search

JungangGe / DRL_for_DDBC

Simulation codes for the manuscript "Deep Reinforcement Learning for Distributed Dynamic MISO Downlink-Beamforming Coordination" submitted to IEEE Transactions on Communications
38 stars 19 forks source link

DRL_for_DDBC variable: "ir_change" issue #1

Open hanchi-gao opened 3 years ago

hanchi-gao commented 3 years ago

Hi Jungang

I am a beginner of AI

  1. I don't understand what is "ir_change" with \DRL_for_DDBC-master\code\train_manner\DTDE\DRL_for_DDBC\drl.py . cn.update(ir_change=False, actions=actions) cn.update(ir_change=True))

why use the ir_change= False & True to update cn?

  1. variable of channel.py have a 11 or 01 ? self.r_power11 self.r_power01

What do those two numbers mean?

JungangGe commented 3 years ago

Hi Jungang

I am a beginner of AI

  1. I don't understand what is "ir_change" with \DRL_for_DDBC-master\code\train_manner\DTDE\DRL_for_DDBC\drl.py . cn.update(ir_change=False, actions=actions) cn.update(ir_change=True))

why use the ir_change= False & True to update cn?

  1. variable of channel.py have a 11 or 01 ? self.r_power11 self.r_power01

What do those two numbers mean?

  1. The measurements at UEs, e.g., received power, change when BSs take actions or the channels change. The variable "ir_change" indicates that the environment change due to which of the above two triiger events. For example, ir_change=True means the impulse response of the channels change due to fast fading.

  2. BSs update their beamformers at a certain moment during a slot. So, a slot is devided into two parts, or two phases as defined in our paper. I used the digital numbers to index the received power in different phases of different time slots. For example, r_power11 denotes the received power in the second phase of the previous time slot, r_power10 denotes the received power in the first phase of the previous time slot. I can not remember clearly, and it should be like this, you can try to make it clear with this train of throught.

hanchi-gao commented 3 years ago

Hi Jungang I am a beginner of AI

  1. I don't understand what is "ir_change" with \DRL_for_DDBC-master\code\train_manner\DTDE\DRL_for_DDBC\drl.py . cn.update(ir_change=False, actions=actions) cn.update(ir_change=True))

why use the ir_change= False & True to update cn?

  1. variable of channel.py have a 11 or 01 ? self.r_power11 self.r_power01

What do those two numbers mean?

  1. The measurements at UEs, e.g., received power, change when BSs take actions or the channels change. The variable "ir_change" indicates that the environment change due to which of the above two triiger events. For example, ir_change=True means the impulse response of the channels change due to fast fading.
  2. BSs update their beamformers at a certain moment during a slot. So, a slot is devided into two parts, or two phases as defined in our paper. I used the digital numbers to index the received power in different phases of different time slots. For example, r_power11 denotes the received power in the second phase of the previous time slot, r_power10 denotes the received power in the first phase of the previous time slot. I can not remember clearly, and it should be like this, you can try to make it clear with this train of throught.

Thank you a lot ! This answer is very accurate ! I will continue to learn