TianLin0509 / DNN_detection_via_keras

This is the simplest implementation of Power of Deep Learning for Channel Estimation and Signal Detection in OFDM Systems using keras.
155 stars 65 forks source link

导频插入方式 #9

Open Yonglinz77 opened 3 years ago

Yonglinz77 commented 3 years ago

当插入导频长度为8时,注意到导频长度只占子载波(长度为64)的一部分,且插入方式为头部插入,其余部分用随机生成的数据补位,这样是否会存在导频只能估计一段频域的信道信息,我发现在原作者的代码中也存在这个问题,或许在dl的框架中这种插入导频的方式可行,但很好奇与之对比的传统方法又是如何保证信道估计的准确性的?是否应该将其改为梳状导频插入方式,如有理解上的错误,还望指点!感谢!

TianLin0509 commented 3 years ago

我并不太清楚作者对比的传统方案中是如何实现的, 但似乎看起来运行结果非常差。 而至于作者自己的DNN网络,应该是均匀地插入8个导频。

BassantTolba1234 commented 3 years ago

Dear Sir, I really appreciate your hard work..and please I have a question, in main file line 48, why do you make (model.evaluate) on function called (validation_gen) which exactly contains the generation of training datasets not the test datasets??..

why did not you apply evaluation the model on testing datasets you provided , instead of training datasets ? I'm waiting for your reply.. thanks in advance.

BassantTolba1234 commented 3 years ago

Dear sir, please what is the value of SNR (Signal Noise to ratio used in training the model ) ??

crazyrayLing commented 8 months ago

你好 @Yonglinz77 ,作者的代码中的导频是均匀分布的状态,但是我对比于直接头部插入,DL跑出来的效果居然比均匀分布导频的效果要更好,而头部插入导频也正是你说的,只是学习了某一段频率上的特点,我很疑惑,请问你现在有想法了吗,还请指教!