The post covers various neural network based word embedding models. Starting from the Neural Probabilistic Language Model from Bengio et al 2003, then reduction of complexity using Hierarchical softmax and Noise Contrastive Estimation. Further works like CBoW, GlOVe, Skip Gram and Negative Sampling which helped to train on much higher data.
In TETRA specifications, one of the modulation technique used is Differential Quaternary Phase Shift Keying (DQPSK). We will discuss the bit error rate with non-coherent demodulation of DQPSK in Additive White Gaussian Noise (AWGN) channel.
The earlier posts on phase noise discussed about phase noise in oscillators, conversion of phase noise profile to jitter and the impact of phase noise on the error vector magnitude (evm). This post discuss the impact of phase noise on the spectrum of the transmit waveform. A simple random QPSK modulated symbols, oversampled and passed…
Minimum shift keying (MSK) is an important concept to learn in digital communications. It is a form of continuous phase frequency shift keying . In minimum phase shift keying, two key concepts are used. (a) The frequency separation of the sinusoidals used for representing bits 1’s and 0’s are , where is the symbol period….
This is the first post in the series discussing receiver diversity in a wireless link. Receiver diversity is a form of space diversity, where there are multiple antennas at the receiver. The presence of receiver diversity poses an interesting problem – how do we use ‘effectively‘ the information from all the antennas to demodulate the…
In the past, we had discussed several posts on two transmit two receive MIMO communication, where the transmission was based on V-BLAST. The details about V-BLAST can be read from the landmark paper V-BLAST: An architeture for realizing very high data rates over the rich scattering wireless channel – P. W. Wolniansky, G. J. Foschini,…
In this post, let us try to understand Gray codes and their usage in digital communication. Quoting from Wiki entry on Gray code [Gray-Wiki], The reflected binary code, also known as Gray code after Frank Gray, is a binary numeral system where two successive values differ in only one digit. In a digital communication system,…
In May 2008, we derived the theoretical symbol error rate for a general M-QAM modulation (in Embedded.com, DSPDesignLine.com and dsplog.com) under Additive White Gaussian Noise. While re-reading that post, felt that the article is nice and warrants a re-run, using OFDM as the underlying physical layer. This post discuss the derivation of symbol error rate for a general…
From the previous posts on Linear Regression (using Batch Gradient descent, Stochastic Gradient Descent, Closed form solution), we discussed couple of different ways to estimate the parameter vector in the least square error sense for the given training set. However, how does the least square error criterion work when the training set is corrupted by…
Understanding gradients is essential in machine learning, as they indicate the direction and rate of change in the loss function with respect to model parameters. This post covers the gradients for the vanilla Linear Regression case taking two loss functions Mean Square Error (MSE) and Mean Absolute Error (MAE) as examples. The gradients computed analytically…
Hi, Those visiting the blog might have noticed a fresh look to the dspLog. This new feel is thanks to the Thesis Magazine Skin provided by FourBlogger Skins. Click here to view more details. There some more tinkering required at some places. But, in general most of the settings are taken care. Hope you like…
Thanks to the nice article from Xilinx TechXclusives [XLNX-TECH], let us try to understand the probable digital implementation of resistor-capacitor based low pass filter. Consider a simple RC filter shown in the figure below. Assuming that there is no load across the capacitor, the capacitor charges and discharges through the resistor path. Figure: RC low…
In the past, we had discussed BER for BPSK in flat fading Rayleigh channel. In this post, lets discuss a frequency selective channel with the use of Zero Forcing (ZF) equalization to compensate for the inter symbol interference (ISI). For simplifying the discussion, we will assume that there is no pulse shaping at the transmitter….
Let us derive the theoretical 16QAM bit error rate (BER) with Gray coded constellation mapping in additive white Gaussian noise conditions. Further, the Matlab/Octave simulation script can be used to confirm that the simulation is in good agreement with theory.