专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US12021572B2 Systems and methods for wireless signal configuration by a neural network 有权
公开(公告)号：US12021572B2
公开(公告)日：2024-06-25
申请号：US18080932
申请日：2022-12-14
申请人： Yiqun Ge , Wuxian Shi , Wen Tong
发明人： Yiqun Ge , Wuxian Shi , Wen Tong
IPC分类号： H04B17/26 , H04B17/373 , H04L25/02 , H04L25/03 , H04W24/10
CPC分类号： H04B17/26 , H04B17/373 , H04L25/0254 , H04L25/03165 , H04W24/10
摘要： A wireless network can generate candidate signal configurations for physical transmissions to or from a user equipment (UE) in a radio environment. The generation of candidate signal configurations can be performed using a first neural network that is associated with the UE. These signal configurations can then be evaluated using a second neural network that is associated with the radio environment. The second neural network can be trained using measurements from previous physical transmissions in the radio environment. The trained second neural network generates a reward value that is associated with the candidate signal configurations. The first neural network is then trained using the reward values from the second neural network to produce improved candidate signal configurations. When a signal configuration that produces a suitable reward value is generated, this signal configuration can be used for the physical transmission in the radio environment.

2. 发明申请

US20210297094A1 METHOD, SYSTEM, AND APPARATUS FOR A SEGMENTED POLARIZATION-ADJUSTED CONVOLUTIONAL (PAC) CODE 有权
公开(公告)号：US20210297094A1
公开(公告)日：2021-09-23
申请号：US16825319
申请日：2020-03-20
申请人： LOUIS-PHILIPPE HAMELIN , HARSH AURORA , YIQUN GE
发明人： LOUIS-PHILIPPE HAMELIN , HARSH AURORA , YIQUN GE
IPC分类号： H03M13/29 , H04L1/00 , H03M13/13 , H03M13/27
摘要： A codeword is generated based on a segmentation transform and a Polarization-Assisted Convolutional (PAC) code that includes an outer convolutional code and a polar code, and based on separate encoding of respective different segments of convolutionally encoded input bits according to the polar code. Each segment of the respective segments includes multiple bits of the convolutionally encoded input bits for which the separate encoding of the segment is independent of the separate encoding of other segments. Separate decoding may be applied to segments of such a codeword to decode convolutionally encoded input bits corresponding to the separately encoded segments of the convolutionally encoded input bits.

3. 发明授权

US10447435B2 Reduced-stage polar decoding 有权
公开(公告)号：US10447435B2
公开(公告)日：2019-10-15
申请号：US15651390
申请日：2017-07-17
申请人： Wuxian Shi , Yiqun Ge
发明人： Wuxian Shi , Yiqun Ge
IPC分类号： H04L1/00 , H03M13/45 , H03M13/13 , H03M13/00 , H04W88/02 , H04W88/08
摘要： In reduced-stage polar decoding, a received word that is based on an N-bit codeword of a polar code is decoded using fewer than log2N Log Likelihood Ratio (LLR) stages. Decoding uses a reduced stage decoding configuration. In an embodiment, such a configuration includes at least one higher-order LLR stage with nodes implementing functions that are based on a combination of lower-order polar code kernels.

4. 发明申请

US20180278389A1 METHODS AND APPARATUS FOR CODING SUB-CHANNEL SELECTION 审中-公开
公开(公告)号：US20180278389A1
公开(公告)日：2018-09-27
申请号：US15901238
申请日：2018-02-21
申请人： LOUIS-PHILIPPE HAMELIN , YIQUN GE , WUXIAN SHI , RAN ZHANG , NAN CHENG
发明人： LOUIS-PHILIPPE HAMELIN , YIQUN GE , WUXIAN SHI , RAN ZHANG , NAN CHENG
IPC分类号： H04L5/00 , H04L1/00
CPC分类号： H04L5/0046 , H03M13/033 , H03M13/13 , H04L1/0009 , H04L1/0041 , H04L1/0057 , H04L1/0068
摘要： Coding sub-channel selection involves, in an embodiment, determining, from sub-channels that are defined by a code and that have associated reliabilities for input bits at input bit positions, a first number of the sub-channels to carry bits that are to be encoded. A second number of the sub-channels, greater than the first number, are selected. The second number of sub-channels are selected to provide exactly the first number sub-channels to be available to carry the bits that are to be encoded.

5. 发明申请

US20180191465A1 APPARATUS AND METHODS FOR TRAINING-BASED CHANNEL CODE DESIGN 审中-公开
公开(公告)号：US20180191465A1
公开(公告)日：2018-07-05
申请号：US15852174
申请日：2017-12-22
申请人： HAMID SABER , YIQUN GE
发明人： HAMID SABER , YIQUN GE
IPC分类号： H04L1/00 , H04J13/10
CPC分类号： H04L1/0076 , H04J13/004 , H04J13/10 , H04L1/0041 , H04L1/0045 , H04L1/0057
摘要： One or more codewords are generated by encoding input bits at input bit positions onto sub-channels that are provided by a code. Each of the sub-channels has an associated reliability of correct decoding of an input bit at an input bit position. Each codeword is transmitted to a decoder, and a word based on each transmitted codeword is received at the decoder. Each received word is decoded, and the reliabilities of the sub-channels are determined based on decoded bits and known input bits from which each codeword was generated. An indication that is based on the determined reliabilities of the sub-channels, such as an indication of the determined reliabilities, is transmitted to the encoder, and may be used by the encoder in selecting the sub-channels for encoding subsequent input bits, for example.

6. 发明申请

US20170155405A1 SIGNATURE-ENABLED POLAR ENCODER AND DECODER 审中-公开
公开(公告)号：US20170155405A1
公开(公告)日：2017-06-01
申请号：US15364521
申请日：2016-11-30
申请人： YIQUN GE , WUXIAN SHI
发明人： YIQUN GE , WUXIAN SHI
IPC分类号： H03M13/13 , H04L1/00
摘要： A signature-enabled Polar code encoder and decoder are provided. Signature bits are inserted in some unreliable bit positions. Different signature bits are inserted for different receivers. For a given codeword, only the receiver with knowledge of the signature can decode the codeword. Cyclic redundancy check (CRC) bits may be included in the input vector to assist in decoding.

7. 发明申请

US20170019222A1 METHOD OF REDUCED STATE DECODING AND DECODER THEREOF 审中-公开
标题翻译：减少状态解码的方法及其解码器
公开(公告)号：US20170019222A1
公开(公告)日：2017-01-19
申请号：US15203498
申请日：2016-07-06
申请人： Yiqun Ge , Wuxian Shi
发明人： Yiqun Ge , Wuxian Shi
IPC分类号： H04L5/00 , H04B7/26
CPC分类号： H04L5/0007 , H04B7/2628
摘要： Methods and devices are disclosed for receiving and decoding sparsely encoded data sequences using a message passing algorithm (MPA) or maximum likelihood sequence estimation (MLSE). Such data sequences may be used in wireless communications systems supporting multiple access, such as sparse code multiple access (SCMA) systems. The Methods and devices reduce the number of states in a search space for each received signal and associated function node based on a search threshold based on a characteristic related to the received signal and/or to a quality of a resource element over which the received signal is transmitted.
摘要翻译：公开了使用消息传递算法（MPA）或最大似然序列估计（MLSE）来接收和解码稀疏编码的数据序列的方法和装置。这样的数据序列可以用于支持多个接入的无线通信系统，例如稀疏码多址（SCMA）系统。方法和装置基于与接收信号相关的特征和/或资源元素的质量，基于搜索阈值来减少每个接收信号和相关联功能节点在搜索空间中的状态数量，被传送。

8. 发明申请

US20070245209A1 DECODING METHOD FOR TAIL-BITING CONVOLUTIONAL CODES USING A SEARCH DEPTH VITERBI ALGORITHM 有权
标题翻译：使用搜索深度维特比算法解码转换代码的解码方法
公开(公告)号：US20070245209A1
公开(公告)日：2007-10-18
申请号：US11687543
申请日：2007-03-16
申请人： Wuxian SHI , Juan DU , Yiqun GE , Guobin SUN
发明人： Wuxian SHI , Juan DU , Yiqun GE , Guobin SUN
IPC分类号： H03M13/00
CPC分类号： H03M13/3938 , H03M13/413 , H03M13/6362 , H03M13/6502 , H03M13/6505 , H04L1/0054 , H04L1/0059
摘要： A method for decoding tail-biting convolutional codes. The method includes initializing a correction depth, selecting a first starting state from a set of encoding states, and initializing a metric value for the selected starting state as zero and the other states as infinity. The input bit stream is read and a Search Depth Viterbi algorithm (SDVA) is performed to determine path metrics and identify a minimum-metric path. The ending state for the minimum-metric path is determined and the output for this ending state is identified as “previous output.” A second starting state is set to the ending state of the minimum-metric path, and symbols equal to the correction depth from the previous output are read. The SDVA is performed on the second set of read symbols to generate a corrected output. A decoded output is generated by replacing symbols at the beginning of the previous output with the corrected output.
摘要翻译：一种用于解码尾巴卷积码的方法。该方法包括初始化校正深度，从一组编码状态中选择第一起始状态，以及将所选择的起始状态的度量值初始化为零，将其他状态初始化为无穷大。读取输入比特流，执行搜索深度维特比算法（SDVA）以确定路径度量并识别最小度量路径。确定最小度量路径的结束状态，将该结束状态的输出识别为“先前输出”。第二起始状态被设置为最小度量路径的结束状态，并且读取等于来自先前输出的校正深度的符号。在第二组读符号上执行SDVA以产生校正输出。通过使用校正输出替换先前输出开始处的符号来产生解码输出。

9. 发明授权

US11533115B2 Systems and methods for wireless signal configuration by a neural network 有权
公开(公告)号：US11533115B2
公开(公告)日：2022-12-20
申请号：US16412908
申请日：2019-05-15
申请人： Yiqun Ge , Wuxian Shi , Wen Tong
发明人： Yiqun Ge , Wuxian Shi , Wen Tong
IPC分类号： H04B17/26 , H04L25/02 , H04W24/10 , H04B17/373 , H04L25/03
摘要： A wireless network can generate candidate signal configurations for physical transmissions to or from a user equipment (UE) in a radio environment. The generation of candidate signal configurations can be performed using a first neural network that is associated with the UE. These signal configurations can then be evaluated using a second neural network that is associated with the radio environment. The second neural network can be trained using measurements from previous physical transmissions in the radio environment. The trained second neural network generates a reward value that is associated with the candidate signal configurations. The first neural network is then trained using the reward values from the second neural network to produce improved candidate signal configurations. When a signal configuration that produces a suitable reward value is generated, this signal configuration can be used for the physical transmission in the radio environment.

10. 发明授权

US10979202B2 Neural-network-based distance metric for use in a communication system 有权
公开(公告)号：US10979202B2
公开(公告)日：2021-04-13
申请号：US16534653
申请日：2019-08-07
申请人： Yiqun Ge , Wuxian Shi , Wen Tong
发明人： Yiqun Ge , Wuxian Shi , Wen Tong
IPC分类号： H04L5/00 , G06N3/08 , G06F17/18 , G06N20/20 , G06N3/04
摘要： Methods and apparatuses for measuring a distance between two signal distributions in a common signal space are described. A measurement network is configured to receive first and second signal distributions as input and output a measurement of a Wasserstein distance between the first and second signal distributions. The measurement network may be implemented using a neural network.

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式