Our radio physical layers are actually pretty simplistic and boring in the world right now, PSK and QAM are well defined expert representations of information to transit a wireless channel. Systems using OFDM and SC-FDMA are a bit more involved, but use some of the same constructs underneath with a bit of shuffling sub-carriers. Forward error correction (FEC), equalization, randomization, and a number of other functions are generally bolted onto this as separate and independent blocks and transforms to make up for performance properties or assumptions of each other layer in order to form an effective end-to-end system.
Enter machine learning … rethink all the things …
We’ve just pre-pubbed a paper to arXiv focusing on trying to learn entire communications systems using unsupervised reconstruction learning (autoencoders). We seek to reconstruct transmitted information bits at a receiver while introducing channel impairments in the hidden layer of the network to simulate a wireless channel. By doing this we force learned representations in the encoder and decoder to adapt jointly to optimize for reconstruction performance of the information bits (we refer to this as channel regularization). The high level design looks something like this:
We evaluate a number of different autoencoder network structures and also consider keeping the CNN layer constrained to a relatively low number of filters to emulate the relatively low number of communications symbols typically used in communications system (although this is not necessarily optimal, but helps with intuition). The structure of our DNN-CNN network candidate looks something like this:
Once we learn a transmit/receive representation in the autoencoder we can evaluate its performance across a range of channel conditions. Traditional wireless channel performance measures such as BER vs SNR and spectral efficiency can be easily compared to legacy expert modulation techniques as shown below.
We discuss a handful of other issues including how to start jointly learning synchronization methods on the front of the decoder using radio transformer networks and how to start simulating channel effects beyond simple additive Gaussian noise. I’m pretty excited about the future of this form of unsupervised communications system learning, there’s a ton of work to do to make it work way better over the air and amongst harsh channel conditions. Hoping to see what others do with this, and finalize a conference version of it for submission soon.
Check out the paper at: https://arxiv.org/abs/1608.06409