Short Codes with Near-ML Universal Decoding: Are Random Codes Good Enough?

Abstract. The ionospheric D-layer is a narrow bandwidth radar target often with a very small scattering cross section. The target autocorrelation function can be obtained by transmitting a series of relatively short coded pulses and computing the correlation between data obtained from different pulses. The spatial resolution should be as high as possible and the spatial side lobes of the codes used should be as small as possible. However, due to the short pulse repetition period (in the order of milliseconds) at any instant, the radar receives detectable scattered signals not only from the pulse illuminating the D-region but also from 3–5 ambiguous-range pulses, which makes it difficult to produce a reliable estimate near zero lag of the autocorrelation function. A new experimental solution to this measurement problem, using a selected set of 40-bit random codes with 4 µs elements giving 600 m spatial resolution is presented. The zero lag is approximated by dividing the pulse into two 20-bit codes and computing the correlation between those two pulses. The lowest altitudes of the E-layer are measured by dividing the pulse into 5 pieces of 8 bits, which allows for computation of 4 lags. In addition, coherent integration of data from four pulses is used for obtaining separately the autocorrelation function estimate for the lowest altitudes and in cases when the target contains structures with a long coherence time. Design details and responses of the experiment are given, and analysed test data are shown.Key words. Radio science (signal processing); Ionosphere (plasma temperature and density; instruments and techniques)

Download Full-text

Deep Polarized Network for Supervised Learning of Accurate Binary Hashing Codes

Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence ◽

10.24963/ijcai.2020/115 ◽

2020 ◽

Cited By ~ 1

Author(s):

Lixin Fan ◽

Kam Woh Ng ◽

Ce Ju ◽

Tianyu Zhang ◽

Chee Seng Chan

Keyword(s):

Large Deviations ◽

Hamming Distance ◽

State Of The Art ◽

Metric Learning ◽

Binary Codes ◽

Distance Metric Learning ◽

Learning Framework ◽

Label Information ◽

Learning To Hash ◽

Random Codes

This paper proposes a novel deep polarized network (DPN) for learning to hash, in which each channel in the network outputs is pushed far away from zero by employing a differentiable bit-wise hinge-like loss which is dubbed as polarization loss. Reformulated within a generic Hamming Distance Metric Learning framework [Norouzi et al., 2012], the proposed polarization loss bypasses the requirement to prepare pairwise labels for (dis-)similar items and, yet, the proposed loss strictly bounds from above the pairwise Hamming Distance based losses. The intrinsic connection between pairwise and pointwise label information, as disclosed in this paper, brings about the following methodological improvements: (a) we may directly employ the proposed differentiable polarization loss with no large deviations incurred from the target Hamming distance based loss; and (b) the subtask of assigning binary codes becomes extremely simple --- even random codes assigned to each class suffice to result in state-of-the-art performances, as demonstrated in CIFAR10, NUS-WIDE and ImageNet100 datasets.

Download Full-text

A Decoupling Approach to the Quantum Capacity

Open Systems & Information Dynamics ◽

10.1142/s1230161208000043 ◽

2008 ◽

Vol 15 (01) ◽

pp. 7-19 ◽

Cited By ~ 68

Author(s):

Patrick Hayden ◽

Michał Horodecki ◽

Andreas Winter ◽

Jon Yard

Keyword(s):

Short Proof ◽

Achievable Rate ◽

Projective Measurement ◽

Tensor Power ◽

Quantum Capacity ◽

Decoupling Approach ◽

Coherent Information ◽

Random Codes ◽

Channel Environment ◽

Noisy Quantum Channel

We give a short proof that the coherent information is an achievable rate for the transmission of quantum information through a noisy quantum channel. Our method is to produce random codes by performing a unitarily covariant projective measurement on a typical subspace of a tensor power state. We show that, provided the rank of each measurement operator is sufficiently small, the transmitted data will, with high probability, be decoupled from the channel environment. We also show that our construction leads to random codes whose average input is close to a product state and outline a modification yielding unitarily invariant ensembles of maximally entangled codes.

Download Full-text

Network Coding for Multi-Hop Wireless Networks

Cooperative Communications for Improved Wireless Network Transmission ◽

10.4018/978-1-60566-665-5.ch006 ◽

2010 ◽

pp. 187-205

Author(s):

Meng Yu ◽

Jing (Tiffany) Li ◽

Haidong Wang

Keyword(s):

Wireless Networks ◽

Network Coding ◽

Mobile Networks ◽

Ad Hoc ◽

Graph Matching ◽

Hybrid Network ◽

Time Varying ◽

General Network ◽

Coded Cooperation ◽

Random Codes

We consider practical network coding, a useful generalization of routing, in multi-hop multicast wireless networks. The model of interest comprises a set of nodes transmitting data wirelessly to a set of destinations across an arbitrary, unreliable, and possibly time-varying network. This model is general and subsumes peer-to-peer, ad-hoc, sensory, and mobile networks. It is first shown that, in the singlehop case, the idea of adaptively matching code-on-graph with network-on-graph, first developed in the adaptive-network-coded-cooperation (ANCC) protocol, provides a significant improvement over the conventional strategies. To generalize the idea to the multi-hop context, we propose to transform an arbitrarily connected network to a possibly time-varying “trellis network,” such that routing design for the network becomes equivalent to path discovery in the trellis. Then, exploiting the distributed, real-time graph-matching technique in each stage of the trellis, a general network coding framework is developed. Depending on whether or not the intermediate relays choose to decode network codes, three practical network coding categories, progress network coding, concatenated network coding and hybrid network coding, are investigated. Analysis shows that the proposed framework can be as dissemination-efficient as those with random codes, but only more practical.

Download Full-text