John Q. Liu

Generalized Kasami Sequences: The Large Set

In this correspondence, new binary sequence families F^k of period 2ⁿ-1 are constructed for even n and any k with gcd(k,n)=2 if n/2 is odd or gcd(k,n)=1 if n/2 is even. The distribution of their correlation values is completely determined. These families have maximum correlation 2^n/2+1 and family size 2^3n/2 + 2^n/2 for odd n/2 or 2^3n/2+2^n/2-1 for even n/2. The proposed families include the large set of Kasami sequences, where the k is taken as k=n/2+1.

Signal Detection in Optical Communications through the Atmospheric Turbulence Channel

The probability of a miss in the detection of a signal in an optical communications system through the turbulent atmosphere using intensity modulation is studied. The turbulence of the atmosphere causes scintillation of the received signal intensity which is treated as a lognormal random process. The received background radiation and electronic noise in the receiver is treated as additive white Gaussian noise (AWGN). A Chernoff bound is derived. For practical values of the signal-to-noise power ratio (SNR), a series solution for the characteristic function of the lognormal random variable is used to find the probability of miss. Simulation results agree with theoretical results. The method developed in this paper can be used by the system designer to choose the proper signal length and meet the system specifications for signal detection.

Optical signal detection in the turbulent atmosphere using p-i-n photodiodes

Performance of signal detection for intensity modulated direct detection optical communications systems through the turbulent atmosphere is studied considering statistics of photoelectron count in p-i-n photodiodes. The effect of atmospheric turbulence is modeled as a lognormal process. In this system, the photoelectron count is a conditional Poisson process, where the mean count is lognormal. A normal approximation is derived for this conditional Poisson process, which results in a simple calculation for the probability of miss. The probabilities of miss for quantum-limited detection and detection in the presence of thermal electronic noise are simulated and compared with calculations using the normal approximation. Simulation results match the analytical results. For the thermal noise case, the probability of miss is compared with the probability of false alarm to determine appropriate signal length and detection threshold settings for a required performance. Applying results in this paper, a system designer can determine appropriate signal length and detection threshold settings.

Uncoded Error Performance of eCall Modem through AMR Codec and AWGN

Data communications through the voice channel of the cellular network are important for telematics applications such as automatic vehicle crash notification, an implementation of which is the pan- European eCall system scheduled for deployment in 2015. The cellular voice channel includes the vocoder which is designed for compressing speech waveforms and is a hindrance for data communications. No mathematical model exists for the vocoder as a data communications channel. Thus, in this paper, the uncoded bit error rate (BER) is found by simulation using the bit-exact C- code implementation of the vocoder. The adaptive multi-rate (AMR) vocoder is used in GSM, the network on which eCall runs and is the one used in these simulations. The uncoded BER is presented for different compression rates along with other effects of the vocoder such as a wake-up effect and the signal-to-distortion ratio.

Design of Binary Multiple Level Sequences

Constructing long multiple level sequences using much shorter component codes is proposed. Correlation properties and constructions of multiple level sequences are studied. Upper bounds and lower bounds are derived for autocorrelation functions and cross-correlation functions. A-quantized sequences and P-quantized sequences are proposed as criteria for the construction. Multiple level sequences can help to reduce the hardware complexity of signal detection to L1/M/L compared with the complexity of traditional signal detectors, with M*to2 the number of levels and L*bj1 the signal length, while maintaining the same level of correlation properties as other well-known sequences in spread spectrum communications.

Signal detection for optical communications through the turbulent atmosphere

The probability of a miss in the detection of a signal in an optical communications system through the turbulent atmosphere using intensity modulation is studied. The turbulence of the atmosphere causes scintillation of the received signal intensity which is treated as a lognormal random process. The received background radiation and electronic noise in the receiver is treated as additive white Gaussian noise (AWGN). A Chernoff bound is derived for the lognormal sum distribution. An approximation for the lognormal sum distribution is investigated for its utility in calculating the probability of miss. For practical values of the signal-to-noise power ratio (SNR), a series solution for the characteristic function of the lognormal random variable is used to find the probability of miss. Simulation results agree with theoretical results. The method developed in this paper can be used by the system designer to choose the proper signal length and meet the system specifications for signal detection.

Effect of Atmospheric Turbulence on Packet Detection in Optical Communications

The effect of atmospheric turbulence on packet detection is studied in the range of the scintillation index ¿ relevant to laser communications. The effect of scintillation is treated as a lognormal random process. Therefore, finding the probability of miss is equivalent to finding the cumulative distribution function of a sum of L independent lognormal random variables. Two techniques for finding this sum are compared with simulation, and probability of miss curves are presented for a correlation detector with preamble length L = 108 and ¿ ranging from 0.2 to 1.0.

Uncoded bit error rate of eCall modem through AMR codec and AWGN channel

Data communications through the voice channel of the cellular network are important for telematics applications such as automatic vehicle crash notification, an implementation of which is the pan-European eCall system scheduled for deployment in 2015. The cellular voice channel includes the vocoder which is designed for compressing speech waveforms and is a hindrance for data communications. No mathematical model exists for the vocoder as a data communications channel. Thus, in this paper, the uncoded bit error rate (BER) is found by simulation using the bit-exact C-code implementation of the vocoder. The adaptive multi-rate (AMR) vocoder is used in GSM, the network on which eCall runs, and is the one used in these simulations. The BER is first presented for the vocoder-only channel at different AMR compression rates. Then it is presented for the vocoder and additive white Gaussian noise (AWGN) channel. Finally, an attempt is made to find the optimum receiver filter for the vocoder channel.

Demodulation performance of multiband UWB communication system with one RF and ADC module receiver

This paper studies the BER performance of a multiband UWB system with a common RF and ADC module in the receiver. Communication signals in multiple RF bands are received by one UWB antenna. The ADC samples the received signal in a swath. This means that the RF signals with lower frequencies are oversampled. Using the ADC output signal, the demodulator demodulates each RF signal individually. The quantization noise process depends on the resolution of the ADC and is considered as a uniform distribution. The bit error rate of each RF signal is analyzed. Numerical results are provided for different ADC resolution. The results can help system designer to choose the right resolution for ADC in multiband UWB communications systems.

Packet Detection for Onboard Switching Broadband IP Satellite Networks

The signal detection performance is studied for an onboard switching broadband IP satellite network. The probability of false alarm, miss and early detection is analyzed and simulated using either CFAR or the classic method. The unique word length is 108 symbols and the SNR is Eb/N0=1.8 dB. It is shown that the overall probability of packet loss can be less than 10-9