Showing papers by "Devendra Jalihal published in 2018"
TL;DR: The minimum average length or energy of the sync frame necessary for error-free frame synchronization is characterized and the results under relaxed assumptions are evaluated, including the imperfect knowledge of arrival distribution and symbol timing error.
Abstract: We study the problem of frame synchronization over a discrete memoryless channel (DMC) in an asynchronous setup. A sync frame is transmitted at a random time $V$ with a known distribution $\{a_{v} \}$ and entropy $H$ . We seek to characterize the minimum average length or energy of the sync frame necessary for error-free frame synchronization, as $H$ tends to infinity. We present a variable length sync frame, where the length of the sync frame is adapted based on $H$ and $\{a_{v} \}$ , for the general arrival distribution and show error-free frame synchronization when the average sync frame length $\bar {N}$ scales as $\Omega (({H}/{\alpha (Q)})$ , where $\alpha (Q)$ is the synchronization threshold of the DMC. We then generalize the framework and study a tradeoff between $\bar {N}$ and $\alpha (Q)$ for optimal frame synchronization and characterize the scaling needed of both $\bar {N}$ and $\alpha (Q)$ with $H$ . We illustrate our results with the AWGN channel and discuss the adapting sync frame length and symbol power for optimal frame synchronization. Finally, using numerical work and simulations, we evaluate the results under relaxed assumptions, including the imperfect knowledge of arrival distribution and symbol timing error.
2 citations
01 Oct 2018
TL;DR: The Jammer to Signal (J/S) strength required to disrupt the Radio link under each of these duplexing schemes is examined to provide useful insight into maximum throughput achievable for various FEC schemes under jamming conditions.
Abstract: It is well known that modern battlefields are subject to intense jamming by enemy forces to deny use of the electromagnetic spectrum. To combat jamming, the Line of Sight (LoS) wireless communication is typically employs Frequency Hopping (FH) technology. The FH radio links are invariably duplex in nature employing either Time Division Duplexing (TDD) or Frequency Division Duplexing (FDD) techniques. The choice of appropriate duplexing technique has a direct bearing on resilience to combat jamming (and hence resultant effective throughput) as well as impacts achievable ranges. This paper examines the impact of the choice of duplexing method under these conditions. In particular we examine the Jammer to Signal (J/S) strength required to disrupt the Radio link under each of these duplexing schemes. We also provide useful insight into maximum throughput achievable for various FEC schemes under jamming conditions.
15 Apr 2018
TL;DR: It is shown that the average sync frame length must scale with the entropy of v and/or logarithm of cardinality of the sample space of 7 and the scaling needed of the sync frame energy for the asynchronous setup is derived.
Abstract: We study one shot frame synchronization in a discrete memoryless channel where, a sync frame of known length N v symbols is transmitted at a random time v. For a random misalignment 7 between the transmitter clock and the receiver clock, we seek to characterize the sync frame (length) necessary for error-free frame synchronization. We assume that the random times v and 7 are independent and have a general distribution. Using a variable length sync word and a multi-stage sequential decoder for the setup, we derive necessary and sufficient scaling of the average sync frame length for asymptotic error-free frame synchronization. In particular, we show that the average sync frame length must scale with the entropy of v and/or logarithm of cardinality of the sample space of 7. We illustrate our results for the AWGN channel and also derive the scaling needed of the sync frame energy for the asynchronous setup.