The performance of packet-level iterative decoding is examined for a slow-frequency-hop spread-spectrum system using interleaved Reed-Solomon code words and per-dwell differential encoding. It is shown that the use of per-dwell soft-output detection in conjunction with successive-erasures decoding results in better performance in the presence of partial-band interference than previously considered iterative decoding techniques for the same packet format. Several alternatives are considered for soft-output symbol detection and the estimator of the per-dwell signal-to-interference-plus-noise ratio.

A stream processor is a power-efficient, high-level-language programmable option for embedded applications that are computation intensive and admit high levels of data parallelism. Many signal-processing algorithms for communications are well matched to stream-processor architectures, including partially parallel implementations of layered decoding algorithms such as the turbo-decoding message-passing (TDMP) algorithm. Communication among clusters of functional units in the stream processor impose a latency cost during both the message-passing phase and the parity-check phase of the TDMP algorithm with early termination; the inter-cluster communications latency is a significant factor in limiting the throughput of the decoder. We consider two modifications of the schedule for the TDMP algorithm with early termination; each halves the communication required between functional-unit clusters of the stream processor in each iteration. We show that these can provide a substantial increase in the information throughput of the decoder without increasing the probability of error.

Most digital signal processors contain one or more functional units with a single-instruction, multiple-data architecture that supports saturating fixed-point arithmetic with two or more options for the arithmetic precision. The processors designed for the highest performance contain many such functional units connected through an on-chip network. The selection of the arithmetic precision provides a trade-off between the task-level throughput and the quality of the output of many signal-processing algorithms, and utilization of the interconnection network during execution of the algorithm introduces a latency that can also limit the algorithm’s throughput. In this dissertation, we consider the turbo-decoding message-passing algorithm for iterative decoding of low-density parity-check codes and investigate its performance in parallel execution on a processor of interconnected functional units employing fast, low-precision fixed-point arithmetic. It is shown that the frequent occurrence of saturation when 8-bit signed arithmetic is used severely degrades the performance of the algorithm compared with decoding using higher-precision arithmetic. A technique of limiting the magnitude of certain intermediate variables of the algorithm, the extrinsic values, is proposed and shown to eliminate most occurrences of saturation, resulting in performance with 8-bit decoding nearly equal to that achieved with higher-precision decoding. We show that the interconnection latency can have a significant detrimental

The performance of serial acquisition is evaluated for direct-sequence spread-spectrum packet radio communications using dual antenna transmissions with orthogonal preamble sequences. Non-coherent diversity combining of the outputs from a corresponding pair of preamble matched filters is considered in conjunction with threshold-based acquisition using both fixed and adaptive acquisition thresholds. The effect of the intermediate-frequency filter and the automatic gain control subsystem on the acquisition performance is addressed, and the performance is evaluated in a slow fading channel. The effect of fixed and adaptive acquisition thresholds on the link outage probability is compared for the dual antenna system, and both are compared with the outage probability in a single antenna system.

Most high-throughput, fixed-point processors offer at least two options for arithmetic operations: 8-bit arithmetic, and 16-bit arithmetic. The lower resolution provides higher computational throughput at the cost of poorer performance in many applications. We investigate the effect of the resolution of saturating, fixed-point arithmetic on the performance of the turbo-decoding message-passing algorithm with quasi-cyclic low-density parity-check codes. We consider limits on the magnitude of extrinsic updates as a means to mitigate the effect of posterior-value saturation on the decoder's performance. We show that a fixed limit on updates only partially overcomes the greater effect of saturation in 8-bit operations, whereas a limit that depends on the degree of the variable node results in performance almost as good as what is possible with 16-bit operations.

A packet scheduler and a medium access control (MAC) protocol are presented for a direct-sequence spread-spectrum, wireless ad hoc network that contains a mix of nodes with directional antennas and nodes with omnidirectional antennas. The scheduler and MAC protocol are designed to prevent the co-site interference problem that arises in some types of nodes employing directional antennas. It is shown that the presence of nodes with directional antennas exacerbates the vulnerability of the network to the receiver blocking problem. A modification of the MAC protocol is presented that mitigates the receiver blocking problem, and it is shown to improve the performance of a network that includes nodes with directional antennas.

The TDMP layered belief-propagation algorithm is investigated for decoding a quasi-cyclic low-density parity-check code on a stream processor using fixed-point arithmetic. The effect of the processor's fixed-point resolution on the decoder performance is determined, and a simple technique is described for minimizing the performance penalty incurred when using the (highest throughput) lowest-resolution arithmetic mode of the processor. A reordering of the decoder schedule and a modification of the parity checks are also considered which permit increased software pipelining and improved latency hiding, with a corresponding increase in the data throughput. The fixed-point Storm-1 stream processor is used for comparative throughput results.

New, simple bounds are presented for the probability of error in a binary hypothesis test for communications using diversity signaling in correlated Rayleigh fading. The bounds are developed in the context of pairwise error-event probabilities in decoding an error-correction code. A long-standing conjecture regarding the form of worst-case error events in exponentially correlated Rayleigh fading is also proven. The utility of the results is illustrated by their application to transfer-function bounds on the probability of bit error for a system using a convolutional code. The closed-form transfer-function bounds are shown to be tighter than previously developed transfer-function bounds for communications in exponentially correlated Rayleigh fading.

A low-complexity, packet-level iterative detection technique is considered for slow-frequency-hop (SFH) spread-spectrum communications in intersymbol-interference (ISI) channels. Maximum-likelihood sequence estimation (MLSE) equalization with state pinning and bounded-distance errors-and-erasures decoding of Reed–Solomon code words are employed within each iteration of equalization and decoding. The design of a bit interleaver is examined for use with the iterative detection technique. The effect of state pinning on the equalizer performance is characterized, and the probability of packet error and the detection complexity of the SFH system are evaluated for a range of static and fading ISI channels. The use of an early-termination criterion is also considered as a way to achieve a tradeoff between performance and detection complexity with the iterative detection technique.

The performance of a system with a quasi-cyclic low-density parity-check code is examined under various constraints on the decoding delay. Two alternatives are considered for the message passing decoding algorithm: the sum-product algorithm and the turbo-decoding message passing algorithm. It is shown that their relative performance depends heavily on the stringency of the delay constraint; the TDMP algorithm results in substantially better performance than the SPA if the constraint is stringent.

New bounds on the probability of bit error are presented for a communication system with binary antipodal modulation and soft-decision maximum-likelihood decoding over a correlated Rayleigh-fading channel. The bounds are illustrated by considering a system using convolutional encoding. The bounds are closed-form expressions in terms of the code's transfer function. They are shown to be tighter than previously developed closed-form bounds for communications in correlated Rayleigh fading.

In a mobile ad hoc network using broadcast transmission scheduling, a terminal that is ready to join the network must first determine the current state of the transmission-scheduling protocol. In order to transmit and receive packets without interference, the new terminal needs to notify other terminals of its presence and exchange information with them so that it can form a collision-free broadcast transmission schedule. A new distributed protocol is described that allows a new terminal to acquire sufficient information about the terminals in its local neighborhood so that it can efficiently integrate itself into the existing transmission schedule. The protocol is also designed to enable a group of terminals to form a new transmission schedule if one does not already exist. We show that a large group of terminals can quickly form a new and efficient collision-free transmission schedule.

Slow-frequency-hop (SFH) spread-spectrum communications provide a high level of robustness in packet-radio networks for both military and commercial applications. Reed-Solomon (R-S) coding has proven to be a good choice for countering the critical channel impairments of partial-band fading and partial-band interference in a SFH system. In particular, it is effective if information about the reliability of individual code-symbol decisions or the content of entire dwell intervals is obtained at the receiver and used in errorsand-erasures (EE) decoding of the R-S code words. In this dissertation, we consider high-data-rate SFH communications for which the channel in each frequency slot is frequency selective, manifesting itself as intersymbol interference (ISI) at the receiver. The use of a packetlevel iterative equalization-and-decoding technique is considered in conjunction with a SFH system employing R-S coding. In each packet-level iteration, MLSE equalization is used in each dwell interval and is followed by boundeddistance EE decoding of the R-S code words. Several per-dwell interleaver designs are considered for the SFH systems. It is shown that packet-level iterations result in a significant improvement in performance with only a modest increase in detection complexity for a variety of ISI channels. The use of differential encoding in conjunction with the SFH system and packet-level iterations is also considered, and it is shown to provide further improvements in performance with only a modest additional increase in detection complexity. The performance SFH systems employing packet-level iterations with and without differential encoding is also evaluated for channels with partial-band interference. Comparisons are made between the performance of this system and the performance of SFH systems using some other codes and iterative decoding techniques.

Transmission-scheduling protocols can support contention-free link-level broadcast transmissions and delay sensitive traffic in mobile, multiple-hop packet radio networks. Use of transmission-scheduling protocols, however, can be very inefficient in mobile environments due to the difficulty in adapting transmission schedules. The paper defines a new adaptive and distributed protocol that permits a terminal to adapt transmission assignments to changes in topology using information it collects from its local neighborhood only. Because global coordination among all the terminals is not required and changes to transmission assignments are distributed to nearby terminals only, the protocol can adapt quickly to changes in the network connectivity. The two key parameters that affect the ability of the protocol to adapt to changes in connectivity are the rate of connectivity changes and the number of terminals near the connectivity changes. Using simulation, we determine the ranges for these parameters for which our adaptive protocol can maintain collision-free schedules with an acceptable level of overhead. The stability of the protocol is also characterized by showing that the protocol can quickly return to a collision-free transmission schedule after a period of very rapid changes in connectivity. Our channel-access protocol does not require a contention-based random-access phase to adapt the transmission schedules, and thus its ability to adapt quickly does not deteriorate with an increase in the traffic load.

We present a new distributed routing protocol for direct-sequence spread-spectrum, mobile ad hoc packet radio networks that contain a mix of nodes with directional antennas and nodes with omnidirectional antennas. Two components of the routing protocol are jointly designed: a new congestion-based link metric that is utilized to identify multiple routes with low levels of congestion and a new forwarding protocol that can dynamically split traffic among the multiple routes based on the relative capabilities of the routes. To be able to efficiently exploit the nodes with the directional antennas, our routing approach accounts for the additional capabilities of these nodes and utilizes multiple routes to a destination if the routes exhibit low mutual coupling. We show that our joint routing and forwarding approach provides substantial improvements in network performance compared to a scheme that simply selects minimum-hop routes. Our investigations demonstrate that the most significant gains in network performance are achieved in networks in which there is a mix of nodes with different types of antennas, but gains in network performance are also achieved in networks in which all nodes employ omnidirectional antennas only

In this paper, we introduce a simple method to improve the performance of threshold-based serial, matched-filter acquisition in packet radio communications. Each packet transmission includes a fixed-length acquisition preamble, and the preamble sequence used in packet transmissions is changed at predetermined times based on a sequence-generation algorithm. Two classes of sequence-generation algorithms are considered: random sequence generation, and preamble sequence generation based on m-sequences. Both are used in conjunction with a sequence-acceptance criterion based on the sidelobe energy of the preamble sequence. The tradeoff between the acquisition performance and the stringency of the acceptance criterion is examined for each sequence-generation algorithm. The computational burden of the search for an acceptable preamble sequence is also considered. It is shown that the use of the m-sequence generation algorithm results in much poorer acquisition performance for a given percentage of accepted candidate preamble sequences than is predicted using the model of a random sequence generator. Moreover, it is shown that that the tradeoff between acquisition performance and the computation required to find an acceptable preamble sequence is comparable for all of the m-sequence generators considered. Thus within the class of m-sequence generation algorithms, the use of a longer-period m-sequence generator results in a larger number of acceptable preamble sequences for a given level of acquisition performance and sequence-search computation

A SFH system using Reed-Solomon coding and the parity-bit method is considered with a receiver employing packet-level iterative detection and decoding, and the performance of the system in a channel with partial-band interference is evaluated. It is shown that the packet-level iterative reception technique results in substantially better performance than one-shot detection and decoding at a minimal cost in increased computation at the receiver. The use of an inner differential encoder is also examined, and its inclusion is shown to result in improved performance with packet-level iterative reception. Furthermore, alternative approaches to the parity-bit method are evaluated. Finally, the performance is compared with that of several other choices for coding and either one-shot or iterative decoding in SFH systems

A packet-level iterative detection technique that employs errors-and-erasures decoding has been described previously for SFH communications using Reed-Solomon coding. The technique enhances the performance of the SFH system in intersymbol-interference channels with only a minimal increase in complexity over one-shot errors-and-erasures decoding. In this paper, the performance of iterative EE decoding is considered for a SFH system with differentially encoded transmissions. It is shown that the use of differential encoding improves the performance of packet-level iterative detection in an AWGN channel with only a modest increase in detection complexity, and it also improves the performance in an intersymbol-interference channel in many instances. The packet size, the target probability of error, and the channel impulse response are considered, and the effect of each on the performance gain and the complexity is examined