech codec is subjected to rigorous bit-sensitivity analysis in terms of both cepstral distance (CD) and segmental signal to noise ratio degradations. A twin-class embedded binary Bose-Chaudhuri-Hocqunghem bit protection scheme is devised to improve robustness. The 6.2-kBd system proposed guarantees low-complexity, high-quality speech transmissions in a bandwidth of 12 kHz for vehicular speeds of 30 mph and channel signal to noise ratios in excess of 25 dB, a value readily maintained in a microcellular PCN.<<ETX>>

Performance of PRMA schemes via fading channels

A packet reservation multiple access (PRMA) assisted microcellular cordless telecommunications (CT) system is investigated for office type environments. The objective speech quality is found as a function of channel signal-to-noise ratio (SNR) for transmissions over narrowband Rayleigh fading channels, parametrized with the number of PRMA users. A moderate-complexity 16-ary CT scheme comprising a 4-b/symbol 32-kb/s adaptive differential pulse code modulation (ADPCM) speech codec, a Reed-Solomon forward error correction (FEC) codec, and a diversity-assisted 16-level star quadrature amplitude modulation (16-StQAM) modem is proposed. The 264-kBd 20-slot PRMA scheme supports 36 users while maintaining virtually imperceptible speech degradation for channel SNRs in excess of about 25 dB and for mobile speeds above 2 mph.

A low-rate multi-level voice/video transceiver for personal communications

A personal communication system (PCS) transceiver is proposed and investigated. A 4.8 kbit/s transformed binary pulse excited (TBPE) linear predictive speech codec, embedded source sensitivity-matched binary Bose-Chaudhuri-Hocquenghem (BCH) block error correction codecs, non-coherent differentially coded 16-level quadrature amplitude modulation (16-QAM) modem and packet reservation multiple access (PRMA) are deployed. The 2.15 kBd transceiver requires a signal-to-noise ratio (SNR) and signal-to-interference ratio (SIR) in excess of about 24 dB over Rayleigh-fading channels in order to support 10–11 nearly un-impaired voice conversations within a bandwidth of 30 kHz. Additionally, by reserving two PRMA time slots for video telephony, an 8.52 kbps videophone user can also be supported.

A mobile hi-fi digital audio broadcasting scheme

The audio quality, robustness and complexity issues of a novel mobile digital audio broadcast (DAB) scheme are addressed. The audio codec is based on a combination of subband coding (SBC) and multipulse excited linear predictive coding (MPLPC), where the bit allocation is dynamically adapted according to both the signal power in different subbands and a perceptual hearing model. Typically a segmental signal to noise ratio (SEGSNR) in excess of 30 dB associated with high fidelity (hi-fi) subjective quality was achieved for 2.67 bits/sample transmissions at a mono bit rate of 86 kbits/s. Four different source-matched forward error correction (FEC) schemes were investigated in order to explore the complexity, bit rate and robustness trade-offs. When using 4 bit/symbol 16-level star-constellation quadrature amplitude modulation (16-StQAM) the overall signalling rate became approximately 30 kBaud, accommodating two stereo DAB channels in a conventional 200 kHz analogue FM channels bandwidth. The diversity assisted DAB scheme required a channel signal to noise ratio (SNR) of about 25 dB for unimpaired audio quality via the worst-case Rayleigh fading mobile channel, when the mobile speed was 30 mph and the propagation frequency was 1.5 GHz. In case of the stationary Gaussian scenario an SNR of about 20 dB was required.<<ETX>>

Classic QAM Modems for Bandlimited AWGN Channels

This chapter contains sections titled: Introduction Trellis Coding Principles V.29 Modem V.32 Modem V.33 Modem Summary

Timing Recovery for Fading Channels

This chapter contains sections titled: Introduction Times-two Clock Recovery for QAM Basic Early-Late Clock Recovery for Star 16-QAM Modified Early-Late Clock Recovery for QAM Clock Recovery in the Presence of ISI Early-Late Clock Recovery Implementation Details Carrier Recovery Summary

Coded Modulation Assisted Channel Equalised Systems

This chapter contains sections titled: Introduction Intersymbol Interference Decision Feedback Equaliser Decision Feedback Equaliser Aided Adaptive Coded Modulation Radial Basis Function-Based Equalisation Turbo Equalisation Using Symbol-Based MAP Decoder RBF Assisted Turbo Equalisation of Coded Modulation Schemes In-phase/Quadrature-phase Turbo Equalisation RBF Assisted Reduced Complexity In-phase and Quadrature-phase Turbo Equalisation of Coded Modulation Schemes Summary

A narrowband mobile multimedia system

Publisher Summary This chapter is devoted to specific algorithmic and system aspects of complete voice/video phone transceiver suitable for the future third-generation PCS. The system bandwidth was assumed to be 30 kHz, as in the American IS-54 standard system, which allowed assessing the potential of the proposed scheme in comparison to a well-known benchmarker. The chapter addresses speech- and video-compression issues and focuses on the choice of modulation, in particular, on 16-level quadrature amplitude modulation (16-QAM) [1]. Forward error correction coding (FEC) is considered in this chapter. The proposed transceiver scheme and the system performance are analyzed. The system performance can be further improved at the cost of higher implementation complexity, when using a more sophisticated pilot, symbol-assisted, block-coded, coherent-square 16-QAM modem. The systems bandwidth efficiency can be further improved, for example, by employing the 2.4 kbps DoD MELP codec, while its robustness can be enhanced by coherent detection. Future work will be targeted at improving the speech quality, implementation complexity, bit rate, bandwidth occupancy, and error resilience trade-off achieved invoking the adaptive transceiver reconfiguration algorithms used in reference.