公开(公告)号：WO2015065686A3

公开(公告)日：2015-06-25

申请号：PCT/US2014060234

申请日：2014-10-13

Applicant: QUALCOMM INC

Inventor： GUPTA VIKRAM ,  TOWAL REGAN BLYTHE ,  CHAN VICTOR HOKKIU ,  PATWARDHAN RAVINDRA MANOHAR ,  LEVIN JEFFREY

IPC: G06N3/04

CPC classification number: G06N3/049 ,  G06N3/0454 ,  G06N3/08

Abstract: Certain aspects of the present disclosure provide methods and apparatus for creating tags (static or dynamic) for input/output classes of a neural network model using supervised learning. The method includes augmenting a neural network model with a plurality of neurons and training the augmented network using spike timing dependent plasticity (STDP) to determine one or more tags.

Abstract translation: 本公开的某些方面提供了使用监督学习为神经网络模型的输入/输出类创建标签（静态或动态）的方法和装置。 该方法包括用多个神经元增强神经网络模型，并使用尖峰定时相关可塑性（STDP）训练增强网络来确定一个或多个标签。

公开(公告)号：WO2009137650A8

公开(公告)日：2010-12-02

申请号：PCT/US2009043095

申请日：2009-05-07

Applicant: QUALCOMM INC ,  YAVUZ MEHMET ,  TINNAKORNSRISUPHAP PEERAPOL ,  REZAIIFAR RAMIN ,  GUPTA VIKRAM

Inventor： YAVUZ MEHMET ,  TINNAKORNSRISUPHAP PEERAPOL ,  REZAIIFAR RAMIN ,  GUPTA VIKRAM

IPC: H04W28/10

CPC classification number: H04W28/14 ,  H04L47/10 ,  H04L47/14 ,  H04L47/26 ,  H04L47/30

Abstract: Techniques for enhanced backhaul flow control are provided. In an exemplary embodiment, a backhaul control system is described that comprises a base station controller (BSC), a backhaul network, and a base transceiver station (BTS). Each is responsive to data and messaging transmitted and received. In one aspect, the BTS includes a queue and a controller. The amount of data in a queue is adjusted by a controller based upon calculating a target queue size value. The controller non-uniformly adjusts the amount of data in a queue based upon a target queue size value which is based upon communication system parameters. The target queue size and amount of data in a queue is adjusted so as to reduce buffer underrun, decrease system latency, and increase communication system throughput.

Abstract translation: 提供了用于增强回程流量控制的技术。 在示例性实施例中，描述了一种回程控制系统，其包括基站控制器（BSC），回程网络和基站收发器（BTS）。 每个都响应于发送和接收的数据和消息。 在一个方面，BTS包括队列和控制器。 基于计算目标队列大小值，控制器调整队列中的数据量。 控制器基于基于通信系统参数的目标队列大小值，不均匀地调整队列中的数据量。 调整目标队列大小和队列数据量，以减少缓冲区欠载，减少系统延迟并提高通信系统吞吐量。

公开(公告)号：WO2010033835A3

公开(公告)日：2010-05-14

申请号：PCT/US2009057527

申请日：2009-09-18

Applicant: QUALCOMM INC ,  PALANKI RAVI ,  AGASHE PARAG A ,  GUPTA VIKRAM ,  GUPTA RAJARSHI ,  BHUSHAN NAGA

Inventor： PALANKI RAVI ,  AGASHE PARAG A ,  GUPTA VIKRAM ,  GUPTA RAJARSHI ,  BHUSHAN NAGA

IPC: H04W56/00

CPC classification number: H04W56/0015

Abstract: A method for synchronizing a wireless communication system is disclosed. A silence duration for a base station is determined based on the time required for a neighbor base station to obtain or maintain synchronization. All transmissions from the base station are ceased for the silence duration. Multiple base stations level may cease transmissions at the same time, thus mitigating interference.

Abstract translation: 公开了一种用于同步无线通信系统的方法。 基于相邻基站获得或维持同步所需的时间来确定基站的静默持续时间。 来自基站的所有传输在静默持续时间内停止。 多个基站级别可以同时停止传输，从而减轻干扰。

公开(公告)号：WO2015148254A3

公开(公告)日：2016-01-07

申请号：PCT/US2015021428

申请日：2015-03-19

Applicant: QUALCOMM INC

Inventor： AGRAWAL PULKIT ,  MAJUMDAR SOMDEB ,  GUPTA VIKRAM

IPC: G06N3/04 ,  G06K9/46

CPC classification number: G06K9/4647 ,  G06K9/4619 ,  G06K9/6273 ,  G06N3/049

Abstract: A method for generating a histogram in a spiking neural network includes counting spikes associated with a latency encoded representation of an object. The method also includes generating the histogram based on the spike count.

Abstract translation: 用于在加标神经网络中生成直方图的方法包括计数与对象的等待时间编码表示相关联的峰值。 该方法还包括基于尖峰计数生成直方图。

公开(公告)号：WO2015088774A2

公开(公告)日：2015-06-18

申请号：PCT/US2014067343

申请日：2014-11-25

Applicant: QUALCOMM INC

Inventor： GUPTA VIKRAM ,  LEVIN JEFFREY ALEXANDER ,  LIAO EDWARD HANYU

IPC: G06N3/04

CPC classification number: G06N3/08 ,  G06N3/049

Abstract: A method for pattern recognition in a spiking neural network robust to initial network conditions includes creating a set of diverse neurons in a first layer to increase a diversity in a set of spike timings. An input corresponding to a pattern plus noise is presented at an input layer and represented as spikes. The spikes are received at the first layer and spikes are produced at the first layer based on the received spikes. The method also includes updating a weight of each synapse between an input layer neuron and an output layer neuron based on a spike timing difference between a spike at the input layer neuron and a spike at the output layer neuron. Further, the method includes classifying a spike pattern represented by a set of inter-spike intervals, regardless of noise in the spike pattern.

Abstract translation: 在对初始网络条件稳健的加标神经网络中的模式识别的方法包括在第一层中创建一组不同的神经元以增加一组尖峰定时的分集。 对应于图案加噪声的输入在输入层处呈现并表示为尖峰。 尖峰在第一层被接收，并且基于接收的尖峰在第一层产生尖峰。 该方法还包括基于输入层神经元的尖峰与输出层神经元的尖峰之间的尖峰定时差来更新输入层神经元和输出层神经元之间的每个突触的权重。 此外，该方法包括分类由一组间穗间隔表示的尖峰图案，而不管尖峰图案中的噪声如何。