Design and Prototyping of Neural Network Compression for Non-Orthogonal IoT Signals

The non-orthogonal IoT signal, following the bandwidth compression spectrally efficient frequency division multiplexing (SEFDM) characteristics, can bring benefits in enhanced massive device connections, signal coverage extension and data rate increase, but at the cost of computational complexity. Resource-constrained IoT devices have limited memory storage and complex signal processing is not allowed. Machine learning can simplify signal detection by training a general data-driven signal detection model. However, fully connected neural networks would introduce processing latency and extra power consumption. Therefore, the motivation of this work is to investigate different neural network compression schemes for system simplification. Three compression strategies are studied including topology compression, weight compression and quantization compression. These methods show efficient neural network compression with trade-offs between computational complexity and bit error rate (BER) performance. Practical neural network prototyping is evaluated as well on a software defined radio (SDR) platform. Results show that the practical weight compression neural network can achieve similar performance as the fully connected neural network but with great resource saving.

[1]  Yoshua Bengio,et al.  BinaryNet: Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1 , 2016, ArXiv.

[2]  Mohsen Guizani,et al.  Deep Learning for IoT Big Data and Streaming Analytics: A Survey , 2017, IEEE Communications Surveys & Tutorials.

[3]  Izzat Darwazeh,et al.  Uplink Narrowband IoT Data Rate Improvement: Dense Modulation Formats or Non-Orthogonal Signal Waveforms? , 2018, 2018 IEEE 29th Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC).

[4]  Timothy J. O'Shea,et al.  Deep Learning Based MIMO Communications , 2017, ArXiv.

[5]  Jakob Hoydis,et al.  An Introduction to Deep Learning for the Physical Layer , 2017, IEEE Transactions on Cognitive Communications and Networking.

[6]  Song Han,et al.  Deep Compression: Compressing Deep Neural Network with Pruning, Trained Quantization and Huffman Coding , 2015, ICLR.

[7]  Izzat Darwazeh,et al.  Non-Orthogonal Narrowband Internet of Things: A Design for Saving Bandwidth and Doubling the Number of Connected Devices , 2018, IEEE Internet of Things Journal.

[8]  Geoffrey Ye Li,et al.  Power of Deep Learning for Channel Estimation and Signal Detection in OFDM Systems , 2017, IEEE Wireless Communications Letters.

[9]  Tao Zhang,et al.  A Survey of Model Compression and Acceleration for Deep Neural Networks , 2017, ArXiv.

[10]  Yixin Chen,et al.  Compressing Neural Networks with the Hashing Trick , 2015, ICML.

[11]  Christos Masouros,et al.  Waveform and Space Precoding for Next Generation Downlink Narrowband IoT , 2019, IEEE Internet of Things Journal.

[12]  Jun Xu,et al.  Narrowband Internet of Things: Evolutions, Technologies, and Open Issues , 2018, IEEE Internet of Things Journal.

[13]  Stephan ten Brink,et al.  Deep Learning Based Communication Over the Air , 2017, IEEE Journal of Selected Topics in Signal Processing.

[14]  Izzat Darwazeh,et al.  Deep Learning for Interference Cancellation in Non-Orthogonal Signal Based Optical Communication Systems , 2018, 2018 Progress in Electromagnetics Research Symposium (PIERS-Toyama).