John Eric Kleider

An adaptive-rate digital communication system for speech

Current digital voice communication systems allow only modest levels of protection of the coded speech and often do not follow the dynamic changes that occur in the transmission channel. We present a method that provides optimal voice quality and intelligibility for any given transmission channel condition. The approach is performed via adaptive-rate voice (ARV) coding using an adaptive-rate modem, channel coding, and a multimode sinusoidal transform coder. In general, the receiver utilizes channel state information to not only optimally demodulate and decode the currently corrupted symbols from the channel, but also to inform the transmitter, via a feedback channel, of the optimal strategy for voice/channel coding and modulation format. We compare several source-channel coding schemes at multiple transmission symbol rates and compare the performance to fixed aggregate-rate channel-controlled variable rate voice coding systems.

Speaker independent speech recognition system and method

An improved method of training a SISRS uses less processing and memory resources by operating on vectors instead of matrices which represent spoken commands. Memory requirements are linearly proportional to the number of spoken commands for storing each command model. A spoken command is identified from the set of spoken commands by a command recognition procedure (200). The command recognition procedure (200) includes sampling the speakers speech, deriving cepstral coefficients and delta-cepstral coefficients, and performing a polynomial expansion on cepstral coefficients. The identified spoken command is selected using the dot product of the command model data and the average command structure representing the unidentified spoken command.

Broadband OFDM using 16-bit precision on a SDR platform

This paper discusses our development of a robust broadband model of OFDM using 16-bit fixed-point ANSI C code that can be easily ported to a variety of software defined radio (SDR) platforms. The fixed-point model allows fast runtime, low power consumption, and low cost implementation. The receiver analog to digital converter (ADC) resolution and 16-bit fixed-point issues are explored. The OFDM system consists of DQPSK encoded data symbols, pilot symbols and null symbols to minimize aliasing and reduce filter requirements. The total number of OFDM subchannels is set to 256. A cyclic extension scheme is used to reduce the intersymbol interference (ISI) caused by long multipath delay spreads. The channel model consists of additive white Gaussian noise (AWGN) and continuously variable channel delay. The fixed-point OFDM algorithm was optimized with a system that uses no compensation for peak to average ratio (PAR). The model supports post-detection spatial diversity, which can be easily implemented by the multi-channel architecture provided by many SDRs.

Embedded Synchronization/Pilot Sequence Creation Using Pocs

In this paper we build on the orthogonal frequency division multiplexing (OFDM) peak-to-average power ratio (PAR) reduction work by Chen and Zhou. It has been demonstrated that pilot sequences that are constant modulus in the time domain can lead to an ensemble PAR-reduction across all data realizations. However, the problem of creating constant modulus sequences from arbitrary frequency domain power profiles has never been addressed. Often, it is desirable to have a some freedom of choice in how pilot and, possibly synchronization, energy is allocated in the frequency domain. In this paper we present a projection on to convex sets (POCS) method for creating low-PAR synchronization/pilot (S/P) sequences with arbitrary frequency-domain power profiles

An adaptive-rate anti-jam system for optimal voice communication

Systems for the digital transmission of voice have traditionally been fixed rate in anti-jam situations. This design severely limits the potential of the communication system in the case of voice since compression of speech is a lossy process. Bit errors in the transmission of voice coded speech cause degradation which can be compensated for in several ways. The standard method is to trade off channel coding and source coding bit rates to achieve an accepted level of quality in a fixed rate system. We present a method that provides optimal voice quality and intelligibility for any given transmission channel condition. The approach is performed via a fully adaptive-rate system using an adaptive-rate modem, channel coding, and a multimode voice coder. In general, the receiver utilizes channel state information to not only optimally demodulate and decode the currently corrupted symbols from the channel, but also to inform the transmitter, via a feedback channel, of the optimal strategy for voice/channel coding and modulation format. We compare several joint source-channel coding schemes at multiple transmission symbol rates and compare the performance to fixed aggregate-rate channel-controlled variable rate voice coding systems. We show through simulation that the reduction in speech distortion is greater for a modulation change than a channel coding change in many realistic situations because of the variability of coding gain.

Speaker identification system and method

A speaker identification system (10) employs a supervised training process (100) that uses row action projection (RAP) to generate speaker model data for a set of speakers. The training process employing RAP uses less memory and processing resources by operating on a single row of a matrix at a time. Memory requirements are linearly proportional to number of speakers for storing each speakers information. A speaker is identified from the set of speakers by sampling the speakers speech (202), deriving cepstral coefficients (208), and performing a polynomial expansion (212) on cepstral coefficients. The identified speaker (228) is selected using the product of the speaker model data (213) and the polynomial expanded coefficients from the speech sample.

Robust image transmission using source adaptive modulation and trellis-coded quantization

The paper presents a robust, low-complexity method of transmitting digitally compressed imagery through very noisy channels. The method is applicable to AWGN and fading channels. The proposed method uses a robust wavelet based image coder employing trellis-coded quantization (TCQ), and optimal source adaptive modulation (SAM) which is implemented through time-frequency diversity. We compare the performance of the SAM-TCQ system to that of a system that utilizes unequal error protection (UEP). We show that for the binary symmetric channel (BSC), the SAM-TCQ system performs as well as UEP-TCQ at high bit error rates (BER), and within 1 dB of the UEP-TCQ system at low bit error rates.

Emerging software defined radio architectures supporting wireless high data rate OFDM

Orthogonal frequency division multiplexing (OFDM) has been identified as one potential method to enhance the performance of wireless communication links degraded by cosite interference, impulsive noise, and frequency-selective fading. In the past, implementation complexity slowed the development of OFDM for useful commercial and handheld applications. With the advances in semiconductor processing technology and digital signal processing, OFDM is now practical for system solutions including wireless LANs, audio and television broadcasting, and land mobile services. We describe the application of OFDM within software defined radio (SDR) platforms to provide multi-service land mobile radio capabilities. The simple combination of OFDM and SDR provides great flexibility in system configuration, significant reduction in product development cycle time, and very high potential for software reuse.

Robust time-frequency synchronization for OFDM mobile applications

Orthogonal frequency division multiplexing (OFDM) has become a popular multicarrier transmission scheme for transmission of data requiring high data rates. We present a digital synchronization technique that provides accurate timing and frequency tracking estimates for severely degraded and noisy channel environments. In addition, the synchronization technique can correct large timing misalignment of the received OFDM symbols, and provides acquisition of frequency offsets on the order of multiple sub-channel spacings. Algorithm complexity is minimized by deriving timing and frequency error estimates from the same basic algorithm, making it suitable to handheld radios. The estimation accuracy is excellent, even for very low signal-to-noise ratios. The algorithmic converges within one baud interval, and thus provides a good solution for systems requiring rapid acquisition and symbol timing alignment at system startup.

Optimal error protection for image transmission using source-adaptive modulation

This paper presents a low-complexity method of transmitting digitally compressed imagery through AWGN and fading channels. The proposed method combines a wavelet-based image coder that employs phase scrambling and trellis-coded quantization (TCQ), and source adaptive modulation (SAM). We present two versions of SAM that utilize BPSK (SAM-TCQ) and 16-ary PPM (OSAM-TCQ), respectively. We then compare the performance of the SAM systems to that of a system that utilizes unequal error protection (UEP). We show that for the binary symmetric channel, the SAM-TCQ system performs as well as UEP-TCQ at high bit error rates, and within 1 dB of the UEP-TCQ system at low bit error rates. Additionally, we show that for the AWGN channel, the OSAM-TCQ system performs nearly 4 dB better than UEP-TCQ at high bit error rates, and the same as the UEP-TCQ system at low bit error rates, with much lower complexity than the UEP-based system.