2 results about "Mean square" patented technology

The invention discloses a method for maintaining liquid crystal screen brightness uniformity. The method comprises steps of obtaining backlight source detection data of each sensor hrough uniformly arranging a plurality of sensors inside a backlight module of a display device, determining an average value and a mean square error of all backlight brightness detection data; determining an absolute difference value between the backlight source brightness detection data and the average value when the mean square error is greater than a first threshold value; taking a sensor that the backlight source brightness detection data belongs to as a target user, wherein a maximum absolute difference value corresponds to the backlight source brightness detection data; and regulating the backlight brightness according to the backlight source brightness detection data of the target sensor and other sensors adjacent to the target sensor. As a result, the method and the display device can always guarantee uniformity of the display screen brightness under an application scene where the backlight brightness uniformity requirement for the displays screen is very high.

The invention discloses a self-adaptive signal processing method based on a polymorphic variable step size minimum mean square, which relates to the technical field of digital signal processing, and is characterized by comprising the following steps of: in a first stage, calculating a steady-state MSD between a filter coefficient W(n) and an optimal filter coefficient H(n), and selecting a largerstep size [mu]1 to obtain a higher convergence speed, calculating according to the step length factor mu1 to obtain the steady state MSD1 of the initial state in the first stage; in the third stage, calculating the minimum final steady-state MSDmin value is calculated, and calculating the optimal [mu] opt according to the final steady-state MSDmin value; in the second stage, adding a plurality oftransient states between the initial state MSD1 and the final state MSDmin, adjusting the transient state step length factor according to the multiple factor of the step length factor mu1 and the optimal step length factor muopt, calculating the transient state according to the transient state step length factor so that the minimum MSD value under the channel, and convergence can also be carried out at a high convergence speed, and the convergence speed is high. Besides, good performance is maintained under the condition of signal mutation of a stable channel.