Carl W. Baum

A sequential procedure for multihypothesis testing

The sequential testing of more than two hypotheses has important applications in direct-sequence spread spectrum signal acquisition, multiple-resolution-element radar, and other areas. A useful sequential test which we term the MSPRT is studied in this paper. The test is shown to be a generalization of the sequential probability ratio test. Under Bayesian assumptions, it is argued that the MSPRT approximates the much more complicated optimal test when error probabilities are small and expected stopping times are large. Bounds on error probabilities are derived, and asymptotic expressions for the stopping time and error probabilities are given. A design procedure is presented for determining the parameters of the MSPRT. Two examples involving Gaussian densities are included, and comparisons are made between simulation results and asymptotic expressions. Comparisons with Bayesian fixed sample size tests are also made, and it is found that the MSPRT requires two to three times fewer samples on average. >

The effect of automatic gain control on serial, matched-filter acquisition in direct-sequence packet radio communications

The performance of a noncoherent serial acquisition technique is evaluated for direct-sequence spread-spectrum packet communications. The acquisition technique that is considered uses threshold crossing of a matched-filter output to detect a fixed-length preamble at the start of each packet. The analysis accounts for frequency mismatch between the transmitter and the receiver due to oscillator inaccuracies and mobility-induced Doppler shifts. It also accounts for the effects of automatic gain control (AGC) in the receiver. The role of the AGC system in determining the acquisition performance is examined. In addition, selection of the optimal acquisition threshold is considered, and a simple method for selection of a good suboptimal threshold is presented. It is shown that use of this threshold results in performance close to that obtained with the optimal threshold over a wide range of channels.

A decision-theoretic approach to the generation of side information in frequency-hop multiple-access communications

The use of side information can provide increased coding gains in frequency-hop multiple-access communication systems. The internal generation of this side information by the receiver is investigated in the paper. Methods to determine which received signals to erase prior to decoding are derived from Bayesian decision theory. The performance of synchronous and asynchronous frequency-hop systems with Reed-Solomon coding is analyzed for channels with Rayleigh fading, multiple-access interference, and wideband Gaussian noise. The performance of the Bayesian technique is compared with that of receivers that make erasure decisions using the Viterbi ratio-threshold test. The Bayesian method is found to provide better performance. >

A reliability output Viterbi algorithm with applications to hybrid ARQ

We present an algorithm that permits a receiver to calculate the probability of packet error in parallel with the Viterbi decoding process. This packet reliability value may be used to request retransmissions in a type-I hybrid ARQ scheme. It is shown that this scheme can be used to guarantee any required bound on the packet error probability. In addition, this scheme can be used in conjunction with packet combining. The performance of this scheme is compared with that of the Yamamoto-Itoh algorithm and is shown to provide a significant improvement in throughput.

Asymptotic efficiency of a sequential multihypothesis test

A sequential multihypothesis test known as the M-ary sequential probability ratio test (MSPRT) is generalized to account for nonuniform decision costs. Bounds on error probabilities and asymptotic expressions for the stopping time and error probabilities are given. A key result of this correspondence is a proof that the generalized MSPRT is asymptotically efficient.

Hybrid acquisition schemes for direct sequence CDMA systems

The acquisition of chip timing in direct sequence code division multiple-access (DS/CDMA) systems must occur before the demodulation of incoming messages can begin. Recent work has shown that the problem of acquisition may be the dominant factor limiting the capacity of DS/CDMA systems and networks. Passive matched filters and parallel search schemes have been shown to be able to acquire signals rapidly, but they do so at the cost of high or prohibitive complexity. In contrast, straight serial search schemes have lower complexity but acquire the signal much more slowly. In this paper, we present several hybrid active correlation schemes that provide flexibility in the trade-off between acquisition speed and complexity. These techniques test several phases concurrently, and either decide that a particular phase is correct, in which case the decision is verified by a binary hypothesis test, or it is decided that none of the phases are correct, in which case another group of phases is tested. Several fixed sample size (FSS) tests are considered, as well as a test based on a new M-ary sequential hypothesis test called the MSPRT. It is found that the MSPRT provides the best performance.<<ETX>>

Erasure insertion in frequency-hop communications with fading and partial-band interference

The use of block coding and errors-and-erasures decoding can enhance performance substantially in frequency-hop communication systems, provided that a good scheme is employed to determine which symbols to erase. In this paper, methods for determining erasures derived from Bayesian decision theory are applied to the mitigation of fading and partial-band interference. The performance of receivers using the Bayesian technique is compared with that of receivers that make erasure decisions using Viterbis (1982) ratio-threshold test. The performance of hard-decision demodulation and the theoretical performance of receivers with access to perfect side information are also compared. It is found that the Bayesian receiver provides the best performance, and that error probabilities for the Bayesian receiver are lower than those for hard-decision demodulation by as much as six orders of magnitude.

Methods for packet combining in HARQ systems over bursty channels

The performance of ARQ systems can be improved by combining current and prior transmissions at the receiver. Two techniques for combining outputs in a packet-based communication system are presented. In both techniques the fundamental unit of retransmission is a packet, and the fundamental unit of combining is a codeword. The techniques are analyzed for a bursty channel and a system that employs Reed–Solomon coding and bounded-distance errors-and-erasures decoding. Performance results show that the packet-combining schemes provide significant gains in throughput and reductions in error probability when compared with a system that does not employ combining.

A multicarrier transmission scheme for wireless local communications

This paper presents a novel multitone transmission scheme that employs a nonlinear binary code to specify multitone signal constellations. The scheme, which can be viewed as a generalization of M-ary frequency shift keying, is amenable to noncoherent demodulation, and it is therefore useful for communication over fading channels. We determine the performance of this modulation scheme when it is used in conjunction with a Reed-Solomon code for error control. Methods for determining and erasing unreliable multitone constellations are presented and their performance is evaluated when they are used in conjunction with errors-and-erasures decoding. The transmission scheme is evaluated with respect to complexity, bandwidth efficiency, and performance in the presence of Gaussian noise and frequency- and time-selective fading.

An energy-efficient routing protocol for wireless sensor networks with battery level uncertainty

A wireless microsensor network consists of a large number of low-cost nodes communicating with a base station. The nodes have limited energy and short transmission ranges. By employing estimates of the remaining battery energy in routing, the use of radios nearing energy exhaustion can be discouraged. Knowledge of the rate at which a node is consuming its energy provides additional useful information. This paper investigates novel routing metrics that incorporate both kinds of information to improve network performance. Also, in practice, there may be some error in a nodes estimate of its battery level. The metrics are designed to explicitly account for uncertainty in remaining energy. Numerical results using several statistical models for this uncertainty show that the proposed metrics outperform a variety of alternatives.