Deep reinforcement learning Beidou navigation chip design method based on sparse representation driving

• Main Authors: LI ZHENNI, MOTOHIDE, TANG JIANHAO, XIE SHENGLI, ZHENG SHAOLONG
• Format: Patent
• Language: Chinese; English
• Published: 02.08.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention relates to a Beidou navigation chip design method for deep reinforcement learning based on sparse representation driving. The Beidou navigation chip design method comprises the steps that graph embedding, current macro-cell embedding and netlist metadata embedding are obtained based on macro-cell characteristics, netlist graph information and netlist metadata of a chip, and a three-dimensional state space is obtained through a second full-connection network; adding regular particles to neurons of the last hidden layer of the value network for sparse constraint to obtain a value network based on sparse representation; inputting the three-dimensional state space into a value network based on sparse representation to obtain a value function; and inputting the three-dimensional state space into a strategy network and obtaining an optimal layout strategy of the Beidou navigation chip macro-cell under the guidance of the value function. The value network based on sparse representation alleviates the p