Deep Learning Based Communication Over the Air

End-to-end learning of communications systems is a fascinating novel concept that has so far only been validated by simulations for block-based transmissions. It allows learning of transmitter and receiver implementations as deep neural networks (NNs) that are optimized for an arbitrary differentiab...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	IEEE journal of selected topics in signal processing Ročník 12; číslo 1; s. 132 - 143
Hlavní autoři:	Dorner, Sebastian, Cammerer, Sebastian, Hoydis, Jakob, Brink, Stephan ten
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	New York IEEE 01.02.2018 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:	Artificial neural networks Autoencoder communication Communication systems Communications systems Computer simulation Data transmission Deep learning end-to-end learning Hardware modulation neural network Neural networks over-the-air Receivers software-defined radio Source code Synchronism Synchronization Training Transmitters
ISSN:	1932-4553, 1941-0484
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Abstract	End-to-end learning of communications systems is a fascinating novel concept that has so far only been validated by simulations for block-based transmissions. It allows learning of transmitter and receiver implementations as deep neural networks (NNs) that are optimized for an arbitrary differentiable end-to-end performance metric, e.g., block error rate (BLER). In this paper, we demonstrate that over-the-air transmissions are possible: We build, train, and run a complete communications system solely composed of NNs using unsynchronized off-the-shelf software-defined radios and open-source deep learning software libraries. We extend the existing ideas toward continuous data transmission, which eases their current restriction to short block lengths but also entails the issue of receiver synchronization. We overcome this problem by introducing a frame synchronization module based on another NN. A comparison of the BLER performance of the "learned" system with that of a practical baseline shows competitive performance close to 1 dB, even without extensive hyperparameter tuning. We identify several practical challenges of training such a system over actual channels, in particular, the missing channel gradient, and propose a two-step learning procedure based on the idea of transfer learning that circumvents this issue.
AbstractList	End-to-end learning of communications systems is a fascinating novel concept that has so far only been validated by simulations for block-based transmissions. It allows learning of transmitter and receiver implementations as deep neural networks (NNs) that are optimized for an arbitrary differentiable end-to-end performance metric, e.g., block error rate (BLER). In this paper, we demonstrate that over-the-air transmissions are possible: We build, train, and run a complete communications system solely composed of NNs using unsynchronized off-the-shelf software-defined radios and open-source deep learning software libraries. We extend the existing ideas toward continuous data transmission, which eases their current restriction to short block lengths but also entails the issue of receiver synchronization. We overcome this problem by introducing a frame synchronization module based on another NN. A comparison of the BLER performance of the "learned" system with that of a practical baseline shows competitive performance close to 1 dB, even without extensive hyperparameter tuning. We identify several practical challenges of training such a system over actual channels, in particular, the missing channel gradient, and propose a two-step learning procedure based on the idea of transfer learning that circumvents this issue. End-to-end learning of communications systems is a fascinating novel concept that has so far only been validated by simulations for block-based transmissions. It allows learning of transmitter and receiver implementations as deep neural networks (NNs) that are optimized for an arbitrary differentiable end-to-end performance metric, e.g., block error rate (BLER). In this paper, we demonstrate that over-the-air transmissions are possible: We build, train, and run a complete communications system solely composed of NNs using unsynchronized off-the-shelf software-defined radios and open-source deep learning software libraries. We extend the existing ideas toward continuous data transmission, which eases their current restriction to short block lengths but also entails the issue of receiver synchronization. We overcome this problem by introducing a frame synchronization module based on another NN. A comparison of the BLER performance of the “learned” system with that of a practical baseline shows competitive performance close to [Formula Omitted] dB, even without extensive hyperparameter tuning. We identify several practical challenges of training such a system over actual channels, in particular, the missing channel gradient, and propose a two-step learning procedure based on the idea of transfer learning that circumvents this issue.
Author	Dorner, Sebastian Hoydis, Jakob Brink, Stephan ten Cammerer, Sebastian
Author_xml	– sequence: 1 givenname: Sebastian orcidid: 0000-0002-7614-5156 surname: Dorner fullname: Dorner, Sebastian email: doerner@inue.uni-stuttgart.de organization: Institute of Telecommunications, University of Stuttgart, Stuttgart, Germany – sequence: 2 givenname: Sebastian surname: Cammerer fullname: Cammerer, Sebastian email: cammerer@inue.uni-stuttgart.de organization: Institute of Telecommunications, University of Stuttgart, Stuttgart, Germany – sequence: 3 givenname: Jakob surname: Hoydis fullname: Hoydis, Jakob email: jakob.hoydis@nokia-bell-labs.com organization: Nokia Bell Labs, Nozay, France – sequence: 4 givenname: Stephan ten surname: Brink fullname: Brink, Stephan ten email: tenbrink@inue.uni-stuttgart.de organization: Institute of Telecommunications, University of Stuttgart, Stuttgart, Germany
BookMark	eNp9kE1Lw0AQhhepYFv9A3oJeE7cz-zmWOs3hQqt52WzmeiWdlN3U8F_b9IWDx48zTC8z8zwjNDANx4QuiQ4IwQXNy-L5eI1o5jIjErFicInaEgKTlLMFR_0PaMpF4KdoVGMK4yFzAkfouwOYJvMwATv_HtyayJUybTZbHbeWdO6xifzLwhJ-wHJxIVzdFqbdYSLYx2jt4f75fQpnc0fn6eTWWpZTtq0FLWUdS1KIwDKAowpFaiCsIpyU-e15UIJwxRVPC8qLKztimWs6qYSSszG6Pqwdxuazx3EVq-aXfDdSU0xlhQXkvUpdUjZ0MQYoNbWtfun22DcWhOsezt6b0f3dvTRTofSP-g2uI0J3_9DVwfIAcAvoCjhlDH2AwT9caU
CODEN	IJSTGY
CitedBy_id	crossref_primary_10_1109_MITP_2019_2963491 crossref_primary_10_1002_ett_4220 crossref_primary_10_1049_cmu2_12076 crossref_primary_10_1016_j_comnet_2022_109367 crossref_primary_10_1109_TVT_2022_3162887 crossref_primary_10_1109_TCCN_2022_3147203 crossref_primary_10_1109_JSAC_2020_3036954 crossref_primary_10_1109_TCOMM_2023_3237259 crossref_primary_10_1109_TWC_2020_2970707 crossref_primary_10_1109_TCOMM_2023_3343787 crossref_primary_10_1109_ACCESS_2021_3108625 crossref_primary_10_1109_LCOMM_2019_2915977 crossref_primary_10_1109_TWC_2020_2976004 crossref_primary_10_1186_s13638_022_02195_3 crossref_primary_10_1109_JIOT_2023_3324667 crossref_primary_10_1109_ACCESS_2021_3117902 crossref_primary_10_1109_TSP_2018_2868322 crossref_primary_10_1109_JLT_2024_3403129 crossref_primary_10_54097_3fkgwa24 crossref_primary_10_1109_MWC_2019_1800598 crossref_primary_10_1109_LCOMM_2018_2868103 crossref_primary_10_1002_dac_5606 crossref_primary_10_1134_S1063785019070241 crossref_primary_10_1109_JPROC_2021_3055679 crossref_primary_10_1109_MBITS_2024_3504521 crossref_primary_10_1088_1757_899X_768_7_072022 crossref_primary_10_1109_ACCESS_2023_3244928 crossref_primary_10_1109_JSAC_2020_3036968 crossref_primary_10_1109_ACCESS_2019_2936796 crossref_primary_10_1109_TCOMM_2020_2977280 crossref_primary_10_1109_JSTQE_2022_3163474 crossref_primary_10_1109_TWC_2021_3054520 crossref_primary_10_1364_JOCN_492770 crossref_primary_10_1109_JSAC_2025_3559116 crossref_primary_10_1109_TWC_2020_2977340 crossref_primary_10_1109_ACCESS_2019_2926843 crossref_primary_10_1109_ACCESS_2023_3245330 crossref_primary_10_1109_TCCN_2019_2943455 crossref_primary_10_1109_TSP_2021_3071210 crossref_primary_10_1109_TVT_2023_3327276 crossref_primary_10_1109_ACCESS_2024_3522847 crossref_primary_10_1109_ACCESS_2022_3224922 crossref_primary_10_1109_TWC_2020_3040983 crossref_primary_10_1109_JIOT_2024_3487246 crossref_primary_10_1109_JLT_2023_3236400 crossref_primary_10_1109_TVT_2022_3162624 crossref_primary_10_1007_s11227_023_05355_4 crossref_primary_10_1109_ACCESS_2020_3003059 crossref_primary_10_1002_inst_12416 crossref_primary_10_1109_ACCESS_2023_3316605 crossref_primary_10_1016_j_comcom_2023_11_004 crossref_primary_10_1109_JPROC_2019_2957798 crossref_primary_10_1109_TCCN_2022_3168725 crossref_primary_10_1109_COMST_2020_2965856 crossref_primary_10_1109_TCOMM_2019_2903811 crossref_primary_10_1109_TWC_2021_3101364 crossref_primary_10_1049_el_2018_7691 crossref_primary_10_1109_TVT_2023_3237589 crossref_primary_10_1109_TWC_2025_3552836 crossref_primary_10_1109_ACCESS_2025_3584000 crossref_primary_10_1109_COMST_2020_2996032 crossref_primary_10_1109_TCOMM_2020_2998538 crossref_primary_10_3390_s25165048 crossref_primary_10_3390_electronics12020253 crossref_primary_10_1109_TETCI_2018_2868474 crossref_primary_10_1109_ACCESS_2019_2916701 crossref_primary_10_1016_j_rineng_2025_105409 crossref_primary_10_1109_TVLSI_2022_3225505 crossref_primary_10_1109_TCCN_2022_3164901 crossref_primary_10_1016_j_neucom_2022_09_088 crossref_primary_10_1109_COMST_2023_3249835 crossref_primary_10_1109_JSAC_2023_3242701 crossref_primary_10_1109_TWC_2021_3108666 crossref_primary_10_1109_TWC_2020_3034610 crossref_primary_10_1109_TSP_2020_3035832 crossref_primary_10_1186_s13677_020_00168_9 crossref_primary_10_1007_s12243_025_01087_z crossref_primary_10_1109_ACCESS_2024_3524571 crossref_primary_10_1109_TSP_2019_2899805 crossref_primary_10_1109_ACCESS_2019_2925569 crossref_primary_10_1109_COMST_2023_3243918 crossref_primary_10_1016_j_phycom_2023_102041 crossref_primary_10_1109_TWC_2019_2924220 crossref_primary_10_1109_JIOT_2020_3047613 crossref_primary_10_1109_LCOMM_2022_3207379 crossref_primary_10_1109_JPROC_2021_3053601 crossref_primary_10_1109_ACCESS_2021_3085901 crossref_primary_10_1109_ACCESS_2020_3000599 crossref_primary_10_1109_TCOMM_2020_3002915 crossref_primary_10_1109_MCOM_2019_1800858 crossref_primary_10_1109_TCCN_2021_3118465 crossref_primary_10_1186_s13638_021_01929_z crossref_primary_10_1016_j_comnet_2025_111531 crossref_primary_10_1007_s11235_021_00806_w crossref_primary_10_1049_cmu2_12424 crossref_primary_10_1109_TCCN_2024_3382973 crossref_primary_10_1002_ett_4656 crossref_primary_10_1007_s11277_024_11159_3 crossref_primary_10_1109_TSP_2021_3050564 crossref_primary_10_1109_TWC_2021_3087191 crossref_primary_10_1049_cmu2_12669 crossref_primary_10_1002_ett_4773 crossref_primary_10_1016_j_comnet_2021_107930 crossref_primary_10_1109_ACCESS_2022_3185124 crossref_primary_10_1016_j_eswa_2024_124857 crossref_primary_10_3390_s22041416 crossref_primary_10_3390_s24102993 crossref_primary_10_1109_JSAC_2023_3288249 crossref_primary_10_1109_TVT_2024_3441559 crossref_primary_10_1038_s41598_024_84017_8 crossref_primary_10_1109_TWC_2022_3157467 crossref_primary_10_1007_s11277_022_10111_7 crossref_primary_10_1007_s11277_022_10157_7 crossref_primary_10_1016_j_apacoust_2021_108515 crossref_primary_10_1049_cmu2_12655 crossref_primary_10_1002_dac_70132 crossref_primary_10_1109_TCOMM_2022_3180997 crossref_primary_10_1109_LWC_2022_3175091 crossref_primary_10_1109_TIFS_2020_2978620 crossref_primary_10_1109_LWC_2021_3076398 crossref_primary_10_1109_TSP_2020_3043879 crossref_primary_10_1109_ACCESS_2022_3206814 crossref_primary_10_1109_ACCESS_2021_3074350 crossref_primary_10_1109_JSAC_2021_3087224 crossref_primary_10_1109_TCOMM_2020_2968314 crossref_primary_10_1109_JLT_2024_3488121 crossref_primary_10_1109_ACCESS_2020_3041765 crossref_primary_10_1016_j_advwatres_2025_104952 crossref_primary_10_1016_j_asoc_2024_111672 crossref_primary_10_4218_etrij_2022_0462 crossref_primary_10_1109_TVT_2022_3201961 crossref_primary_10_1109_MCOM_001_2000218 crossref_primary_10_1109_TCCN_2020_2990657 crossref_primary_10_1109_TWC_2020_2979138 crossref_primary_10_1109_TCCN_2024_3454264 crossref_primary_10_1364_OL_553826 crossref_primary_10_1109_TSP_2022_3143372 crossref_primary_10_1109_ACCESS_2020_2995633 crossref_primary_10_1109_TCOMM_2022_3213284 crossref_primary_10_1109_TWC_2023_3243212 crossref_primary_10_1109_TVT_2021_3060459 crossref_primary_10_1109_LWC_2019_2933392 crossref_primary_10_1109_TCOMM_2020_3017020 crossref_primary_10_1109_ACCESS_2020_2964103 crossref_primary_10_1109_TCOMM_2019_2951563 crossref_primary_10_1016_j_dsp_2023_104196 crossref_primary_10_1109_JIOT_2025_3542753 crossref_primary_10_4316_AECE_2025_01007 crossref_primary_10_1109_TWC_2022_3158694 crossref_primary_10_1109_JPROC_2019_2941458 crossref_primary_10_1109_ACCESS_2022_3155879 crossref_primary_10_1109_JLT_2024_3519360 crossref_primary_10_1109_ACCESS_2019_2954551 crossref_primary_10_1109_JSAC_2019_2904361 crossref_primary_10_1007_s11432_020_2955_6 crossref_primary_10_1109_TCOMM_2022_3217777 crossref_primary_10_1109_TVT_2023_3278467 crossref_primary_10_1016_j_sysarc_2021_102180 crossref_primary_10_1016_j_optcom_2023_129997 crossref_primary_10_1109_TMAG_2020_3017373 crossref_primary_10_1109_TCCN_2019_2945774 crossref_primary_10_1109_COMST_2023_3315374 crossref_primary_10_1109_COMST_2024_3443193 crossref_primary_10_1109_TWC_2024_3381211 crossref_primary_10_3233_JIFS_219104 crossref_primary_10_3390_mi13010031 crossref_primary_10_1109_TCOMM_2022_3214236 crossref_primary_10_1109_ACCESS_2023_3238099 crossref_primary_10_1109_TCOMM_2019_2924010 crossref_primary_10_1109_LWC_2025_3538017 crossref_primary_10_1109_TCCN_2025_3527689 crossref_primary_10_1109_TNSM_2021_3123216 crossref_primary_10_1109_TWC_2020_2964661 crossref_primary_10_1007_s10922_022_09652_w crossref_primary_10_1049_cmu2_12588 crossref_primary_10_1109_TCOMM_2022_3228648 crossref_primary_10_1007_s11063_020_10419_0 crossref_primary_10_1109_ACCESS_2021_3070336 crossref_primary_10_1109_JETCAS_2020_2992593 crossref_primary_10_1109_JSAC_2021_3118346 crossref_primary_10_1109_TMC_2022_3153049 crossref_primary_10_1109_TWC_2019_2924424 crossref_primary_10_1109_JLT_2022_3213519 crossref_primary_10_1016_j_phycom_2023_102197 crossref_primary_10_1109_JIOT_2025_3571754 crossref_primary_10_1109_ACCESS_2020_2973318 crossref_primary_10_1117_1_OE_61_12_128101 crossref_primary_10_1109_OJCOMS_2025_3586560 crossref_primary_10_1109_TWC_2019_2956146 crossref_primary_10_1007_s12596_022_00924_x crossref_primary_10_1109_JSAC_2021_3126087 crossref_primary_10_1109_TCCN_2020_2985371 crossref_primary_10_1109_JSAC_2021_3126088 crossref_primary_10_1109_TR_2022_3148114 crossref_primary_10_3390_app9050852 crossref_primary_10_1016_j_phycom_2023_102199 crossref_primary_10_1016_j_optcom_2020_125272 crossref_primary_10_1109_TVT_2022_3226799 crossref_primary_10_1109_LCOMM_2020_3018260 crossref_primary_10_1109_TCCN_2024_3485648 crossref_primary_10_1109_JLT_2023_3251660 crossref_primary_10_1109_MCOM_001_2000804 crossref_primary_10_1109_JPROC_2024_3437730 crossref_primary_10_1016_j_phycom_2022_101835 crossref_primary_10_1016_j_asoc_2024_111759 crossref_primary_10_1109_ACCESS_2022_3221800 crossref_primary_10_1016_j_icte_2022_12_001 crossref_primary_10_1109_TVT_2021_3064868 crossref_primary_10_1109_ACCESS_2025_3566207 crossref_primary_10_1109_LCOMM_2022_3219382 crossref_primary_10_1109_TVT_2020_3027013 crossref_primary_10_3390_s23198139 crossref_primary_10_1109_TWC_2021_3124202 crossref_primary_10_1088_1742_6596_1617_1_012024 crossref_primary_10_3390_s21227716 crossref_primary_10_1109_COMST_2023_3333342 crossref_primary_10_1109_COMST_2021_3135829 crossref_primary_10_1109_TNET_2024_3444602 crossref_primary_10_1109_ACCESS_2020_2967402 crossref_primary_10_1109_LCOMM_2024_3456823 crossref_primary_10_1109_LWC_2022_3216848 crossref_primary_10_1109_TCOMM_2019_2960361 crossref_primary_10_1002_aisy_202500344 crossref_primary_10_1016_j_comcom_2021_09_033 crossref_primary_10_1016_j_phycom_2024_102585 crossref_primary_10_1109_TCCN_2021_3061464 crossref_primary_10_3390_s19163541 crossref_primary_10_1109_ACCESS_2020_3025597 crossref_primary_10_1109_TVT_2019_2951501 crossref_primary_10_1109_LCOMM_2024_3387286 crossref_primary_10_1109_TWC_2021_3138733 crossref_primary_10_1109_JSAC_2021_3126080 crossref_primary_10_1109_TCOMM_2020_3010964 crossref_primary_10_3390_s20102753 crossref_primary_10_1109_JIOT_2024_3425470 crossref_primary_10_1109_ACCESS_2024_3380203 crossref_primary_10_1049_iet_com_2019_0243 crossref_primary_10_3390_en12224300 crossref_primary_10_3390_electronics14061130 crossref_primary_10_1109_TWC_2019_2950026 crossref_primary_10_1002_dac_5109 crossref_primary_10_1049_2023_6610762 crossref_primary_10_1155_2021_5586029 crossref_primary_10_1109_TCCN_2024_3486168 crossref_primary_10_1109_JIOT_2023_3274209 crossref_primary_10_1109_TVT_2023_3241440 crossref_primary_10_1109_TWC_2021_3085303 crossref_primary_10_1109_ACCESS_2019_2935613 crossref_primary_10_1109_COMST_2019_2904897 crossref_primary_10_1109_COMST_2020_3033748 crossref_primary_10_1109_TCOMM_2022_3225460 crossref_primary_10_1109_JSAC_2019_2933890 crossref_primary_10_1109_JSAC_2019_2933891 crossref_primary_10_1109_COMST_2023_3326399 crossref_primary_10_3390_info14020098 crossref_primary_10_1016_j_phycom_2022_101976 crossref_primary_10_1109_TWC_2019_2962474 crossref_primary_10_23919_cje_2023_00_278 crossref_primary_10_1109_TCCN_2022_3225790 crossref_primary_10_1038_s44172_023_00108_w crossref_primary_10_1007_s11276_020_02423_y crossref_primary_10_1109_TCCN_2021_3132609 crossref_primary_10_1109_ACCESS_2019_2914928 crossref_primary_10_1109_TWC_2020_3042074 crossref_primary_10_1109_TCOMM_2024_3394039 crossref_primary_10_1109_TVT_2021_3099640 crossref_primary_10_3390_electronics11071138 crossref_primary_10_1016_j_icte_2022_01_018 crossref_primary_10_1109_TSP_2021_3070186 crossref_primary_10_1109_MWC_012_2200552 crossref_primary_10_1007_s41745_020_00169_2 crossref_primary_10_1109_TWC_2024_3418907 crossref_primary_10_1007_s11277_021_08222_8 crossref_primary_10_1109_MWC_2019_1800601 crossref_primary_10_3390_iot1010003 crossref_primary_10_1109_TVT_2024_3386912 crossref_primary_10_1016_j_yofte_2024_104069 crossref_primary_10_1109_MWC_001_2000005 crossref_primary_10_1109_JLT_2018_2865109 crossref_primary_10_1109_ACCESS_2019_2903130 crossref_primary_10_1109_LCOMM_2021_3108845 crossref_primary_10_1109_LWC_2021_3106039 crossref_primary_10_1007_s12596_024_02006_6 crossref_primary_10_1109_JSYST_2022_3204312 crossref_primary_10_1109_TWC_2024_3412234 crossref_primary_10_3390_make6030084 crossref_primary_10_1109_TCCN_2022_3161936 crossref_primary_10_1145_3494049 crossref_primary_10_1016_j_comnet_2020_107556 crossref_primary_10_1109_ACCESS_2021_3069707 crossref_primary_10_1109_JLT_2022_3148270 crossref_primary_10_1109_JSTSP_2021_3051490 crossref_primary_10_1109_TWC_2024_3462450 crossref_primary_10_3390_s22103666 crossref_primary_10_1109_ACCESS_2019_2909490 crossref_primary_10_1007_s11235_025_01274_2 crossref_primary_10_1016_j_vehcom_2020_100250 crossref_primary_10_1109_JLT_2021_3103339 crossref_primary_10_3390_app13042151 crossref_primary_10_3390_photonics9030190 crossref_primary_10_1109_ACCESS_2022_3208284 crossref_primary_10_1049_cmu2_12258 crossref_primary_10_1109_TCOMM_2019_2957482 crossref_primary_10_1109_JSAC_2019_2933956 crossref_primary_10_1109_JSAC_2019_2934004 crossref_primary_10_1109_JSAC_2025_3559154 crossref_primary_10_1109_JSYST_2022_3232590 crossref_primary_10_1109_TVT_2023_3294756 crossref_primary_10_3390_app10103634 crossref_primary_10_1007_s40012_024_00401_9 crossref_primary_10_1109_ACCESS_2020_3020544 crossref_primary_10_3390_electronics9020294 crossref_primary_10_1109_LWC_2021_3132067 crossref_primary_10_1109_ACCESS_2019_2919983 crossref_primary_10_1109_JETCAS_2020_3000103 crossref_primary_10_1109_TCCN_2020_3018736 crossref_primary_10_1109_TCCN_2021_3067916 crossref_primary_10_3390_s23249848 crossref_primary_10_1007_s11276_020_02412_1 crossref_primary_10_1109_COMST_2022_3223224 crossref_primary_10_3390_make6020057 crossref_primary_10_1016_j_phycom_2021_101402 crossref_primary_10_1109_TVLSI_2023_3274555 crossref_primary_10_1088_1742_6596_1169_1_012060 crossref_primary_10_1109_TCOMM_2020_3007622 crossref_primary_10_1007_s11761_024_00402_1 crossref_primary_10_1109_JSAC_2019_2933964 crossref_primary_10_1109_JLT_2020_3006827 crossref_primary_10_1109_JSAC_2019_2933965 crossref_primary_10_1109_TCSI_2024_3371780 crossref_primary_10_1109_ACCESS_2019_2923321 crossref_primary_10_1109_ACCESS_2019_2891218 crossref_primary_10_1007_s11276_023_03304_w crossref_primary_10_1109_TVT_2022_3151246 crossref_primary_10_1007_s11235_025_01312_z crossref_primary_10_1109_JSYST_2018_2870483 crossref_primary_10_1109_LWC_2019_2943466 crossref_primary_10_1109_ACCESS_2020_2984218 crossref_primary_10_1109_ACCESS_2022_3157651 crossref_primary_10_1109_JSAC_2020_3036943 crossref_primary_10_1016_j_nancom_2023_100456 crossref_primary_10_1109_ACCESS_2020_3000601 crossref_primary_10_1109_JSTSP_2022_3221850 crossref_primary_10_1109_TCOMM_2020_2978072 crossref_primary_10_1109_ACCESS_2019_2956211 crossref_primary_10_1007_s13369_021_05552_4 crossref_primary_10_1109_ACCESS_2020_2981745 crossref_primary_10_1109_TWC_2023_3328437
Cites_doi	10.1016/S0165-1684(00)00030-X 10.1109/TKDE.2009.191 10.1109/ISSPIT.2016.7886039 10.1109/26.141453 10.1109/ICASSP.2017.7952561 10.1109/SPAWC.2017.8227772 10.1109/SURV.2012.100412.00017 10.1214/08-PS141 10.1109/LCOMM.2005.09013 10.1109/ICC.2017.7997434 10.1002/j.1538-7305.1948.tb01338.x 10.1109/GlobalSIP.2016.7905837 10.1109/SPLIM.2016.7528397 10.1109/CISS.2017.7926071 10.1109/ACSSC.2016.7869126 10.1109/ALLERTON.2016.7852251 10.1109/ACCESS.2015.2471178 10.1038/nature16961
ContentType	Journal Article
Copyright	Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2018
Copyright_xml	– notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2018
DBID	97E RIA RIE AAYXX CITATION 7SP 8FD H8D L7M
DOI	10.1109/JSTSP.2017.2784180
DatabaseName	IEEE Xplore (IEEE) IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Xplore CrossRef Electronics & Communications Abstracts Technology Research Database Aerospace Database Advanced Technologies Database with Aerospace
DatabaseTitle	CrossRef Aerospace Database Technology Research Database Advanced Technologies Database with Aerospace Electronics & Communications Abstracts
DatabaseTitleList	Aerospace Database
Database_xml	– sequence: 1 dbid: RIE name: IEEE Xplore url: https://ieeexplore.ieee.org/ sourceTypes: Publisher
DeliveryMethod	fulltext_linktorsrc
Discipline	Engineering
EISSN	1941-0484
EndPage	143
ExternalDocumentID	10_1109_JSTSP_2017_2784180 8214233
Genre	orig-research
GroupedDBID	-~X 0R~ 29I 4.4 5GY 5VS 6IK 97E AAJGR AARMG AASAJ AAWTH ABAZT ABQJQ ABVLG ACIWK AENEX AETIX AGQYO AGSQL AHBIQ AKJIK AKQYR ALMA_UNASSIGNED_HOLDINGS ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 EBS EJD F5P HZ~ IFIPE IPLJI JAVBF LAI M43 O9- OCL RIA RIE RNS AAYXX CITATION 7SP 8FD H8D L7M RIG
ID	FETCH-LOGICAL-c361t-b5f77ff5ba5eeb9eaab8e8913d24af6fc4585a3828469d05cc69dc33d85a7eb03
IEDL.DBID	RIE
ISICitedReferencesCount	526
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000426010500011&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	1932-4553
IngestDate	Mon Jun 30 10:33:44 EDT 2025 Tue Nov 18 21:51:28 EST 2025 Sat Nov 29 04:10:31 EST 2025 Wed Aug 27 02:28:17 EDT 2025
IsPeerReviewed	true
IsScholarly	true
Issue	1
Language	English
License	https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c361t-b5f77ff5ba5eeb9eaab8e8913d24af6fc4585a3828469d05cc69dc33d85a7eb03
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ORCID	0000-0002-7614-5156
PQID	2007209730
PQPubID	75721
PageCount	12
ParticipantIDs	proquest_journals_2007209730 crossref_citationtrail_10_1109_JSTSP_2017_2784180 crossref_primary_10_1109_JSTSP_2017_2784180 ieee_primary_8214233
PublicationCentury	2000
PublicationDate	2018-02-01
PublicationDateYYYYMMDD	2018-02-01
PublicationDate_xml	– month: 02 year: 2018 text: 2018-02-01 day: 01
PublicationDecade	2010
PublicationPlace	New York
PublicationPlace_xml	– name: New York
PublicationTitle	IEEE journal of selected topics in signal processing
PublicationTitleAbbrev	JSTSP
PublicationYear	2018
Publisher	IEEE The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Publisher_xml	– name: IEEE – name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
References	nachmani (ref17) 2017 ref31 ref30 abadi (ref24) 2016 ref10 wang (ref5) 2016 ref2 goodfellow (ref26) 2016 ref1 ref16 ref19 ref18 rumelhart (ref28) 1986; 1 harris (ref32) 2004 collobert (ref15) 0 ref25 ref20 al-rfou (ref11) 2016 o’shea (ref4) 2017 ref22 ref21 kingma (ref33) 2014 ref27 chen (ref14) 2015 ref29 ref8 ref7 george (ref6) 2016 ref9 farsad (ref23) 2017 ref3 abadi (ref12) 2016 chollet (ref13) 2015
References_xml	– ident: ref8 doi: 10.1016/S0165-1684(00)00030-X – year: 2004 ident: ref32 publication-title: Multirate Signal Processing for Communication Systems – year: 2014 ident: ref33 article-title: Adam: A method for stochastic optimization – ident: ref29 doi: 10.1109/TKDE.2009.191 – year: 2015 ident: ref14 article-title: MXNet: A flexible and efficient machine learning library for heterogeneous distributed systems – ident: ref3 doi: 10.1109/ISSPIT.2016.7886039 – ident: ref2 doi: 10.1109/26.141453 – ident: ref20 doi: 10.1109/ICASSP.2017.7952561 – volume: 1 start-page: 318 year: 1986 ident: ref28 article-title: Parallel distributed processing: Explorations in the microstructure of cognition – ident: ref21 doi: 10.1109/SPAWC.2017.8227772 – year: 0 ident: ref15 article-title: Torch7: A MATLAB-like environment for machine learning publication-title: Proc BigLearn Neural Inf Process Syst Workshop – ident: ref9 doi: 10.1109/SURV.2012.100412.00017 – year: 2016 ident: ref26 publication-title: Deep Learning – ident: ref31 doi: 10.1214/08-PS141 – year: 2015 ident: ref13 article-title: Keras – ident: ref27 doi: 10.1109/LCOMM.2005.09013 – year: 2017 ident: ref23 article-title: Detection algorithms for communication systems using deep learning – year: 2016 ident: ref11 article-title: Theano: A python framework for fast computation of mathematical expressions – year: 2016 ident: ref5 article-title: Deep learning for identifying metastatic breast cancer – ident: ref22 doi: 10.1109/ICC.2017.7997434 – ident: ref1 doi: 10.1002/j.1538-7305.1948.tb01338.x – ident: ref19 doi: 10.1109/GlobalSIP.2016.7905837 – ident: ref25 doi: 10.1109/SPLIM.2016.7528397 – year: 2016 ident: ref6 article-title: Deep neural networks to enable real-time multimessenger astrophysics – year: 2017 ident: ref4 article-title: An introduction to machine learning communications systems – ident: ref18 doi: 10.1109/CISS.2017.7926071 – ident: ref30 doi: 10.1109/ACSSC.2016.7869126 – year: 2016 ident: ref12 article-title: TensorFlow: Large-scale machine learning on heterogeneous distributed systems – ident: ref16 doi: 10.1109/ALLERTON.2016.7852251 – year: 2017 ident: ref17 article-title: RNN decoding of linear block codes – ident: ref10 doi: 10.1109/ACCESS.2015.2471178 – ident: ref7 doi: 10.1038/nature16961 – year: 2016 ident: ref24 article-title: Learning to protect communications with adversarial neural cryptography
SSID	ssj0057614
Score	2.6736963
Snippet	End-to-end learning of communications systems is a fascinating novel concept that has so far only been validated by simulations for block-based transmissions....
SourceID	proquest crossref ieee
SourceType	Aggregation Database Enrichment Source Index Database Publisher
StartPage	132
SubjectTerms	Artificial neural networks Autoencoder communication Communication systems Communications systems Computer simulation Data transmission Deep learning end-to-end learning Hardware modulation neural network Neural networks over-the-air Receivers software-defined radio Source code Synchronism Synchronization Training Transmitters
Title	Deep Learning Based Communication Over the Air
URI	https://ieeexplore.ieee.org/document/8214233 https://www.proquest.com/docview/2007209730
Volume	12
WOSCitedRecordID	wos000426010500011&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVIEE databaseName: IEEE Xplore customDbUrl: eissn: 1941-0484 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0057614 issn: 1932-4553 databaseCode: RIE dateStart: 20070101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3dS8MwED_m8EEf_JridEoefNNu7ZI07eP8GD7IHKiwt9IkFxnINvbh32-StmOiCD61lKSUX665u9zd7wCuFMu7nMow4HnkHJQkDtJYsgBDRCUMMwY9Zf6TGAyS0Sgd1uBmXQuDiD75DNvu1sfy9VSt3FFZJ3H8YJRuwZYQcVGrVe261myOyghyN2Cc06pAJkw7VsRfhi6LS7R9mM1RQG4oId9V5cdW7PVLf_9_X3YAe6UdSXrFwh9CDSdHsLvBLtiA9j3ijJQEqu_k1uorTb4VhJBnK8jEmoCkN54fw1v_4fXuMSjbIwSKxtEykNwIYQyXOUeUKea5TNBFHXWX5SY2illXIKfWpbIusA65UvaiKNX2qUAZ0hOoT6YTPAUSxYprgZHBRDFEq8ONjE1OBdMpVUI3IarwylTJHe5aWHxk3ocI08xjnDmMsxLjJlyv58wK5ow_RzccquuRJaBNaFXLkpU_18J1zhRdRzMUnv0-6xx27LuTIrm6BfXlfIUXsK0-l-PF_NLLzRfvd78o
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3dS8MwED_mFNQHv6Y4ndoH37Rb2iRN-zg_xsQ5B07YW2nTiwxkjn3495uk3Zgogk8tJaHll2vuLnf3O4BLyRKf05S4PPGMgxIGbhSkzEWCKIViSqGlzO-IbjccDKJeCa6XtTCIaJPPsG5ubSw_-5Bzc1TWCA0_GKVrsM4Z80lerbXYd7Xh7BUxZN9lnNNFiQyJGlrIX3omj0vUbaDNkECuqCHbV-XHZmw1TGv3f9-2BzuFJek086XfhxKODmB7hV-wAvU7xLFTUKi-OTdaY2XOt5IQ51mLsqONQKc5nBzCa-u-f9t2iwYJrqSBN3NTroRQiqcJR0wjTJI0RBN3zHyWqEBJpp2BhGqnSjvBGeFS6oukNNNPBaaEHkF59DHCY3C8QPJMoKcwlAxRa3GVBiqhgmURlSKrgrfAK5YFe7hpYvEeWy-CRLHFODYYxwXGVbhazhnn3Bl_jq4YVJcjC0CrUFssS1z8XlPTO1P4hmiInPw-6wI22_2nTtx56D6ewpZ-T5inWtegPJvM8Qw25OdsOJ2cWxn6Ag7Iwm8
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Deep+Learning+Based+Communication+Over+the+Air&rft.jtitle=IEEE+journal+of+selected+topics+in+signal+processing&rft.au=Dorner%2C+Sebastian&rft.au=Cammerer%2C+Sebastian&rft.au=Hoydis%2C+Jakob&rft.au=Brink%2C+Stephan+ten&rft.date=2018-02-01&rft.pub=IEEE&rft.issn=1932-4553&rft.volume=12&rft.issue=1&rft.spage=132&rft.epage=143&rft_id=info:doi/10.1109%2FJSTSP.2017.2784180&rft.externalDocID=8214233
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1932-4553&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1932-4553&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1932-4553&client=summon