The authors present a hybrid Automatic Repeat reQuest technique using adaptive Reed–Solomon (RS) codes with packet erasure. This technique suits the transport layer in tactical mobile wireless networks with type I encryption, where encryption erasures the entire Internet Protocol packet. The novelty of the presented technique is the multifaceted optimisation of Reed–Solomon codes at the transport layer for delivery assurance, speed of service (SoS) and network throughput. With this technique, the transport layer in tactical networks can meet the stringent requirements of quality of service imposed by the tactical network user, even under adverse conditions. These requirements define a high level of reliability (delivery assurance), a specific SoS and optimum use of the limited bandwidth (BW) of the wireless network, where the probability of packet erasure can be very high. The provided probabilistic analysis shows that focusing on network throughput alone will result in violating SoS and delivery assurance requirements. On the other hand, focusing on SoS and delivery assurance requirements can result in poor network throughput. The multifaceted optimisation technique, which utilises hybrid ARQ for message delivery, is described using a homogeneous Markov chain.

This paper introduces a new transmission protocol that uses a joint lossless-source and channel coding scheme (JARQ). This protocol utilizes the self-synchronization property of lossless compression to generate indications of errors in the received data. This approach assures the reconstruction of images that were compressed in a lossless fashion and transmitted over a packet switched network, regardless of the error pattern that occurred during transmission. This new protocol is referred to as Go-Back-(N, M) JARQ. We evaluate the protocol by analyzing its throughput, and present simulation results that compare its throughput with those of conventional ARQ protocols. Using still images, we find significant gain in throughput performance for our Go-Back-(N, M) protocol, especially in moderate to low BERs, compared with that of the conventional Go-Back-N protocol. Moreover, by including packet combining, we find large throughput gains at low BERs, as well. This gain is realized for both random and burst errors. This approach provides: an increase in the code rate; a clear indication of error propagation; insensitivity to the type of error pattern-random or burst.

Arithmetic coding with embedded channel coding (ACECC) is an arithmetic coding approach in which the code points selected over the [0-1) interval are kept within a certain minimum Hamming distance of each other. The decoding process utilizes this distance, the knowledge of the source statistics, and the self-synchronization property of arithmetic coding, to overcome channel errors. This paper explains how maximum a posteriori probability symbol decoding (MAPSD) is implemented. We show how this joint source/channel coding approach can avoid the catastrophic effect of unrecoverable errors in both random- and burst-error cases. This coding method suits band-limited channels with moderate to high error rates.

The susceptibility of arithmetic coding to errors is utilized for channel error recovery using an automatic repeat request (ARQ) scheme. The presented scheme optimizes the tradeoff between the redundancy added to overcome channel errors and detection capability. With this approach, the type of error pattern introduced by the channel does not affect the receiver's detection capability. This scheme is suitable for transmitting long files over low bit error rate channels, and it provides the assurance of reconstructing the original data, free from catastrophic errors.

A watermark signal, which is a unique sequence of random variables, by itself does not give a decisive indication of ownership. This work addresses the need to include meaningful information such that a string of English characters, numbers, and punctuation is embedded within the watermark signal, while the signal remains a sequence of random variables. We attempt to provide answers for (1) the maximum number of symbols the watermark signal can carry, and (2) the most practical and reliable way to implement this. We approached this problem as power constrained multiple signaling over an AWGN channel.