• COMPARISON OF NONCOHERENT DETECTORS FOR SOQPSK AND GMSK IN PHASE NOISE CHANNELS

      Perrins, Erik; Syed, Afzal; University of Kansas (International Foundation for Telemetering, 2007-10)
      SOQPSK and GMSK are highly bandwidth efficient continuous phase modulation (CPM) schemes with several desirable qualities. In both cases, coherent detectors are available with good performance in AWGN. In this paper, we develop reduced complexity noncoherent detectors for SOQPSK and GMSK; and discuss a phase noise model. This is followed by a performance comparison of both the noncoherent detectors in channels with phase noise.
    • Multiple Bit Differential Detection of SOQPSK with Diversity Reception

      Perrins, Erik; Ramakrishnan, Madhusudhan; University of Kansas (International Foundation for Telemetering, 2007-10)
      In this paper, we consider multiple bit differential detection (MBDD) of differentially encoded shapedoffset quadrature phase-shift keying (SOQPSK) over slow fading channels, especially Rayleigh fading channels. SOQPSK is a highly bandwidth efficient and popular form of constant envelope continuous phase modulation (CPM). We present two versions of the MBDD algorithm: the full-size version (FMBDD) which uses a detection window that spans the entire N-bit observation window, and an improved version (I-MBDD) which maintains the original N-bit observation window but detects only N − 2 bits within the window. The complexity of both algorithms is shown to increase linearly with the order of diversity reception, L, and exponentially with the size of the observation window, N; the I-MBDD achieves the best performance for given values of L and N. As expected, the performance in the case of diversity reception shows a marked improvement over the single channel case.
    • ON REDUCED COMPLEXITY TECHNIQUES FOR BANDWIDTH EFFICIENT CONTINUOUS PHASE MODULATIONS IN SERIALLY CONCATENATED CODED SYSTEMS

      Perrins, Erik; Kumaraswamy, Dileep; University of Kansas (International Foundation for Telemetering, 2007-10)
      Serially concatenated coded (SCC) systems with continuous phase modulations (CPMs) as recursive inner codes have been known to give very high coding gains at low operative signal to noise ratios (SNRs). Moreover, concatenated coded systems with iterative decoding approach the bit error rate (BER) bounds given by the maximum likelihood (ML) criterion. Although SCC systems by themselves are reduced complexity systems when compared to the ML decoding, when very highly bandwidth efficient CPMs such as pulse code modulation /frequency modulation (PCM/FM) is used [1], they present a problem of extremely high decoding complexity at the receiver. The complexity of a CPM is described by the size of its trellis which is a function of the modulation index, the cardinality of the source alphabet and the length of the frequency pulse used. The surveyed complexity reduction techniques adopt approximations which will reduce the size of the trellis with minimal expense of power. In this paper, we present reduced complexity approaches to sub-optimally decode SCC PCM/FM by mainly two approaches - 1) Frequency pulse truncation. 2) Decision feedback.
    • SERIALLY CONCATENATED HIGH RATE CONVOLUTIONAL CODES WITH CONTINUOUS PHASE MODULATION

      Perrins, Erik; Damodaran, Kanagaraj; University of Kansas (International Foundation for Telemetering, 2007-10)
      We propose serially concatenated convolutional codes with continuous phase modulation for aeronautical telemetry. Such a concatenated code has an outer encoder whose code words are permuted by an interleaver, and a modulation, which is viewed as a code and takes the interleaved words as its input and produces the modulated signal. Since bandwidth expansion is a concern when coding is introduced, we focus on high rate punctured codes of rates 2/3 through 9/10. These are obtained by puncturing the basic rate 1/2 convolutional codes with maximal free distance. At the receiver end we use a reduced complexity iterative decoding algorithm which is essentially a soft input soft output decoding algorithm. These simple highly powerful concatenated codes produce high coding gains with minimum bandwidth expansion.
    • SIMPLIFIED 2-state Detectors for SOQPSK-TG and SOQPSK-MIL

      Perrins, Erik; Kumaraswamy, Balachandra; University of Kansas (International Foundation for Telemetering, 2007-10)
      We study simple trellis-based detectors for SOQPSK that have a minimal level of complexity. In particular, we show that the state complexity can be cut in half relative to previous approaches—from 4 states down to 2—with asymptotically optimum performance. We give two possible means of achieving this: the pulse amplitude modulation (PAM) technique and the pulse truncation (PT) technique; both of these techniques make use of recent advances in SOQPSK technology based on a continuous phase modulation (CPM) interpretation of SOQPSK. The proposed simplifications are significant since trellis-based SOQPSK detectors are 1–2 dB superior to widely-deployed symbol-by-symbol detectors. These performance gains come at the expense of complexity, and the proposed 2-state detectors minimize this expense. Thus, these simple detection schemes are applicable in settings where high-performance and low complexity are needed to meet restrictions on power consumption and cost.
    • SYMBOL TIMING RECOVERY FOR SOQPSK

      Perrins, Erik; Chandran, Prashanth; University of Kansas (International Foundation for Telemetering, 2007-10)
      Shaped offset quadrature phase shift keying (SOQPSK) is a highly bandwidth efficient modulation technique used widely in military and aeronautical telemetry standards. It can be classified as a form of continuous phase modulation (CPM), but its major distinction from other CPM schemes is that it has a constrained (correlated) ternary data alphabet. CPM-based detection models for SOQPSK have been developed only recently. One roadblock standing in the way of these detectors being adopted is that existing symbol timing recovery techniques for CPM are not always applicable since the data symbols are correlated. We investigate the performance of one CPM-based timing error detector (TED) that can be used with SOQPSK, and apply it to the versions of SOQPSK used in military (MIL-STD SOQPSK) and telemetry group (SOQPSK-TG) standards. We derive the theoretical performance limits on the accuracy of timing recovery for SOQPSK, as given by the modified Cramer-Rao bound (MCRB), and show that the proposed TED performs close to these bounds in computer simulations and is free of false-lock points. We also show that the proposed scheme outperforms a non-data aided TED that was recently developed for SOQPSK. These results show that the proposed scheme has great promise in a wide range of applications due to its low complexity, strong performance, and lack of false-lock points.