The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Neural network theory combining with multi-user detection, viz., multi-user detection based on neural network has been a new algorithm in recent years. An adaptive multi-user detection algorithm based on diagonal recursion neural network was proposed, the cost function and transfer function were chosen, the iteration formula was deduced, the computer simulation was made, and the simulation result...
An improved space-time MOE blind multi-user detection algorithm based on RLS was proposed, which is got through expanding the time-only MOE blind multi-user detection algorithm to space-time domain. The structure of multi-user detector was analyzed, and the iteration equation of weight was deduced. This algorithm can avoid using direct eigenvalue decomposition at each iterations, so its operation...
The influence of momentum term to blind equalization algorithm was analyzed in this paper. A new neural network blind equalization algorithm based on variable momentum factor was proposed, which adopted the non-linear function of mean square error as the variable momentum factor. The simulation shows that the new algorithm has faster convergence rate, smaller state residual error and lower bit error...
A variable step size blind equalization algorithm based on minimum error probability(MEP) was proposed. The cost function based on minimum error probability was founded. And the iterative equations of the variable step size blind equalization algorithm were deduced by the steepest gradient. Simulations demonstrate that the novel proposed algorithm possessed faster convergence and smaller steady MSE...
A feed-forward neural network blind multi-user detection algorithm was proposed. Feed-forward neural network and constant modulus algorithm (CMA) were consociated to complete blind multi-user detection. A constant modulus cost function firstly was constructed and the cost function with restrict condition was optimized by augmented Lagrange method. Blind multi-user detection algorithm was realized...
As a key technology of digital broadcast and TV, blind equalization overcomes inter-symbol interference to improve the effect of receiving signals. A new QAM blind equalization algorithm based on fuzzy neural network classifier was proposed. Simulation shows that the new algorithm improves convergence speed and reduces residual error and BER (bit error ratio).
As a key technology in the digital communication system, blind equalization algorithm based on fuzzy neural network classifier is proposed. The algorithm overcomes bad judgment error. Channel estimation algorithm and fuzzy neural network classifier were combined to carry out equalization. The primary signal was attained by de-convolution. Judgment range of fuzzy neural network was adjusted dynamically...
In order to overcome the influence of iteration step-size to algorithm performance, an improved neural network blind equalization algorithm based on momentum term is proposed in this paper. The simulation shows that the improved algorithm has faster convergence rate and lower bit error rate.
In this paper, an improved blind equalization algorithm based on multiple-layer forward neural network is proposed, and a new transfer function for the neural network is designed. The computer simulations show that the improved algorithm have faster convergence speed and smaller bit error rate than traditional forward neural network algorithm.
An improved space-time MOE blind multiuser detection algorithm is proposed in this paper, which is got through expanding the time-only MOE blind multiuser detection algorithm to space-time domain. The simulation shows that the improved MOE blind multiuser detection algorithm has better performance to combat multiple access interference than the time-only MOE blind multiuser detection algorithm.
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.